API tools faq
paste
Advertisement
lamiastella

train result NAN loss

lamiastella
Nov 10th, 2020 (edited)
262
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 2.00 MB | None | 0 0
raw download clone embed print report
  1. size of train loader is: 90
  2. torch.Size([8, 600, 800])
  3. torch.Size([8, 8])
  4. tensor([[0.6200, 0.4070, 0.8938, 0.4183, 0.3538, 0.4567, 0.6175, 0.5400],
  5. [0.6197, 0.3986, 0.8800, 0.4617, 0.4188, 0.4783, 0.5687, 0.5550],
  6. [0.6276, 0.4002, 0.8800, 0.5533, 0.3575, 0.4400, 0.6132, 0.4672],
  7. [0.6162, 0.4014, 0.8800, 0.5333, 0.3750, 0.4817, 0.5988, 0.5283],
  8. [0.6346, 0.4165, 0.9138, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471],
  9. [0.6325, 0.4165, 0.9000, 0.4617, 0.3813, 0.4900, 0.7485, 0.5447],
  10. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  11. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383]],
  12. device='cuda:0', dtype=torch.float64)
  13. predictions are: tensor([[-0.0584, 0.2115, -0.0577, -0.0264, 0.6124, -0.3884, 0.2870, 0.4973],
  14. [-0.0186, 0.2038, -0.0694, -0.0212, 0.5656, -0.3732, 0.2786, 0.4981],
  15. [-0.0844, 0.2453, -0.0447, -0.0736, 0.6318, -0.3736, 0.2809, 0.5259],
  16. [-0.0908, 0.2592, -0.0879, -0.0383, 0.5870, -0.3843, 0.3103, 0.4775],
  17. [-0.0545, 0.2420, -0.0500, -0.0258, 0.6162, -0.3611, 0.2940, 0.5109],
  18. [-0.0442, 0.2109, -0.0649, -0.0483, 0.5915, -0.3324, 0.2890, 0.5157],
  19. [-0.0514, 0.2363, -0.0713, -0.0051, 0.6292, -0.3114, 0.2788, 0.4927],
  20. [-0.0822, 0.2308, -0.0656, -0.0090, 0.6174, -0.3896, 0.2823, 0.5259]],
  21. device='cuda:0', grad_fn=<AddmmBackward>)
  22. landmarks are: tensor([[[0.6200, 0.4070, 0.8938, 0.4183, 0.3537, 0.4567, 0.6175, 0.5400],
  23. [0.6197, 0.3986, 0.8800, 0.4617, 0.4187, 0.4783, 0.5688, 0.5550],
  24. [0.6276, 0.4002, 0.8800, 0.5533, 0.3575, 0.4400, 0.6132, 0.4672],
  25. [0.6162, 0.4014, 0.8800, 0.5333, 0.3750, 0.4817, 0.5987, 0.5283],
  26. [0.6346, 0.4165, 0.9137, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471],
  27. [0.6325, 0.4165, 0.9000, 0.4617, 0.3812, 0.4900, 0.7485, 0.5447],
  28. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  29. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383]]],
  30. device='cuda:0')
  31. loss_train_step before backward: tensor(0.2998, device='cuda:0', grad_fn=<MseLossBackward>)
  32. loss_train_step after backward: tensor(0.2998, device='cuda:0', grad_fn=<MseLossBackward>)
  33. loss_train: 0.2997697591781616
  34. step: 1
  35. running loss: 0.2997697591781616
  36. Train Steps: 1/90 Loss: 0.2998 torch.Size([8, 600, 800])
  37. torch.Size([8, 8])
  38. tensor([[0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  39. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  40. [0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166],
  41. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901],
  42. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  43. [0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  44. [0.6200, 0.3993, 0.8519, 0.4923, 0.3962, 0.4717, 0.6013, 0.5433],
  45. [0.6165, 0.4106, 0.7575, 0.1733, 0.3838, 0.2650, 0.5680, 0.5116]],
  46. device='cuda:0', dtype=torch.float64)
  47. predictions are: tensor([[ 0.1958, 0.3088, 0.2890, 0.1878, 0.4821, -0.0499, 0.4648, 0.5882],
  48. [ 0.2268, 0.3071, 0.2562, 0.1235, 0.4270, -0.0543, 0.4471, 0.5042],
  49. [ 0.2150, 0.3346, 0.2953, 0.1354, 0.4465, -0.0666, 0.4839, 0.5662],
  50. [ 0.2461, 0.2619, 0.2592, 0.1226, 0.4609, -0.0660, 0.4270, 0.4831],
  51. [ 0.1998, 0.3029, 0.2570, 0.1213, 0.4600, -0.0628, 0.4580, 0.5487],
  52. [ 0.2079, 0.2926, 0.2190, 0.1779, 0.4371, -0.0531, 0.4744, 0.5373],
  53. [ 0.2588, 0.2846, 0.2885, 0.1383, 0.4450, -0.0381, 0.4251, 0.5413],
  54. [ 0.1834, 0.2973, 0.3002, 0.1576, 0.4539, -0.0788, 0.4260, 0.5624]],
  55. device='cuda:0', grad_fn=<AddmmBackward>)
  56. landmarks are: tensor([[[0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  57. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  58. [0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166],
  59. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901],
  60. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  61. [0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  62. [0.6200, 0.3993, 0.8519, 0.4923, 0.3963, 0.4717, 0.6012, 0.5433],
  63. [0.6165, 0.4106, 0.7575, 0.1733, 0.3837, 0.2650, 0.5680, 0.5116]]],
  64. device='cuda:0')
  65. loss_train_step before backward: tensor(0.1120, device='cuda:0', grad_fn=<MseLossBackward>)
  66. loss_train_step after backward: tensor(0.1120, device='cuda:0', grad_fn=<MseLossBackward>)
  67. loss_train: 0.4117617905139923
  68. step: 2
  69. running loss: 0.20588089525699615
  70. Train Steps: 2/90 Loss: 0.2059 torch.Size([8, 600, 800])
  71. torch.Size([8, 8])
  72. tensor([[0.6133, 0.4094, 0.8495, 0.4028, 0.3588, 0.3200, 0.5003, 0.5407],
  73. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  74. [0.6286, 0.4078, 0.8063, 0.2267, 0.4788, 0.1533, 0.5953, 0.4913],
  75. [0.6059, 0.4002, 0.7562, 0.2767, 0.3538, 0.3033, 0.5529, 0.5455],
  76. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5637, 0.5633],
  77. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  78. [0.6229, 0.4107, 0.8137, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  79. [ nan, nan, 0.7612, 0.3250, 0.4037, 0.2533, 0.5438, 0.5767]],
  80. device='cuda:0', dtype=torch.float64)
  81. predictions are: tensor([[0.4055, 0.3739, 0.5440, 0.2942, 0.3808, 0.2197, 0.5719, 0.5390],
  82. [0.4301, 0.4093, 0.5301, 0.2674, 0.3338, 0.2265, 0.5864, 0.5637],
  83. [0.3896, 0.3546, 0.4938, 0.2991, 0.3735, 0.1760, 0.5180, 0.5417],
  84. [0.3896, 0.3833, 0.5534, 0.3136, 0.3577, 0.1936, 0.5783, 0.5552],
  85. [0.4145, 0.4122, 0.5246, 0.2921, 0.3729, 0.2434, 0.5558, 0.5490],
  86. [0.3705, 0.3855, 0.5324, 0.3042, 0.3832, 0.1809, 0.5555, 0.5525],
  87. [0.3739, 0.3511, 0.5211, 0.2681, 0.3678, 0.1772, 0.5384, 0.5119],
  88. [0.3903, 0.3680, 0.5187, 0.2795, 0.3428, 0.1927, 0.5294, 0.5423]],
  89. device='cuda:0', grad_fn=<AddmmBackward>)
  90. landmarks are: tensor([[[0.6133, 0.4094, 0.8495, 0.4028, 0.3587, 0.3200, 0.5003, 0.5407],
  91. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  92. [0.6286, 0.4078, 0.8062, 0.2267, 0.4787, 0.1533, 0.5953, 0.4913],
  93. [0.6059, 0.4002, 0.7563, 0.2767, 0.3537, 0.3033, 0.5529, 0.5455],
  94. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5638, 0.5633],
  95. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  96. [0.6229, 0.4107, 0.8138, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  97. [0.0000, 0.0000, 0.7613, 0.3250, 0.4038, 0.2533, 0.5437, 0.5767]]],
  98. device='cuda:0')
  99. loss_train_step before backward: tensor(0.0267, device='cuda:0', grad_fn=<MseLossBackward>)
  100. loss_train_step after backward: tensor(0.0267, device='cuda:0', grad_fn=<MseLossBackward>)
  101. loss_train: 0.43846603110432625
  102. step: 3
  103. running loss: 0.1461553437014421
  104. Train Steps: 3/90 Loss: 0.1462 torch.Size([8, 600, 800])
  105. torch.Size([8, 8])
  106. tensor([[0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  107. [0.6230, 0.4152, 0.7588, 0.2283, 0.4012, 0.2883, 0.6200, 0.5767],
  108. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  109. [0.6075, 0.4000, 0.8513, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  110. [0.6177, 0.4086, 0.8738, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700],
  111. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  112. [0.6212, 0.4171, 0.7875, 0.3633, 0.3813, 0.2933, 0.5675, 0.5700],
  113. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6063, 0.5617]],
  114. device='cuda:0', dtype=torch.float64)
  115. predictions are: tensor([[0.5931, 0.4201, 0.7860, 0.3801, 0.3459, 0.3958, 0.5810, 0.5520],
  116. [0.5350, 0.4044, 0.8002, 0.3776, 0.3409, 0.3038, 0.6013, 0.5565],
  117. [0.5821, 0.4141, 0.8026, 0.3910, 0.3213, 0.3766, 0.6106, 0.5469],
  118. [0.5523, 0.3895, 0.7907, 0.4043, 0.3298, 0.3388, 0.6085, 0.5748],
  119. [0.5915, 0.4425, 0.8049, 0.4194, 0.3449, 0.4237, 0.6408, 0.5606],
  120. [0.6064, 0.4165, 0.7978, 0.3757, 0.3294, 0.4055, 0.6155, 0.5291],
  121. [0.5497, 0.3888, 0.7902, 0.3943, 0.3422, 0.3726, 0.6113, 0.5332],
  122. [0.5783, 0.4244, 0.8106, 0.3968, 0.2945, 0.3802, 0.6230, 0.5543]],
  123. device='cuda:0', grad_fn=<AddmmBackward>)
  124. landmarks are: tensor([[[0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  125. [0.6230, 0.4152, 0.7588, 0.2283, 0.4013, 0.2883, 0.6200, 0.5767],
  126. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  127. [0.6075, 0.4000, 0.8512, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  128. [0.6177, 0.4085, 0.8737, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700],
  129. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  130. [0.6212, 0.4171, 0.7875, 0.3633, 0.3812, 0.2933, 0.5675, 0.5700],
  131. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6062, 0.5617]]],
  132. device='cuda:0')
  133. loss_train_step before backward: tensor(0.0052, device='cuda:0', grad_fn=<MseLossBackward>)
  134. loss_train_step after backward: tensor(0.0052, device='cuda:0', grad_fn=<MseLossBackward>)
  135. loss_train: 0.4436452961526811
  136. step: 4
  137. running loss: 0.11091132403817028
  138.  
  139. Train Steps: 4/90 Loss: 0.1109 torch.Size([8, 600, 800])
  140. torch.Size([8, 8])
  141. tensor([[0.6264, 0.4071, 0.9038, 0.3867, 0.3663, 0.3917, 0.6338, 0.5283],
  142. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  143. [0.6229, 0.4107, 0.8137, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  144. [0.6115, 0.3998, 0.7063, 0.2383, 0.4037, 0.1950, 0.5320, 0.4993],
  145. [0.6160, 0.4093, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  146. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  147. [0.6064, 0.4019, 0.8650, 0.4517, 0.4037, 0.5367, 0.5703, 0.5609],
  148. [0.6353, 0.4128, 0.9138, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991]],
  149. device='cuda:0', dtype=torch.float64)
  150. predictions are: tensor([[0.6595, 0.4149, 0.9635, 0.4405, 0.3608, 0.5043, 0.6468, 0.5533],
  151. [0.6573, 0.4416, 0.9378, 0.4602, 0.3851, 0.4643, 0.6446, 0.5442],
  152. [0.6465, 0.3929, 0.9284, 0.3990, 0.3773, 0.3864, 0.6122, 0.5066],
  153. [0.6654, 0.4107, 0.9472, 0.4188, 0.3765, 0.4518, 0.6117, 0.5490],
  154. [0.6871, 0.4540, 0.9970, 0.4859, 0.3264, 0.5708, 0.6297, 0.5511],
  155. [0.6877, 0.4320, 0.9677, 0.5030, 0.3692, 0.5202, 0.6485, 0.5374],
  156. [0.6752, 0.4174, 1.0047, 0.5007, 0.3579, 0.5584, 0.6603, 0.5398],
  157. [0.6805, 0.4097, 0.9533, 0.4593, 0.3685, 0.4398, 0.6234, 0.5440]],
  158. device='cuda:0', grad_fn=<AddmmBackward>)
  159. landmarks are: tensor([[[0.6264, 0.4071, 0.9038, 0.3867, 0.3663, 0.3917, 0.6338, 0.5283],
  160. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  161. [0.6229, 0.4107, 0.8138, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  162. [0.6115, 0.3998, 0.7063, 0.2383, 0.4038, 0.1950, 0.5320, 0.4993],
  163. [0.6160, 0.4092, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  164. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  165. [0.6064, 0.4019, 0.8650, 0.4517, 0.4038, 0.5367, 0.5703, 0.5609],
  166. [0.6353, 0.4128, 0.9137, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991]]],
  167. device='cuda:0')
  168. loss_train_step before backward: tensor(0.0123, device='cuda:0', grad_fn=<MseLossBackward>)
  169. loss_train_step after backward: tensor(0.0123, device='cuda:0', grad_fn=<MseLossBackward>)
  170. loss_train: 0.455948734190315
  171. step: 5
  172. running loss: 0.091189746838063
  173. Train Steps: 5/90 Loss: 0.0912 torch.Size([8, 600, 800])
  174. torch.Size([8, 8])
  175. tensor([[0.6111, 0.3995, 0.8788, 0.4567, 0.3813, 0.4833, 0.5450, 0.5700],
  176. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  177. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  178. [0.6204, 0.4049, 0.7975, 0.2700, 0.3937, 0.2567, 0.5700, 0.5183],
  179. [0.6154, 0.4112, 0.7037, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  180. [0.6204, 0.4055, 0.8438, 0.5733, 0.4574, 0.4801, 0.5487, 0.5617],
  181. [0.6207, 0.4081, 0.7662, 0.2067, 0.3962, 0.3200, 0.6312, 0.5300],
  182. [0.6200, 0.4071, 0.7338, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517]],
  183. device='cuda:0', dtype=torch.float64)
  184. predictions are: tensor([[0.7639, 0.4666, 1.0737, 0.5197, 0.4200, 0.6167, 0.6571, 0.5344],
  185. [0.7446, 0.4824, 1.0979, 0.4950, 0.4316, 0.6142, 0.6310, 0.5616],
  186. [0.7546, 0.4313, 1.0392, 0.4391, 0.4339, 0.5028, 0.6431, 0.5494],
  187. [0.7280, 0.4014, 1.0350, 0.3954, 0.4080, 0.4104, 0.6010, 0.5530],
  188. [0.7477, 0.4200, 1.0253, 0.4783, 0.4805, 0.4765, 0.5870, 0.5082],
  189. [0.7686, 0.4448, 1.0690, 0.4399, 0.4344, 0.5180, 0.6374, 0.5681],
  190. [0.7443, 0.4067, 1.0424, 0.4304, 0.4043, 0.5326, 0.6042, 0.5293],
  191. [0.7320, 0.4131, 1.0113, 0.4355, 0.3841, 0.4534, 0.6299, 0.5128]],
  192. device='cuda:0', grad_fn=<AddmmBackward>)
  193. landmarks are: tensor([[[0.6111, 0.3995, 0.8788, 0.4567, 0.3812, 0.4833, 0.5450, 0.5700],
  194. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  195. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  196. [0.6204, 0.4049, 0.7975, 0.2700, 0.3938, 0.2567, 0.5700, 0.5183],
  197. [0.6154, 0.4112, 0.7038, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  198. [0.6204, 0.4055, 0.8438, 0.5733, 0.4574, 0.4801, 0.5487, 0.5617],
  199. [0.6207, 0.4081, 0.7663, 0.2067, 0.3963, 0.3200, 0.6313, 0.5300],
  200. [0.6200, 0.4071, 0.7337, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517]]],
  201. device='cuda:0')
  202. loss_train_step before backward: tensor(0.0178, device='cuda:0', grad_fn=<MseLossBackward>)
  203. loss_train_step after backward: tensor(0.0178, device='cuda:0', grad_fn=<MseLossBackward>)
  204. loss_train: 0.4737188364379108
  205. step: 6
  206. running loss: 0.07895313940631847
  207. Train Steps: 6/90 Loss: 0.0790 torch.Size([8, 600, 800])
  208. torch.Size([8, 8])
  209. tensor([[0.6350, 0.4043, 0.8738, 0.5650, 0.3850, 0.4750, 0.6401, 0.4950],
  210. [0.6276, 0.4095, 0.8237, 0.2250, 0.4662, 0.1783, 0.6171, 0.4869],
  211. [0.6131, 0.4037, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  212. [0.6202, 0.4066, 0.8398, 0.2648, 0.3925, 0.2627, 0.5845, 0.5124],
  213. [0.6196, 0.4088, 0.8888, 0.4583, 0.4500, 0.5683, 0.6138, 0.5883],
  214. [0.6267, 0.4065, 0.8313, 0.2467, 0.4788, 0.1733, 0.6312, 0.5133],
  215. [0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364],
  216. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575]],
  217. device='cuda:0', dtype=torch.float64)
  218. predictions are: tensor([[0.7992, 0.4712, 1.0351, 0.5252, 0.4813, 0.6162, 0.6746, 0.5770],
  219. [0.7076, 0.4036, 0.9749, 0.3386, 0.4551, 0.3386, 0.5951, 0.5499],
  220. [0.7864, 0.4010, 1.0021, 0.3990, 0.4437, 0.4824, 0.5861, 0.5256],
  221. [0.6992, 0.3757, 0.9816, 0.3626, 0.4514, 0.2998, 0.5664, 0.5662],
  222. [0.8054, 0.4612, 1.0290, 0.4557, 0.4737, 0.5697, 0.6536, 0.5601],
  223. [0.7407, 0.4508, 0.9957, 0.4088, 0.4786, 0.4903, 0.6357, 0.5672],
  224. [0.7307, 0.4101, 1.0154, 0.4157, 0.4688, 0.3993, 0.5921, 0.5474],
  225. [0.7518, 0.3969, 1.0725, 0.4365, 0.4336, 0.5652, 0.6162, 0.5677]],
  226. device='cuda:0', grad_fn=<AddmmBackward>)
  227. landmarks are: tensor([[[0.6350, 0.4043, 0.8737, 0.5650, 0.3850, 0.4750, 0.6401, 0.4950],
  228. [0.6276, 0.4095, 0.8238, 0.2250, 0.4663, 0.1783, 0.6171, 0.4869],
  229. [0.6131, 0.4036, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  230. [0.6202, 0.4066, 0.8398, 0.2648, 0.3925, 0.2627, 0.5845, 0.5124],
  231. [0.6196, 0.4088, 0.8888, 0.4583, 0.4500, 0.5683, 0.6137, 0.5883],
  232. [0.6266, 0.4065, 0.8313, 0.2467, 0.4787, 0.1733, 0.6313, 0.5133],
  233. [0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364],
  234. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575]]],
  235. device='cuda:0')
  236. loss_train_step before backward: tensor(0.0135, device='cuda:0', grad_fn=<MseLossBackward>)
  237. loss_train_step after backward: tensor(0.0135, device='cuda:0', grad_fn=<MseLossBackward>)
  238. loss_train: 0.48724996810778975
  239. step: 7
  240. running loss: 0.06960713830111283
  241. Train Steps: 7/90 Loss: 0.0696 torch.Size([8, 600, 800])
  242. torch.Size([8, 8])
  243. tensor([[0.6134, 0.4090, 0.6926, 0.2819, 0.3538, 0.3233, 0.5563, 0.5667],
  244. [0.6246, 0.4028, 0.8738, 0.4867, 0.4088, 0.5667, 0.6362, 0.5200],
  245. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  246. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  247. [ nan, nan, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  248. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  249. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  250. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100]],
  251. device='cuda:0', dtype=torch.float64)
  252. predictions are: tensor([[0.7125, 0.4141, 0.9276, 0.3407, 0.4545, 0.3371, 0.6088, 0.5133],
  253. [0.7724, 0.4321, 0.9407, 0.3951, 0.4934, 0.5374, 0.6366, 0.5840],
  254. [0.7000, 0.4080, 0.9320, 0.3456, 0.4463, 0.3606, 0.5961, 0.5657],
  255. [0.7373, 0.4212, 0.9778, 0.3996, 0.4907, 0.4384, 0.6298, 0.5426],
  256. [0.6694, 0.3810, 0.9105, 0.2660, 0.4332, 0.2168, 0.5700, 0.5416],
  257. [0.7486, 0.4425, 0.9655, 0.4365, 0.4873, 0.4980, 0.6413, 0.5653],
  258. [0.6991, 0.4299, 0.9115, 0.3282, 0.5000, 0.3444, 0.6179, 0.5303],
  259. [0.7601, 0.4259, 0.9607, 0.4066, 0.4690, 0.5100, 0.6555, 0.5402]],
  260. device='cuda:0', grad_fn=<AddmmBackward>)
  261. landmarks are: tensor([[[0.6134, 0.4090, 0.6926, 0.2819, 0.3537, 0.3233, 0.5562, 0.5667],
  262. [0.6246, 0.4028, 0.8737, 0.4867, 0.4087, 0.5667, 0.6363, 0.5200],
  263. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  264. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  265. [0.0000, 0.0000, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  266. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  267. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  268. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100]]],
  269. device='cuda:0')
  270. loss_train_step before backward: tensor(0.0159, device='cuda:0', grad_fn=<MseLossBackward>)
  271. loss_train_step after backward: tensor(0.0159, device='cuda:0', grad_fn=<MseLossBackward>)
  272. loss_train: 0.503159002866596
  273. step: 8
  274. running loss: 0.0628948753583245
  275.  
  276. Train Steps: 8/90 Loss: 0.0629 torch.Size([8, 600, 800])
  277. torch.Size([8, 8])
  278. tensor([[0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  279. [0.6193, 0.4034, 0.7757, 0.2347, 0.3733, 0.2919, 0.5930, 0.4926],
  280. [0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  281. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600],
  282. [0.6329, 0.4196, 0.9238, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748],
  283. [0.6230, 0.4152, 0.7588, 0.2283, 0.4012, 0.2883, 0.6200, 0.5767],
  284. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  285. [0.6111, 0.4033, 0.8300, 0.3267, 0.3588, 0.3333, 0.5444, 0.5637]],
  286. device='cuda:0', dtype=torch.float64)
  287. predictions are: tensor([[0.6310, 0.4255, 0.8500, 0.3970, 0.4594, 0.4231, 0.6480, 0.5211],
  288. [0.5883, 0.3990, 0.8303, 0.3005, 0.4513, 0.3195, 0.6211, 0.5154],
  289. [0.6312, 0.3964, 0.8382, 0.3922, 0.4255, 0.4006, 0.6227, 0.4975],
  290. [0.6345, 0.4232, 0.8789, 0.3901, 0.4579, 0.4217, 0.6088, 0.5509],
  291. [0.5915, 0.3512, 0.8164, 0.3049, 0.4660, 0.2282, 0.5638, 0.5399],
  292. [0.6027, 0.3788, 0.8326, 0.2586, 0.4733, 0.2217, 0.6128, 0.5120],
  293. [0.6407, 0.4505, 0.8766, 0.4017, 0.4244, 0.4355, 0.6682, 0.5646],
  294. [0.6337, 0.3537, 0.8414, 0.3109, 0.4235, 0.2654, 0.6164, 0.5033]],
  295. device='cuda:0', grad_fn=<AddmmBackward>)
  296. landmarks are: tensor([[[0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  297. [0.6193, 0.4034, 0.7757, 0.2347, 0.3733, 0.2919, 0.5930, 0.4926],
  298. [0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  299. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600],
  300. [0.6329, 0.4196, 0.9237, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748],
  301. [0.6230, 0.4152, 0.7588, 0.2283, 0.4013, 0.2883, 0.6200, 0.5767],
  302. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  303. [0.6111, 0.4033, 0.8300, 0.3267, 0.3587, 0.3333, 0.5444, 0.5637]]],
  304. device='cuda:0')
  305. loss_train_step before backward: tensor(0.0041, device='cuda:0', grad_fn=<MseLossBackward>)
  306. loss_train_step after backward: tensor(0.0041, device='cuda:0', grad_fn=<MseLossBackward>)
  307. loss_train: 0.5072252913378179
  308. step: 9
  309. running loss: 0.05635836570420199
  310. Train Steps: 9/90 Loss: 0.0564 torch.Size([8, 600, 800])
  311. torch.Size([8, 8])
  312. tensor([[0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  313. [0.6250, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6088, 0.5183],
  314. [0.6128, 0.4084, 0.8738, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  315. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  316. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  317. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  318. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  319. [ nan, nan, 0.8850, 0.3000, 0.5363, 0.2250, 0.7343, 0.5771]],
  320. device='cuda:0', dtype=torch.float64)
  321. predictions are: tensor([[0.5854, 0.4194, 0.7982, 0.4484, 0.3842, 0.4339, 0.6975, 0.4881],
  322. [0.5636, 0.3870, 0.7519, 0.3707, 0.4280, 0.3838, 0.6772, 0.5203],
  323. [0.5463, 0.3677, 0.7634, 0.3431, 0.3938, 0.2582, 0.6202, 0.5056],
  324. [0.5107, 0.3751, 0.7688, 0.3725, 0.3725, 0.3147, 0.6341, 0.5228],
  325. [0.5468, 0.3651, 0.7660, 0.3440, 0.3912, 0.2587, 0.6207, 0.5004],
  326. [0.5300, 0.3686, 0.7474, 0.3380, 0.3827, 0.3088, 0.6002, 0.5400],
  327. [0.5577, 0.3624, 0.7734, 0.3745, 0.3982, 0.3396, 0.6820, 0.4983],
  328. [0.5084, 0.3917, 0.7298, 0.2798, 0.4158, 0.2044, 0.6265, 0.4828]],
  329. device='cuda:0', grad_fn=<AddmmBackward>)
  330. landmarks are: tensor([[[0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  331. [0.6251, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6087, 0.5183],
  332. [0.6127, 0.4084, 0.8737, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  333. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  334. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  335. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  336. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  337. [0.0000, 0.0000, 0.8850, 0.3000, 0.5362, 0.2250, 0.7343, 0.5771]]],
  338. device='cuda:0')
  339. loss_train_step before backward: tensor(0.0133, device='cuda:0', grad_fn=<MseLossBackward>)
  340. loss_train_step after backward: tensor(0.0133, device='cuda:0', grad_fn=<MseLossBackward>)
  341. loss_train: 0.5205279276706278
  342. step: 10
  343. running loss: 0.05205279276706278
  344. Train Steps: 10/90 Loss: 0.0521 torch.Size([8, 600, 800])
  345. torch.Size([8, 8])
  346. tensor([[0.6193, 0.4108, 0.7438, 0.2700, 0.3650, 0.3683, 0.6238, 0.5717],
  347. [0.6216, 0.4099, 0.7225, 0.2033, 0.4188, 0.2217, 0.5975, 0.5283],
  348. [0.6274, 0.4270, 0.8938, 0.4967, 0.3550, 0.4283, 0.5700, 0.5733],
  349. [0.6236, 0.3966, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  350. [0.6310, 0.4017, 0.8563, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006],
  351. [0.6260, 0.4120, 0.8013, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895],
  352. [0.6201, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  353. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297]],
  354. device='cuda:0', dtype=torch.float64)
  355. predictions are: tensor([[0.4604, 0.3238, 0.7617, 0.3943, 0.3457, 0.3056, 0.6460, 0.4995],
  356. [0.4523, 0.3381, 0.6931, 0.3316, 0.3268, 0.2285, 0.6105, 0.5138],
  357. [0.4482, 0.3509, 0.7564, 0.4134, 0.3853, 0.3224, 0.5696, 0.5136],
  358. [0.5103, 0.3802, 0.7495, 0.4832, 0.3225, 0.3799, 0.6293, 0.5123],
  359. [0.5484, 0.3973, 0.7444, 0.5327, 0.3770, 0.5066, 0.6837, 0.5201],
  360. [0.4322, 0.3602, 0.7118, 0.3229, 0.3856, 0.1895, 0.5927, 0.5381],
  361. [0.4152, 0.3308, 0.6963, 0.2808, 0.3751, 0.1732, 0.5842, 0.4984],
  362. [0.4961, 0.3820, 0.7515, 0.4521, 0.3516, 0.3834, 0.6456, 0.5185]],
  363. device='cuda:0', grad_fn=<AddmmBackward>)
  364. landmarks are: tensor([[[0.6193, 0.4108, 0.7437, 0.2700, 0.3650, 0.3683, 0.6237, 0.5717],
  365. [0.6216, 0.4099, 0.7225, 0.2033, 0.4187, 0.2217, 0.5975, 0.5283],
  366. [0.6274, 0.4270, 0.8938, 0.4967, 0.3550, 0.4283, 0.5700, 0.5733],
  367. [0.6236, 0.3965, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  368. [0.6310, 0.4017, 0.8562, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006],
  369. [0.6259, 0.4120, 0.8012, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895],
  370. [0.6202, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  371. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297]]],
  372. device='cuda:0')
  373. loss_train_step before backward: tensor(0.0071, device='cuda:0', grad_fn=<MseLossBackward>)
  374. loss_train_step after backward: tensor(0.0071, device='cuda:0', grad_fn=<MseLossBackward>)
  375. loss_train: 0.5276685687713325
  376. step: 11
  377. running loss: 0.04796986988830296
  378. Train Steps: 11/90 Loss: 0.0480 torch.Size([8, 600, 800])
  379. torch.Size([8, 8])
  380. tensor([[0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5463, 0.5800],
  381. [0.6361, 0.4076, 0.8862, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297],
  382. [0.6203, 0.4096, 0.8862, 0.4267, 0.3538, 0.4117, 0.6025, 0.5650],
  383. [0.6109, 0.4015, 0.7668, 0.3639, 0.3513, 0.3667, 0.5200, 0.5641],
  384. [0.6201, 0.4064, 0.8688, 0.5050, 0.4225, 0.5100, 0.6138, 0.5500],
  385. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  386. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6188, 0.5283],
  387. [ nan, nan, 0.6469, 0.1943, 0.4025, 0.2000, 0.5125, 0.5533]],
  388. device='cuda:0', dtype=torch.float64)
  389. predictions are: tensor([[0.4242, 0.3276, 0.7231, 0.3277, 0.3530, 0.2147, 0.5563, 0.5227],
  390. [0.5232, 0.4021, 0.7501, 0.5769, 0.3382, 0.5162, 0.6495, 0.5368],
  391. [0.4466, 0.3568, 0.7542, 0.4222, 0.3525, 0.2569, 0.5686, 0.5441],
  392. [0.4663, 0.3350, 0.7140, 0.4204, 0.3233, 0.3100, 0.5935, 0.5369],
  393. [0.5148, 0.3751, 0.7542, 0.5413, 0.3297, 0.4700, 0.6317, 0.5378],
  394. [0.4329, 0.3714, 0.7269, 0.3483, 0.3543, 0.2071, 0.5617, 0.5350],
  395. [0.4669, 0.3490, 0.7035, 0.3924, 0.3518, 0.2807, 0.5944, 0.5369],
  396. [0.4827, 0.3401, 0.7154, 0.3492, 0.3307, 0.2902, 0.5846, 0.5344]],
  397. device='cuda:0', grad_fn=<AddmmBackward>)
  398. landmarks are: tensor([[[0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5462, 0.5800],
  399. [0.6361, 0.4076, 0.8863, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297],
  400. [0.6203, 0.4096, 0.8863, 0.4267, 0.3537, 0.4117, 0.6025, 0.5650],
  401. [0.6109, 0.4015, 0.7668, 0.3639, 0.3512, 0.3667, 0.5200, 0.5641],
  402. [0.6201, 0.4064, 0.8687, 0.5050, 0.4225, 0.5100, 0.6137, 0.5500],
  403. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  404. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6187, 0.5283],
  405. [0.0000, 0.0000, 0.6469, 0.1943, 0.4025, 0.2000, 0.5125, 0.5533]]],
  406. device='cuda:0')
  407. loss_train_step before backward: tensor(0.0119, device='cuda:0', grad_fn=<MseLossBackward>)
  408. loss_train_step after backward: tensor(0.0119, device='cuda:0', grad_fn=<MseLossBackward>)
  409. loss_train: 0.5395649061538279
  410. step: 12
  411. running loss: 0.044963742179485656
  412.  
  413. Train Steps: 12/90 Loss: 0.0450 torch.Size([8, 600, 800])
  414. torch.Size([8, 8])
  415. tensor([[0.6203, 0.4096, 0.8862, 0.4267, 0.3538, 0.4117, 0.6025, 0.5650],
  416. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  417. [0.6250, 0.4054, 0.8770, 0.4723, 0.4662, 0.5367, 0.6162, 0.5433],
  418. [0.6350, 0.4043, 0.8738, 0.5650, 0.3850, 0.4750, 0.6401, 0.4950],
  419. [0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  420. [0.6082, 0.4024, 0.8738, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  421. [0.6040, 0.4002, 0.7338, 0.2267, 0.3975, 0.2100, 0.5231, 0.4778],
  422. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334]],
  423. device='cuda:0', dtype=torch.float64)
  424. predictions are: tensor([[0.4855, 0.3495, 0.7835, 0.4110, 0.3501, 0.2785, 0.5376, 0.5742],
  425. [0.4499, 0.3128, 0.7569, 0.3151, 0.3372, 0.1992, 0.5607, 0.5833],
  426. [0.5349, 0.3750, 0.7566, 0.4635, 0.3470, 0.4020, 0.6041, 0.5436],
  427. [0.5596, 0.3703, 0.7619, 0.5797, 0.3526, 0.4759, 0.5958, 0.5680],
  428. [0.5413, 0.3735, 0.7778, 0.4743, 0.3496, 0.4225, 0.5655, 0.5585],
  429. [0.4563, 0.3362, 0.7333, 0.3678, 0.3327, 0.2827, 0.5468, 0.5548],
  430. [0.4906, 0.3384, 0.7265, 0.3375, 0.3288, 0.2491, 0.5441, 0.5544],
  431. [0.5201, 0.3942, 0.7543, 0.4075, 0.3626, 0.3394, 0.5407, 0.5594]],
  432. device='cuda:0', grad_fn=<AddmmBackward>)
  433. landmarks are: tensor([[[0.6203, 0.4096, 0.8863, 0.4267, 0.3537, 0.4117, 0.6025, 0.5650],
  434. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  435. [0.6250, 0.4054, 0.8770, 0.4723, 0.4663, 0.5367, 0.6162, 0.5433],
  436. [0.6350, 0.4043, 0.8737, 0.5650, 0.3850, 0.4750, 0.6401, 0.4950],
  437. [0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  438. [0.6082, 0.4024, 0.8737, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  439. [0.6040, 0.4002, 0.7337, 0.2267, 0.3975, 0.2100, 0.5231, 0.4778],
  440. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334]]],
  441. device='cuda:0')
  442. loss_train_step before backward: tensor(0.0066, device='cuda:0', grad_fn=<MseLossBackward>)
  443. loss_train_step after backward: tensor(0.0066, device='cuda:0', grad_fn=<MseLossBackward>)
  444. loss_train: 0.5461384528316557
  445. step: 13
  446. running loss: 0.042010650217819676
  447. Train Steps: 13/90 Loss: 0.0420 torch.Size([8, 600, 800])
  448. torch.Size([8, 8])
  449. tensor([[0.6339, 0.4149, 0.8800, 0.5000, 0.3900, 0.5283, 0.7541, 0.5424],
  450. [0.6205, 0.4016, 0.8350, 0.2717, 0.3987, 0.2550, 0.5787, 0.5133],
  451. [0.6160, 0.4093, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  452. [0.6273, 0.4143, 0.8750, 0.5700, 0.3987, 0.4717, 0.6013, 0.5467],
  453. [0.6064, 0.4019, 0.8650, 0.4517, 0.4037, 0.5367, 0.5703, 0.5609],
  454. [0.6284, 0.4093, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  455. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  456. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947]],
  457. device='cuda:0', dtype=torch.float64)
  458. predictions are: tensor([[0.6138, 0.4089, 0.8334, 0.4689, 0.3602, 0.4501, 0.5896, 0.5509],
  459. [0.5220, 0.3308, 0.7897, 0.2446, 0.3819, 0.1887, 0.5110, 0.5884],
  460. [0.5544, 0.3636, 0.8083, 0.4162, 0.3470, 0.3582, 0.5382, 0.5543],
  461. [0.5467, 0.3849, 0.8156, 0.3669, 0.3635, 0.3404, 0.5649, 0.5676],
  462. [0.5364, 0.3596, 0.8030, 0.4386, 0.3820, 0.3932, 0.5638, 0.5476],
  463. [0.5485, 0.3434, 0.7971, 0.3553, 0.3534, 0.2809, 0.5138, 0.5525],
  464. [0.5894, 0.3773, 0.8297, 0.4457, 0.3587, 0.4137, 0.5796, 0.5698],
  465. [0.5982, 0.4032, 0.8306, 0.4746, 0.3600, 0.4449, 0.5974, 0.5406]],
  466. device='cuda:0', grad_fn=<AddmmBackward>)
  467. landmarks are: tensor([[[0.6339, 0.4149, 0.8800, 0.5000, 0.3900, 0.5283, 0.7541, 0.5424],
  468. [0.6205, 0.4015, 0.8350, 0.2717, 0.3988, 0.2550, 0.5788, 0.5133],
  469. [0.6160, 0.4092, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  470. [0.6273, 0.4143, 0.8750, 0.5700, 0.3988, 0.4717, 0.6012, 0.5467],
  471. [0.6064, 0.4019, 0.8650, 0.4517, 0.4038, 0.5367, 0.5703, 0.5609],
  472. [0.6284, 0.4092, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  473. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  474. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947]]],
  475. device='cuda:0')
  476. loss_train_step before backward: tensor(0.0045, device='cuda:0', grad_fn=<MseLossBackward>)
  477. loss_train_step after backward: tensor(0.0045, device='cuda:0', grad_fn=<MseLossBackward>)
  478. loss_train: 0.5506306039169431
  479. step: 14
  480. running loss: 0.03933075742263879
  481. Train Steps: 14/90 Loss: 0.0393 torch.Size([8, 600, 800])
  482. torch.Size([8, 8])
  483. tensor([[ nan, nan, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  484. [ nan, nan, 0.6512, 0.1717, 0.4100, 0.1983, 0.5253, 0.5240],
  485. [0.6239, 0.4123, 0.8313, 0.2550, 0.4500, 0.2050, 0.6175, 0.5400],
  486. [0.6111, 0.3995, 0.8788, 0.4567, 0.3813, 0.4833, 0.5450, 0.5700],
  487. [0.6308, 0.3990, 0.8688, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  488. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  489. [0.6353, 0.4128, 0.8488, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  490. [0.6299, 0.4008, 0.8450, 0.5350, 0.4213, 0.5000, 0.6350, 0.5100]],
  491. device='cuda:0', dtype=torch.float64)
  492. predictions are: tensor([[0.5937, 0.3509, 0.8326, 0.3062, 0.3551, 0.3064, 0.5233, 0.5584],
  493. [0.5701, 0.3368, 0.8274, 0.2523, 0.3597, 0.2419, 0.5122, 0.5743],
  494. [0.5733, 0.3884, 0.8194, 0.2436, 0.4069, 0.2326, 0.5367, 0.5481],
  495. [0.6927, 0.4126, 0.8742, 0.5447, 0.3724, 0.5729, 0.6022, 0.5212],
  496. [0.6857, 0.4112, 0.8971, 0.5565, 0.3780, 0.5510, 0.5959, 0.5286],
  497. [0.6102, 0.3915, 0.8287, 0.3693, 0.3949, 0.3350, 0.5436, 0.5626],
  498. [0.5731, 0.3627, 0.8294, 0.3020, 0.4315, 0.2867, 0.5511, 0.5545],
  499. [0.6631, 0.4388, 0.8668, 0.5153, 0.3947, 0.5607, 0.6025, 0.5215]],
  500. device='cuda:0', grad_fn=<AddmmBackward>)
  501. landmarks are: tensor([[[0.0000, 0.0000, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  502. [0.0000, 0.0000, 0.6513, 0.1717, 0.4100, 0.1983, 0.5253, 0.5240],
  503. [0.6239, 0.4123, 0.8313, 0.2550, 0.4500, 0.2050, 0.6175, 0.5400],
  504. [0.6111, 0.3995, 0.8788, 0.4567, 0.3812, 0.4833, 0.5450, 0.5700],
  505. [0.6308, 0.3990, 0.8687, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  506. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  507. [0.6353, 0.4128, 0.8487, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  508. [0.6299, 0.4008, 0.8450, 0.5350, 0.4212, 0.5000, 0.6350, 0.5100]]],
  509. device='cuda:0')
  510. loss_train_step before backward: tensor(0.0177, device='cuda:0', grad_fn=<MseLossBackward>)
  511. loss_train_step after backward: tensor(0.0177, device='cuda:0', grad_fn=<MseLossBackward>)
  512. loss_train: 0.5683316038921475
  513. step: 15
  514. running loss: 0.03788877359280984
  515. Train Steps: 15/90 Loss: 0.0379 torch.Size([8, 600, 800])
  516. torch.Size([8, 8])
  517. tensor([[0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  518. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  519. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  520. [0.6200, 0.3993, 0.8519, 0.4923, 0.3962, 0.4717, 0.6013, 0.5433],
  521. [ nan, nan, 0.8938, 0.2850, 0.4662, 0.3117, 0.7406, 0.5528],
  522. [0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921],
  523. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  524. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285]],
  525. device='cuda:0', dtype=torch.float64)
  526. predictions are: tensor([[0.6547, 0.4026, 0.9127, 0.4880, 0.4210, 0.5077, 0.6168, 0.5180],
  527. [0.6555, 0.3753, 0.8744, 0.4401, 0.4125, 0.4705, 0.6017, 0.5382],
  528. [0.5861, 0.3916, 0.8767, 0.2952, 0.4540, 0.3381, 0.5853, 0.5346],
  529. [0.6865, 0.4021, 0.9131, 0.5099, 0.3946, 0.5068, 0.6111, 0.5421],
  530. [0.6070, 0.3931, 0.8990, 0.3118, 0.4520, 0.3536, 0.5994, 0.5259],
  531. [0.6044, 0.3817, 0.8638, 0.2735, 0.4256, 0.2824, 0.5731, 0.5519],
  532. [0.6471, 0.3808, 0.8958, 0.4194, 0.4224, 0.4481, 0.5902, 0.5492],
  533. [0.6469, 0.3749, 0.8770, 0.3937, 0.4066, 0.3749, 0.5731, 0.5123]],
  534. device='cuda:0', grad_fn=<AddmmBackward>)
  535. landmarks are: tensor([[[0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  536. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  537. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  538. [0.6200, 0.3993, 0.8519, 0.4923, 0.3963, 0.4717, 0.6012, 0.5433],
  539. [0.0000, 0.0000, 0.8938, 0.2850, 0.4663, 0.3117, 0.7406, 0.5528],
  540. [0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921],
  541. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  542. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285]]],
  543. device='cuda:0')
  544. loss_train_step before backward: tensor(0.0110, device='cuda:0', grad_fn=<MseLossBackward>)
  545. loss_train_step after backward: tensor(0.0110, device='cuda:0', grad_fn=<MseLossBackward>)
  546. loss_train: 0.5793465515598655
  547. step: 16
  548. running loss: 0.03620915947249159
  549.  
  550. Train Steps: 16/90 Loss: 0.0362 torch.Size([8, 600, 800])
  551. torch.Size([8, 8])
  552. tensor([[0.6198, 0.3997, 0.8582, 0.5361, 0.4117, 0.5016, 0.5942, 0.5134],
  553. [ nan, nan, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  554. [0.6346, 0.4144, 0.9088, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  555. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650],
  556. [0.6072, 0.4029, 0.7037, 0.2150, 0.3912, 0.2267, 0.5516, 0.5507],
  557. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  558. [0.6272, 0.4045, 0.8538, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  559. [0.6273, 0.4143, 0.8750, 0.5700, 0.3987, 0.4717, 0.6013, 0.5467]],
  560. device='cuda:0', dtype=torch.float64)
  561. predictions are: tensor([[0.6627, 0.4136, 0.9392, 0.5165, 0.4523, 0.5500, 0.6682, 0.4953],
  562. [0.5376, 0.3240, 0.8913, 0.2640, 0.4416, 0.3313, 0.5895, 0.5246],
  563. [0.6395, 0.3805, 0.9361, 0.4163, 0.4485, 0.4339, 0.6147, 0.5254],
  564. [0.5680, 0.3594, 0.9200, 0.2960, 0.4363, 0.3497, 0.5919, 0.5198],
  565. [0.5743, 0.3581, 0.8822, 0.2888, 0.4754, 0.3589, 0.6170, 0.5224],
  566. [0.6266, 0.3792, 0.9241, 0.3970, 0.4373, 0.4108, 0.6244, 0.4996],
  567. [0.6290, 0.3848, 0.9422, 0.4540, 0.4268, 0.4926, 0.6146, 0.5069],
  568. [0.6318, 0.4059, 0.9420, 0.4500, 0.4560, 0.4884, 0.6553, 0.5268]],
  569. device='cuda:0', grad_fn=<AddmmBackward>)
  570. landmarks are: tensor([[[0.6198, 0.3997, 0.8582, 0.5361, 0.4117, 0.5016, 0.5942, 0.5134],
  571. [0.0000, 0.0000, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  572. [0.6346, 0.4144, 0.9087, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  573. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650],
  574. [0.6072, 0.4029, 0.7038, 0.2150, 0.3913, 0.2267, 0.5516, 0.5507],
  575. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  576. [0.6271, 0.4045, 0.8537, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  577. [0.6273, 0.4143, 0.8750, 0.5700, 0.3988, 0.4717, 0.6012, 0.5467]]],
  578. device='cuda:0')
  579. loss_train_step before backward: tensor(0.0105, device='cuda:0', grad_fn=<MseLossBackward>)
  580. loss_train_step after backward: tensor(0.0105, device='cuda:0', grad_fn=<MseLossBackward>)
  581. loss_train: 0.5897985212504864
  582. step: 17
  583. running loss: 0.034694030661793315
  584. Train Steps: 17/90 Loss: 0.0347 torch.Size([8, 600, 800])
  585. torch.Size([8, 8])
  586. tensor([[0.6139, 0.4019, 0.7137, 0.2150, 0.4375, 0.1533, 0.5293, 0.5006],
  587. [0.6147, 0.4107, 0.8137, 0.3333, 0.3750, 0.2683, 0.5006, 0.5412],
  588. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6038, 0.4833],
  589. [0.6213, 0.4131, 0.8438, 0.3550, 0.3513, 0.4400, 0.5716, 0.5123],
  590. [0.6187, 0.4104, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  591. [0.6109, 0.4003, 0.8650, 0.4883, 0.4775, 0.4867, 0.5175, 0.5683],
  592. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  593. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617]],
  594. device='cuda:0', dtype=torch.float64)
  595. predictions are: tensor([[0.4989, 0.3019, 0.8501, 0.2146, 0.4736, 0.2612, 0.5998, 0.5084],
  596. [0.5352, 0.3228, 0.8733, 0.2890, 0.4420, 0.2892, 0.5881, 0.5260],
  597. [0.6540, 0.3960, 0.9527, 0.5137, 0.4338, 0.5415, 0.6720, 0.5053],
  598. [0.5784, 0.3489, 0.9258, 0.4035, 0.4336, 0.4318, 0.6304, 0.5092],
  599. [0.5422, 0.3244, 0.9022, 0.2565, 0.4607, 0.3007, 0.6357, 0.5141],
  600. [0.5923, 0.3676, 0.9305, 0.4436, 0.4626, 0.4719, 0.6405, 0.5138],
  601. [0.6616, 0.3972, 0.9572, 0.5468, 0.4315, 0.5406, 0.6886, 0.4973],
  602. [0.6289, 0.3807, 0.9102, 0.4968, 0.4496, 0.4921, 0.6804, 0.4749]],
  603. device='cuda:0', grad_fn=<AddmmBackward>)
  604. landmarks are: tensor([[[0.6139, 0.4019, 0.7138, 0.2150, 0.4375, 0.1533, 0.5293, 0.5006],
  605. [0.6147, 0.4107, 0.8138, 0.3333, 0.3750, 0.2683, 0.5006, 0.5412],
  606. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6037, 0.4833],
  607. [0.6213, 0.4131, 0.8438, 0.3550, 0.3512, 0.4400, 0.5716, 0.5123],
  608. [0.6187, 0.4103, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  609. [0.6109, 0.4003, 0.8650, 0.4883, 0.4775, 0.4867, 0.5175, 0.5683],
  610. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  611. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617]]],
  612. device='cuda:0')
  613. loss_train_step before backward: tensor(0.0042, device='cuda:0', grad_fn=<MseLossBackward>)
  614. loss_train_step after backward: tensor(0.0042, device='cuda:0', grad_fn=<MseLossBackward>)
  615. loss_train: 0.59402234852314
  616. step: 18
  617. running loss: 0.03300124158461889
  618. Train Steps: 18/90 Loss: 0.0330 torch.Size([8, 600, 800])
  619. torch.Size([8, 8])
  620. tensor([[0.6257, 0.4034, 0.8287, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  621. [0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921],
  622. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  623. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864],
  624. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  625. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  626. [0.6265, 0.4091, 0.8950, 0.3533, 0.3600, 0.3967, 0.6295, 0.4901],
  627. [0.6339, 0.4123, 0.8638, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436]],
  628. device='cuda:0', dtype=torch.float64)
  629. predictions are: tensor([[0.5489, 0.3691, 0.8602, 0.3525, 0.4167, 0.3724, 0.6237, 0.5266],
  630. [0.5204, 0.3422, 0.8550, 0.2708, 0.4328, 0.2782, 0.6180, 0.5341],
  631. [0.4912, 0.3260, 0.8521, 0.2680, 0.4374, 0.2647, 0.6071, 0.5039],
  632. [0.6073, 0.3701, 0.8993, 0.4969, 0.4102, 0.4751, 0.6399, 0.5054],
  633. [0.5643, 0.3735, 0.8997, 0.4671, 0.4200, 0.4458, 0.6410, 0.5132],
  634. [0.6377, 0.3858, 0.9165, 0.5669, 0.4071, 0.5251, 0.6724, 0.4992],
  635. [0.5725, 0.3525, 0.8803, 0.3685, 0.4274, 0.3686, 0.6311, 0.5081],
  636. [0.6164, 0.3833, 0.9266, 0.5551, 0.4389, 0.5435, 0.6636, 0.5159]],
  637. device='cuda:0', grad_fn=<AddmmBackward>)
  638. landmarks are: tensor([[[0.6257, 0.4034, 0.8288, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  639. [0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921],
  640. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  641. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864],
  642. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  643. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  644. [0.6265, 0.4091, 0.8950, 0.3533, 0.3600, 0.3967, 0.6295, 0.4901],
  645. [0.6339, 0.4123, 0.8637, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436]]],
  646. device='cuda:0')
  647. loss_train_step before backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  648. loss_train_step after backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  649. loss_train: 0.5967375976033509
  650. step: 19
  651. running loss: 0.03140724197912373
  652. Train Steps: 19/90 Loss: 0.0314 torch.Size([8, 600, 800])
  653. torch.Size([8, 8])
  654. tensor([[0.6149, 0.4054, 0.6713, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  655. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  656. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5363, 0.5550],
  657. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  658. [0.6200, 0.3978, 0.8900, 0.4550, 0.3775, 0.5200, 0.6150, 0.5367],
  659. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  660. [0.6307, 0.4029, 0.8650, 0.5200, 0.3763, 0.4017, 0.7311, 0.5366],
  661. [0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297]],
  662. device='cuda:0', dtype=torch.float64)
  663. predictions are: tensor([[0.5002, 0.3322, 0.7828, 0.2725, 0.3797, 0.2560, 0.5804, 0.5275],
  664. [0.6001, 0.3807, 0.8608, 0.4400, 0.3863, 0.4352, 0.6208, 0.5364],
  665. [0.5115, 0.3182, 0.8096, 0.2736, 0.3886, 0.2493, 0.6107, 0.5172],
  666. [0.5968, 0.4046, 0.8681, 0.5250, 0.4023, 0.4724, 0.6819, 0.5299],
  667. [0.6252, 0.3984, 0.8836, 0.5122, 0.4340, 0.4717, 0.6479, 0.5141],
  668. [0.6535, 0.4172, 0.8849, 0.5469, 0.3984, 0.5496, 0.6890, 0.4915],
  669. [0.5809, 0.3773, 0.8423, 0.3697, 0.3968, 0.3501, 0.6085, 0.5313],
  670. [0.5965, 0.3755, 0.8642, 0.4635, 0.3780, 0.3989, 0.6192, 0.5332]],
  671. device='cuda:0', grad_fn=<AddmmBackward>)
  672. landmarks are: tensor([[[0.6149, 0.4054, 0.6712, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  673. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  674. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5362, 0.5550],
  675. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  676. [0.6199, 0.3978, 0.8900, 0.4550, 0.3775, 0.5200, 0.6150, 0.5367],
  677. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  678. [0.6307, 0.4029, 0.8650, 0.5200, 0.3762, 0.4017, 0.7311, 0.5366],
  679. [0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297]]],
  680. device='cuda:0')
  681. loss_train_step before backward: tensor(0.0033, device='cuda:0', grad_fn=<MseLossBackward>)
  682. loss_train_step after backward: tensor(0.0033, device='cuda:0', grad_fn=<MseLossBackward>)
  683. loss_train: 0.6000654962845147
  684. step: 20
  685. running loss: 0.030003274814225732
  686.  
  687. Train Steps: 20/90 Loss: 0.0300 torch.Size([8, 600, 800])
  688. torch.Size([8, 8])
  689. tensor([[0.6273, 0.4105, 0.8988, 0.4517, 0.3912, 0.2550, 0.5894, 0.4811],
  690. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5637, 0.5633],
  691. [0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133],
  692. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  693. [0.6084, 0.3981, 0.8588, 0.5233, 0.4600, 0.5367, 0.5680, 0.5006],
  694. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  695. [ nan, nan, 0.6512, 0.1717, 0.4100, 0.1983, 0.5253, 0.5240],
  696. [0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090]],
  697. device='cuda:0', dtype=torch.float64)
  698. predictions are: tensor([[0.5916, 0.3844, 0.7633, 0.3877, 0.3732, 0.3227, 0.6046, 0.5243],
  699. [0.6335, 0.4340, 0.7941, 0.4725, 0.3441, 0.4270, 0.6307, 0.5186],
  700. [0.6787, 0.4430, 0.8263, 0.5454, 0.3485, 0.4909, 0.6363, 0.4990],
  701. [0.5771, 0.3782, 0.7535, 0.3500, 0.3570, 0.3527, 0.6062, 0.5452],
  702. [0.6281, 0.4197, 0.8233, 0.5323, 0.3730, 0.4784, 0.6468, 0.5321],
  703. [0.5682, 0.4123, 0.8094, 0.3721, 0.4299, 0.3495, 0.6410, 0.5398],
  704. [0.4449, 0.3098, 0.7400, 0.2139, 0.3699, 0.2040, 0.5648, 0.5597],
  705. [0.6228, 0.4093, 0.8242, 0.4646, 0.3890, 0.4224, 0.6368, 0.5439]],
  706. device='cuda:0', grad_fn=<AddmmBackward>)
  707. landmarks are: tensor([[[0.6273, 0.4105, 0.8988, 0.4517, 0.3913, 0.2550, 0.5894, 0.4811],
  708. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5638, 0.5633],
  709. [0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133],
  710. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  711. [0.6084, 0.3981, 0.8587, 0.5233, 0.4600, 0.5367, 0.5680, 0.5006],
  712. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  713. [0.0000, 0.0000, 0.6513, 0.1717, 0.4100, 0.1983, 0.5253, 0.5240],
  714. [0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090]]],
  715. device='cuda:0')
  716. loss_train_step before backward: tensor(0.0075, device='cuda:0', grad_fn=<MseLossBackward>)
  717. loss_train_step after backward: tensor(0.0075, device='cuda:0', grad_fn=<MseLossBackward>)
  718. loss_train: 0.6075229728594422
  719. step: 21
  720. running loss: 0.028929665374259155
  721. Train Steps: 21/90 Loss: 0.0289 torch.Size([8, 600, 800])
  722. torch.Size([8, 8])
  723. tensor([[0.6159, 0.4085, 0.6900, 0.2283, 0.4088, 0.1950, 0.5123, 0.5397],
  724. [0.6272, 0.4045, 0.8538, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  725. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  726. [0.6257, 0.4167, 0.8775, 0.3433, 0.3563, 0.4133, 0.6200, 0.5667],
  727. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  728. [0.6100, 0.4071, 0.7601, 0.3444, 0.3400, 0.4117, 0.5625, 0.5617],
  729. [0.6277, 0.4103, 0.8087, 0.5717, 0.4188, 0.4750, 0.5663, 0.6083],
  730. [0.6124, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485]],
  731. device='cuda:0', dtype=torch.float64)
  732. predictions are: tensor([[0.5361, 0.3473, 0.7224, 0.2576, 0.3896, 0.2200, 0.5687, 0.5548],
  733. [0.6341, 0.4344, 0.8141, 0.4878, 0.3341, 0.4040, 0.5881, 0.5413],
  734. [0.6328, 0.4104, 0.7829, 0.5221, 0.3337, 0.4532, 0.6095, 0.5360],
  735. [0.6438, 0.4458, 0.8056, 0.4583, 0.3897, 0.4038, 0.6523, 0.5541],
  736. [0.6684, 0.4341, 0.8065, 0.5662, 0.3743, 0.4831, 0.6147, 0.5662],
  737. [0.5448, 0.4118, 0.7564, 0.3263, 0.3703, 0.3139, 0.5909, 0.5578],
  738. [0.6425, 0.4230, 0.8266, 0.5468, 0.3626, 0.4794, 0.6113, 0.5741],
  739. [0.5344, 0.3745, 0.7279, 0.2517, 0.3772, 0.2505, 0.5578, 0.5618]],
  740. device='cuda:0', grad_fn=<AddmmBackward>)
  741. landmarks are: tensor([[[0.6159, 0.4085, 0.6900, 0.2283, 0.4087, 0.1950, 0.5123, 0.5397],
  742. [0.6271, 0.4045, 0.8537, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  743. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  744. [0.6257, 0.4167, 0.8775, 0.3433, 0.3562, 0.4133, 0.6200, 0.5667],
  745. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  746. [0.6100, 0.4071, 0.7601, 0.3444, 0.3400, 0.4117, 0.5625, 0.5617],
  747. [0.6277, 0.4103, 0.8087, 0.5717, 0.4187, 0.4750, 0.5663, 0.6083],
  748. [0.6123, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485]]],
  749. device='cuda:0')
  750. loss_train_step before backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  751. loss_train_step after backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  752. loss_train: 0.6093606404028833
  753. step: 22
  754. running loss: 0.027698210927403787
  755. Train Steps: 22/90 Loss: 0.0277 torch.Size([8, 600, 800])
  756. torch.Size([8, 8])
  757. tensor([[0.6332, 0.4118, 0.9238, 0.4267, 0.4012, 0.4733, 0.7525, 0.5436],
  758. [0.6128, 0.4118, 0.8638, 0.5333, 0.4625, 0.5267, 0.5193, 0.5475],
  759. [0.6245, 0.4115, 0.8700, 0.4883, 0.4625, 0.5517, 0.6100, 0.5217],
  760. [0.6276, 0.4120, 0.8738, 0.3133, 0.4225, 0.2217, 0.6203, 0.4892],
  761. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  762. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  763. [0.6257, 0.4167, 0.8775, 0.3433, 0.3563, 0.4133, 0.6200, 0.5667],
  764. [0.6064, 0.3953, 0.8738, 0.4417, 0.3663, 0.4683, 0.5511, 0.5416]],
  765. device='cuda:0', dtype=torch.float64)
  766. predictions are: tensor([[0.6037, 0.4042, 0.7861, 0.4525, 0.3799, 0.4002, 0.5853, 0.5947],
  767. [0.6158, 0.3983, 0.7456, 0.4208, 0.3749, 0.3494, 0.5616, 0.5747],
  768. [0.5826, 0.3949, 0.7472, 0.4460, 0.3831, 0.4107, 0.5832, 0.5636],
  769. [0.5835, 0.4133, 0.7313, 0.3065, 0.4241, 0.2709, 0.5836, 0.5425],
  770. [0.6116, 0.4081, 0.7711, 0.4174, 0.3359, 0.3233, 0.5526, 0.5699],
  771. [0.6512, 0.4110, 0.7677, 0.5061, 0.3607, 0.4228, 0.5736, 0.5367],
  772. [0.6322, 0.4209, 0.7510, 0.4096, 0.3741, 0.3559, 0.6041, 0.5573],
  773. [0.6771, 0.3982, 0.7817, 0.4814, 0.3504, 0.4184, 0.5547, 0.5450]],
  774. device='cuda:0', grad_fn=<AddmmBackward>)
  775. landmarks are: tensor([[[0.6332, 0.4118, 0.9237, 0.4267, 0.4013, 0.4733, 0.7525, 0.5436],
  776. [0.6128, 0.4118, 0.8637, 0.5333, 0.4625, 0.5267, 0.5193, 0.5475],
  777. [0.6245, 0.4115, 0.8700, 0.4883, 0.4625, 0.5517, 0.6100, 0.5217],
  778. [0.6276, 0.4120, 0.8737, 0.3133, 0.4225, 0.2217, 0.6203, 0.4892],
  779. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  780. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  781. [0.6257, 0.4167, 0.8775, 0.3433, 0.3562, 0.4133, 0.6200, 0.5667],
  782. [0.6064, 0.3952, 0.8737, 0.4417, 0.3663, 0.4683, 0.5511, 0.5416]]],
  783. device='cuda:0')
  784. loss_train_step before backward: tensor(0.0047, device='cuda:0', grad_fn=<MseLossBackward>)
  785. loss_train_step after backward: tensor(0.0047, device='cuda:0', grad_fn=<MseLossBackward>)
  786. loss_train: 0.6140206111595035
  787. step: 23
  788. running loss: 0.02669654831128276
  789. Train Steps: 23/90 Loss: 0.0267 torch.Size([8, 600, 800])
  790. torch.Size([8, 8])
  791. tensor([[0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431],
  792. [ nan, nan, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733],
  793. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  794. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  795. [0.6189, 0.4029, 0.8375, 0.5767, 0.4745, 0.4829, 0.5551, 0.5598],
  796. [0.6199, 0.4102, 0.8950, 0.4417, 0.4012, 0.5367, 0.6112, 0.5967],
  797. [0.6200, 0.4118, 0.8287, 0.4017, 0.3775, 0.2833, 0.5391, 0.5799],
  798. [ nan, nan, 0.7512, 0.2117, 0.4288, 0.2000, 0.5600, 0.5617]],
  799. device='cuda:0', dtype=torch.float64)
  800. predictions are: tensor([[0.6534, 0.4442, 0.7870, 0.5273, 0.3688, 0.4122, 0.5606, 0.5491],
  801. [0.4899, 0.3300, 0.6995, 0.1907, 0.4047, 0.1847, 0.5049, 0.5413],
  802. [0.5462, 0.3605, 0.7138, 0.3059, 0.3777, 0.2765, 0.5254, 0.5580],
  803. [0.6595, 0.4666, 0.8279, 0.5287, 0.3715, 0.4721, 0.6088, 0.5316],
  804. [0.6401, 0.4335, 0.7881, 0.5154, 0.3938, 0.4018, 0.5882, 0.5655],
  805. [0.7212, 0.4738, 0.8581, 0.6698, 0.3993, 0.5994, 0.6318, 0.5457],
  806. [0.5917, 0.3890, 0.7234, 0.3395, 0.3801, 0.2883, 0.5364, 0.5730],
  807. [0.5050, 0.3630, 0.7010, 0.2187, 0.3996, 0.2131, 0.5094, 0.5673]],
  808. device='cuda:0', grad_fn=<AddmmBackward>)
  809. landmarks are: tensor([[[0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431],
  810. [0.0000, 0.0000, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733],
  811. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  812. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  813. [0.6189, 0.4029, 0.8375, 0.5767, 0.4745, 0.4829, 0.5551, 0.5598],
  814. [0.6199, 0.4102, 0.8950, 0.4417, 0.4013, 0.5367, 0.6112, 0.5967],
  815. [0.6200, 0.4118, 0.8288, 0.4017, 0.3775, 0.2833, 0.5391, 0.5799],
  816. [0.0000, 0.0000, 0.7513, 0.2117, 0.4288, 0.2000, 0.5600, 0.5617]]],
  817. device='cuda:0')
  818. loss_train_step before backward: tensor(0.0145, device='cuda:0', grad_fn=<MseLossBackward>)
  819. loss_train_step after backward: tensor(0.0145, device='cuda:0', grad_fn=<MseLossBackward>)
  820. loss_train: 0.628472588956356
  821. step: 24
  822. running loss: 0.026186357873181503
  823.  
  824. Train Steps: 24/90 Loss: 0.0262 torch.Size([8, 600, 800])
  825. torch.Size([8, 8])
  826. tensor([[0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  827. [0.6286, 0.4055, 0.9000, 0.4717, 0.3763, 0.4683, 0.7018, 0.5494],
  828. [ nan, nan, 0.7335, 0.2569, 0.3788, 0.2667, 0.5066, 0.5578],
  829. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  830. [ nan, nan, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  831. [0.6307, 0.4029, 0.8988, 0.4817, 0.3937, 0.3500, 0.7311, 0.5378],
  832. [0.6153, 0.4117, 0.8688, 0.5167, 0.4895, 0.5647, 0.5524, 0.5136],
  833. [0.6149, 0.4054, 0.6713, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695]],
  834. device='cuda:0', dtype=torch.float64)
  835. predictions are: tensor([[0.6083, 0.4021, 0.8504, 0.4506, 0.4062, 0.4482, 0.5526, 0.5495],
  836. [0.6550, 0.4405, 0.8656, 0.6100, 0.3859, 0.5233, 0.5851, 0.5481],
  837. [0.4780, 0.3301, 0.7199, 0.2762, 0.4161, 0.2419, 0.4989, 0.5769],
  838. [0.5234, 0.3830, 0.8144, 0.3121, 0.4428, 0.3014, 0.5504, 0.5798],
  839. [0.4204, 0.3060, 0.7255, 0.2233, 0.4225, 0.1847, 0.5139, 0.5718],
  840. [0.6194, 0.4022, 0.8339, 0.4905, 0.4203, 0.3756, 0.5566, 0.5277],
  841. [0.5894, 0.4061, 0.8469, 0.5381, 0.4337, 0.5094, 0.5568, 0.5652],
  842. [0.4978, 0.3319, 0.7075, 0.3023, 0.4139, 0.2716, 0.4750, 0.5493]],
  843. device='cuda:0', grad_fn=<AddmmBackward>)
  844. landmarks are: tensor([[[0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  845. [0.6286, 0.4055, 0.9000, 0.4717, 0.3762, 0.4683, 0.7018, 0.5494],
  846. [0.0000, 0.0000, 0.7335, 0.2569, 0.3787, 0.2667, 0.5066, 0.5578],
  847. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  848. [0.0000, 0.0000, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  849. [0.6307, 0.4029, 0.8988, 0.4817, 0.3938, 0.3500, 0.7311, 0.5378],
  850. [0.6154, 0.4117, 0.8687, 0.5167, 0.4895, 0.5647, 0.5524, 0.5136],
  851. [0.6149, 0.4054, 0.6712, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695]]],
  852. device='cuda:0')
  853. loss_train_step before backward: tensor(0.0125, device='cuda:0', grad_fn=<MseLossBackward>)
  854. loss_train_step after backward: tensor(0.0125, device='cuda:0', grad_fn=<MseLossBackward>)
  855. loss_train: 0.6410121768712997
  856. step: 25
  857. running loss: 0.02564048707485199
  858. Train Steps: 25/90 Loss: 0.0256 torch.Size([8, 600, 800])
  859. torch.Size([8, 8])
  860. tensor([[0.6268, 0.4102, 0.8938, 0.3667, 0.4025, 0.2833, 0.6275, 0.5183],
  861. [ nan, nan, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  862. [0.6286, 0.4078, 0.8063, 0.2267, 0.4788, 0.1533, 0.5953, 0.4913],
  863. [0.6111, 0.4019, 0.8350, 0.3867, 0.3500, 0.4283, 0.5480, 0.5719],
  864. [0.6199, 0.4071, 0.7600, 0.2117, 0.4037, 0.2767, 0.6138, 0.5550],
  865. [0.6134, 0.4090, 0.6926, 0.2819, 0.3538, 0.3233, 0.5563, 0.5667],
  866. [0.6199, 0.4102, 0.8950, 0.4417, 0.4012, 0.5367, 0.6112, 0.5967],
  867. [0.6278, 0.4253, 0.8875, 0.5017, 0.4113, 0.2750, 0.5413, 0.6196]],
  868. device='cuda:0', dtype=torch.float64)
  869. predictions are: tensor([[0.5108, 0.3604, 0.8390, 0.3908, 0.4406, 0.3910, 0.5886, 0.5432],
  870. [0.4525, 0.3217, 0.8464, 0.3252, 0.4609, 0.3114, 0.5518, 0.5576],
  871. [0.3797, 0.2495, 0.7300, 0.1931, 0.4586, 0.1687, 0.5035, 0.5400],
  872. [0.5461, 0.3603, 0.8976, 0.4546, 0.4051, 0.4570, 0.5562, 0.5573],
  873. [0.4287, 0.3064, 0.7561, 0.2479, 0.4488, 0.2206, 0.5375, 0.5389],
  874. [0.4644, 0.2900, 0.7424, 0.2811, 0.3841, 0.3158, 0.4938, 0.5445],
  875. [0.6190, 0.3835, 0.9274, 0.5857, 0.4378, 0.6095, 0.5924, 0.5434],
  876. [0.5422, 0.3625, 0.8487, 0.4538, 0.4157, 0.4287, 0.5251, 0.5353]],
  877. device='cuda:0', grad_fn=<AddmmBackward>)
  878. landmarks are: tensor([[[0.6268, 0.4102, 0.8938, 0.3667, 0.4025, 0.2833, 0.6275, 0.5183],
  879. [0.0000, 0.0000, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  880. [0.6286, 0.4078, 0.8062, 0.2267, 0.4787, 0.1533, 0.5953, 0.4913],
  881. [0.6111, 0.4019, 0.8350, 0.3867, 0.3500, 0.4283, 0.5480, 0.5719],
  882. [0.6199, 0.4071, 0.7600, 0.2117, 0.4038, 0.2767, 0.6137, 0.5550],
  883. [0.6134, 0.4090, 0.6926, 0.2819, 0.3537, 0.3233, 0.5562, 0.5667],
  884. [0.6199, 0.4102, 0.8950, 0.4417, 0.4013, 0.5367, 0.6112, 0.5967],
  885. [0.6278, 0.4253, 0.8875, 0.5017, 0.4112, 0.2750, 0.5413, 0.6196]]],
  886. device='cuda:0')
  887. loss_train_step before backward: tensor(0.0109, device='cuda:0', grad_fn=<MseLossBackward>)
  888. loss_train_step after backward: tensor(0.0109, device='cuda:0', grad_fn=<MseLossBackward>)
  889. loss_train: 0.6518932655453682
  890. step: 26
  891. running loss: 0.025072817905591085
  892. Train Steps: 26/90 Loss: 0.0251 torch.Size([8, 600, 800])
  893. torch.Size([8, 8])
  894. tensor([[0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  895. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6362, 0.5200],
  896. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  897. [0.6311, 0.3998, 0.7975, 0.5767, 0.3838, 0.4850, 0.7327, 0.5343],
  898. [ nan, nan, 0.7850, 0.2700, 0.4288, 0.1717, 0.5199, 0.4999],
  899. [0.6213, 0.4001, 0.7712, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183],
  900. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  901. [0.6189, 0.4049, 0.8888, 0.4417, 0.4213, 0.5200, 0.5988, 0.5633]],
  902. device='cuda:0', dtype=torch.float64)
  903. predictions are: tensor([[0.5346, 0.3361, 0.9201, 0.4234, 0.4331, 0.4263, 0.5722, 0.5859],
  904. [0.5568, 0.3413, 0.9471, 0.4525, 0.4453, 0.4915, 0.5743, 0.5571],
  905. [0.3841, 0.2736, 0.8017, 0.2267, 0.4508, 0.2552, 0.5554, 0.5411],
  906. [0.4868, 0.3168, 0.9052, 0.4018, 0.4096, 0.3981, 0.5696, 0.5464],
  907. [0.3279, 0.2467, 0.8089, 0.1724, 0.4432, 0.1667, 0.5400, 0.5439],
  908. [0.3911, 0.2765, 0.7820, 0.1967, 0.4317, 0.2572, 0.5715, 0.5536],
  909. [0.4877, 0.3180, 0.8559, 0.3810, 0.4573, 0.3899, 0.5876, 0.5489],
  910. [0.5775, 0.3616, 0.9427, 0.5111, 0.4411, 0.5220, 0.6233, 0.5373]],
  911. device='cuda:0', grad_fn=<AddmmBackward>)
  912. landmarks are: tensor([[[0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  913. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6363, 0.5200],
  914. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  915. [0.6311, 0.3998, 0.7975, 0.5767, 0.3837, 0.4850, 0.7327, 0.5343],
  916. [0.0000, 0.0000, 0.7850, 0.2700, 0.4288, 0.1717, 0.5199, 0.4999],
  917. [0.6213, 0.4001, 0.7713, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183],
  918. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  919. [0.6189, 0.4049, 0.8888, 0.4417, 0.4212, 0.5200, 0.5987, 0.5633]]],
  920. device='cuda:0')
  921. loss_train_step before backward: tensor(0.0092, device='cuda:0', grad_fn=<MseLossBackward>)
  922. loss_train_step after backward: tensor(0.0092, device='cuda:0', grad_fn=<MseLossBackward>)
  923. loss_train: 0.6610878705978394
  924. step: 27
  925. running loss: 0.024484735948068125
  926. Train Steps: 27/90 Loss: 0.0245 torch.Size([8, 600, 800])
  927. torch.Size([8, 8])
  928. tensor([[0.6132, 0.4066, 0.7259, 0.2402, 0.3588, 0.3300, 0.6000, 0.5600],
  929. [0.6085, 0.4008, 0.8588, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  930. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  931. [0.6200, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  932. [0.6272, 0.4045, 0.8538, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  933. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  934. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  935. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646]],
  936. device='cuda:0', dtype=torch.float64)
  937. predictions are: tensor([[0.4143, 0.2747, 0.8007, 0.2348, 0.4075, 0.2985, 0.5870, 0.5697],
  938. [0.4828, 0.2987, 0.9111, 0.4216, 0.4440, 0.4238, 0.6115, 0.5584],
  939. [0.3344, 0.2688, 0.8159, 0.2095, 0.4573, 0.1910, 0.6030, 0.5582],
  940. [0.5427, 0.3206, 0.9698, 0.4371, 0.4166, 0.4546, 0.6427, 0.5454],
  941. [0.5437, 0.3577, 0.9476, 0.4925, 0.3931, 0.4746, 0.6047, 0.5349],
  942. [0.3526, 0.2414, 0.7833, 0.1797, 0.4681, 0.2100, 0.6073, 0.5477],
  943. [0.4671, 0.3069, 0.9184, 0.3883, 0.4627, 0.3664, 0.6575, 0.5600],
  944. [0.4275, 0.2836, 0.8408, 0.2699, 0.4258, 0.3069, 0.5922, 0.5463]],
  945. device='cuda:0', grad_fn=<AddmmBackward>)
  946. landmarks are: tensor([[[0.6132, 0.4066, 0.7259, 0.2402, 0.3587, 0.3300, 0.6000, 0.5600],
  947. [0.6084, 0.4008, 0.8587, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  948. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  949. [0.6201, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  950. [0.6271, 0.4045, 0.8537, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  951. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  952. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  953. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646]]],
  954. device='cuda:0')
  955. loss_train_step before backward: tensor(0.0080, device='cuda:0', grad_fn=<MseLossBackward>)
  956. loss_train_step after backward: tensor(0.0080, device='cuda:0', grad_fn=<MseLossBackward>)
  957. loss_train: 0.6691014468669891
  958. step: 28
  959. running loss: 0.023896480245249613
  960.  
  961. Train Steps: 28/90 Loss: 0.0239 torch.Size([8, 600, 800])
  962. torch.Size([8, 8])
  963. tensor([[0.6109, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  964. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  965. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  966. [0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510],
  967. [0.6241, 0.4143, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  968. [0.6203, 0.4078, 0.8800, 0.5083, 0.3900, 0.5000, 0.6100, 0.5583],
  969. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  970. [0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726]],
  971. device='cuda:0', dtype=torch.float64)
  972. predictions are: tensor([[0.5390, 0.3826, 0.9603, 0.4170, 0.3944, 0.4088, 0.6443, 0.5350],
  973. [0.4486, 0.2729, 0.8097, 0.2505, 0.4108, 0.2737, 0.6175, 0.5547],
  974. [0.4647, 0.3007, 0.8495, 0.2775, 0.4124, 0.2690, 0.6329, 0.5355],
  975. [0.4015, 0.2568, 0.7904, 0.2029, 0.3834, 0.2554, 0.5896, 0.5526],
  976. [0.5965, 0.3690, 0.9675, 0.5167, 0.4347, 0.5499, 0.7048, 0.5469],
  977. [0.6405, 0.3905, 0.9769, 0.5324, 0.3739, 0.5301, 0.6639, 0.5552],
  978. [0.4013, 0.2692, 0.8289, 0.1844, 0.4672, 0.1578, 0.6531, 0.5266],
  979. [0.4481, 0.2940, 0.8121, 0.2447, 0.4270, 0.2471, 0.6292, 0.5344]],
  980. device='cuda:0', grad_fn=<AddmmBackward>)
  981. landmarks are: tensor([[[0.6108, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  982. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  983. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  984. [0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510],
  985. [0.6241, 0.4142, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  986. [0.6203, 0.4078, 0.8800, 0.5083, 0.3900, 0.5000, 0.6100, 0.5583],
  987. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  988. [0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726]]],
  989. device='cuda:0')
  990. loss_train_step before backward: tensor(0.0062, device='cuda:0', grad_fn=<MseLossBackward>)
  991. loss_train_step after backward: tensor(0.0062, device='cuda:0', grad_fn=<MseLossBackward>)
  992. loss_train: 0.6753336312249303
  993. step: 29
  994. running loss: 0.02328736659396311
  995. Train Steps: 29/90 Loss: 0.0233 torch.Size([8, 600, 800])
  996. torch.Size([8, 8])
  997. tensor([[0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  998. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  999. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  1000. [0.6111, 0.4033, 0.8300, 0.3267, 0.3588, 0.3333, 0.5444, 0.5637],
  1001. [0.6250, 0.4106, 0.8700, 0.3717, 0.3588, 0.4967, 0.6038, 0.5167],
  1002. [0.6293, 0.3982, 0.8700, 0.5300, 0.3763, 0.4717, 0.7050, 0.5297],
  1003. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6188, 0.5283],
  1004. [0.6250, 0.4103, 0.8950, 0.4400, 0.3912, 0.5650, 0.6050, 0.5133]],
  1005. device='cuda:0', dtype=torch.float64)
  1006. predictions are: tensor([[0.6369, 0.4089, 0.8763, 0.3559, 0.3884, 0.3804, 0.6565, 0.5152],
  1007. [0.5575, 0.3820, 0.8672, 0.3558, 0.4400, 0.3587, 0.6304, 0.5472],
  1008. [0.4623, 0.2721, 0.7361, 0.2071, 0.3519, 0.1966, 0.6082, 0.5153],
  1009. [0.5630, 0.3509, 0.8418, 0.3172, 0.3521, 0.2707, 0.6689, 0.4976],
  1010. [0.5934, 0.3768, 0.8642, 0.3091, 0.4111, 0.3737, 0.6373, 0.5525],
  1011. [0.6304, 0.3884, 0.8941, 0.4007, 0.3824, 0.3954, 0.6342, 0.5111],
  1012. [0.5640, 0.3495, 0.8785, 0.3389, 0.4095, 0.2641, 0.7161, 0.5165],
  1013. [0.6403, 0.4074, 0.9030, 0.3953, 0.4031, 0.4607, 0.6779, 0.5264]],
  1014. device='cuda:0', grad_fn=<AddmmBackward>)
  1015. landmarks are: tensor([[[0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  1016. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  1017. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  1018. [0.6111, 0.4033, 0.8300, 0.3267, 0.3587, 0.3333, 0.5444, 0.5637],
  1019. [0.6250, 0.4105, 0.8700, 0.3717, 0.3587, 0.4967, 0.6037, 0.5167],
  1020. [0.6293, 0.3982, 0.8700, 0.5300, 0.3762, 0.4717, 0.7050, 0.5297],
  1021. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6187, 0.5283],
  1022. [0.6250, 0.4103, 0.8950, 0.4400, 0.3913, 0.5650, 0.6050, 0.5133]]],
  1023. device='cuda:0')
  1024. loss_train_step before backward: tensor(0.0046, device='cuda:0', grad_fn=<MseLossBackward>)
  1025. loss_train_step after backward: tensor(0.0046, device='cuda:0', grad_fn=<MseLossBackward>)
  1026. loss_train: 0.6799087165854871
  1027. step: 30
  1028. running loss: 0.022663623886182906
  1029. Train Steps: 30/90 Loss: 0.0227 torch.Size([8, 600, 800])
  1030. torch.Size([8, 8])
  1031. tensor([[0.6223, 0.4028, 0.8988, 0.4200, 0.3763, 0.5733, 0.6375, 0.5167],
  1032. [ nan, nan, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  1033. [0.6240, 0.4217, 0.8150, 0.3133, 0.4425, 0.2650, 0.5650, 0.5817],
  1034. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783],
  1035. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  1036. [0.6264, 0.4071, 0.9038, 0.3867, 0.3663, 0.3917, 0.6338, 0.5283],
  1037. [0.6188, 0.4099, 0.7400, 0.2433, 0.3962, 0.2750, 0.6162, 0.5467],
  1038. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398]],
  1039. device='cuda:0', dtype=torch.float64)
  1040. predictions are: tensor([[0.8508, 0.5153, 0.9209, 0.4967, 0.3942, 0.5981, 0.6709, 0.5179],
  1041. [0.4904, 0.3145, 0.7087, 0.2052, 0.3900, 0.1955, 0.5960, 0.5065],
  1042. [0.4885, 0.3282, 0.7663, 0.2554, 0.4071, 0.1947, 0.6204, 0.5186],
  1043. [0.6714, 0.4204, 0.7941, 0.3738, 0.3507, 0.3638, 0.6271, 0.5192],
  1044. [0.5495, 0.3321, 0.7264, 0.2246, 0.4070, 0.2041, 0.6100, 0.5103],
  1045. [0.6896, 0.4610, 0.8982, 0.4124, 0.3639, 0.4241, 0.6851, 0.5136],
  1046. [0.5575, 0.3861, 0.7116, 0.2491, 0.3762, 0.2665, 0.6054, 0.5234],
  1047. [0.7678, 0.4905, 0.9072, 0.5293, 0.3453, 0.4882, 0.6262, 0.4931]],
  1048. device='cuda:0', grad_fn=<AddmmBackward>)
  1049. landmarks are: tensor([[[0.6223, 0.4028, 0.8988, 0.4200, 0.3762, 0.5733, 0.6375, 0.5167],
  1050. [0.0000, 0.0000, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  1051. [0.6240, 0.4217, 0.8150, 0.3133, 0.4425, 0.2650, 0.5650, 0.5817],
  1052. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783],
  1053. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  1054. [0.6264, 0.4071, 0.9038, 0.3867, 0.3663, 0.3917, 0.6338, 0.5283],
  1055. [0.6188, 0.4099, 0.7400, 0.2433, 0.3963, 0.2750, 0.6162, 0.5467],
  1056. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398]]],
  1057. device='cuda:0')
  1058. loss_train_step before backward: tensor(0.0089, device='cuda:0', grad_fn=<MseLossBackward>)
  1059. loss_train_step after backward: tensor(0.0089, device='cuda:0', grad_fn=<MseLossBackward>)
  1060. loss_train: 0.6888156817294657
  1061. step: 31
  1062. running loss: 0.022219860700950507
  1063. Train Steps: 31/90 Loss: 0.0222 torch.Size([8, 600, 800])
  1064. torch.Size([8, 8])
  1065. tensor([[0.6250, 0.4146, 0.8838, 0.3933, 0.3588, 0.4283, 0.6162, 0.5367],
  1066. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  1067. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  1068. [0.6183, 0.4076, 0.8838, 0.4517, 0.3813, 0.4483, 0.5775, 0.5633],
  1069. [0.6097, 0.3988, 0.8650, 0.5250, 0.4213, 0.5200, 0.5675, 0.5050],
  1070. [0.6104, 0.4029, 0.8738, 0.4900, 0.4088, 0.4533, 0.5070, 0.5510],
  1071. [0.6157, 0.4102, 0.8513, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  1072. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583]],
  1073. device='cuda:0', dtype=torch.float64)
  1074. predictions are: tensor([[0.6710, 0.4203, 0.7945, 0.3961, 0.3727, 0.3562, 0.6165, 0.4802],
  1075. [0.5841, 0.3361, 0.7245, 0.2290, 0.3879, 0.2388, 0.5897, 0.5041],
  1076. [0.6982, 0.4584, 0.7993, 0.3879, 0.3324, 0.3598, 0.5931, 0.5174],
  1077. [0.7280, 0.4381, 0.8203, 0.4481, 0.3542, 0.4475, 0.6107, 0.5190],
  1078. [0.7076, 0.4717, 0.8278, 0.4338, 0.3845, 0.4404, 0.5895, 0.5367],
  1079. [0.7239, 0.4483, 0.8322, 0.4365, 0.3643, 0.4370, 0.5907, 0.5236],
  1080. [0.6941, 0.4412, 0.7852, 0.3663, 0.3613, 0.3664, 0.6066, 0.5027],
  1081. [0.7049, 0.4463, 0.7884, 0.3625, 0.3525, 0.3731, 0.6007, 0.5497]],
  1082. device='cuda:0', grad_fn=<AddmmBackward>)
  1083. landmarks are: tensor([[[0.6250, 0.4146, 0.8838, 0.3933, 0.3587, 0.4283, 0.6162, 0.5367],
  1084. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  1085. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  1086. [0.6183, 0.4076, 0.8838, 0.4517, 0.3812, 0.4483, 0.5775, 0.5633],
  1087. [0.6097, 0.3988, 0.8650, 0.5250, 0.4212, 0.5200, 0.5675, 0.5050],
  1088. [0.6104, 0.4029, 0.8737, 0.4900, 0.4087, 0.4533, 0.5070, 0.5510],
  1089. [0.6157, 0.4102, 0.8512, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  1090. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583]]],
  1091. device='cuda:0')
  1092. loss_train_step before backward: tensor(0.0029, device='cuda:0', grad_fn=<MseLossBackward>)
  1093. loss_train_step after backward: tensor(0.0029, device='cuda:0', grad_fn=<MseLossBackward>)
  1094. loss_train: 0.6917505832388997
  1095. step: 32
  1096. running loss: 0.021617205726215616
  1097.  
  1098. Train Steps: 32/90 Loss: 0.0216 torch.Size([8, 600, 800])
  1099. torch.Size([8, 8])
  1100. tensor([[0.6246, 0.4028, 0.8738, 0.4867, 0.4088, 0.5667, 0.6362, 0.5200],
  1101. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  1102. [0.6289, 0.4024, 0.9088, 0.4567, 0.3937, 0.5633, 0.7058, 0.5609],
  1103. [0.6156, 0.4125, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  1104. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  1105. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  1106. [0.6076, 0.3953, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954],
  1107. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051]],
  1108. device='cuda:0', dtype=torch.float64)
  1109. predictions are: tensor([[0.7848, 0.4827, 0.8367, 0.4659, 0.3955, 0.5450, 0.6055, 0.5506],
  1110. [0.6511, 0.3996, 0.7105, 0.3123, 0.3613, 0.2571, 0.5342, 0.5192],
  1111. [0.7224, 0.4844, 0.8538, 0.4692, 0.3898, 0.5190, 0.5744, 0.5551],
  1112. [0.7083, 0.4858, 0.8719, 0.4633, 0.4123, 0.4938, 0.5730, 0.5529],
  1113. [0.6830, 0.4370, 0.7286, 0.3365, 0.3475, 0.3186, 0.5642, 0.5397],
  1114. [0.6332, 0.4395, 0.7310, 0.3280, 0.3316, 0.2940, 0.5389, 0.5436],
  1115. [0.6753, 0.4355, 0.7696, 0.3689, 0.3314, 0.3119, 0.5551, 0.5516],
  1116. [0.7349, 0.4735, 0.8271, 0.4398, 0.3587, 0.3736, 0.5524, 0.5296]],
  1117. device='cuda:0', grad_fn=<AddmmBackward>)
  1118. landmarks are: tensor([[[0.6246, 0.4028, 0.8737, 0.4867, 0.4087, 0.5667, 0.6363, 0.5200],
  1119. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  1120. [0.6289, 0.4024, 0.9087, 0.4567, 0.3938, 0.5633, 0.7058, 0.5609],
  1121. [0.6155, 0.4124, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  1122. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  1123. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  1124. [0.6076, 0.3952, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954],
  1125. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051]]],
  1126. device='cuda:0')
  1127. loss_train_step before backward: tensor(0.0031, device='cuda:0', grad_fn=<MseLossBackward>)
  1128. loss_train_step after backward: tensor(0.0031, device='cuda:0', grad_fn=<MseLossBackward>)
  1129. loss_train: 0.694860020885244
  1130. step: 33
  1131. running loss: 0.021056364269249818
  1132. Train Steps: 33/90 Loss: 0.0211 torch.Size([8, 600, 800])
  1133. torch.Size([8, 8])
  1134. tensor([[0.6175, 0.4091, 0.7863, 0.2800, 0.3638, 0.3583, 0.6188, 0.5433],
  1135. [0.6299, 0.4008, 0.8450, 0.5350, 0.4213, 0.5000, 0.6350, 0.5100],
  1136. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  1137. [0.6199, 0.4060, 0.8888, 0.4667, 0.3800, 0.5050, 0.6188, 0.5433],
  1138. [0.6229, 0.4198, 0.7662, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783],
  1139. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  1140. [0.6275, 0.4111, 0.8463, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  1141. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117]],
  1142. device='cuda:0', dtype=torch.float64)
  1143. predictions are: tensor([[0.6149, 0.4340, 0.7060, 0.3099, 0.3430, 0.3445, 0.5397, 0.5623],
  1144. [0.7183, 0.4977, 0.8231, 0.5166, 0.3665, 0.5264, 0.5438, 0.5380],
  1145. [0.7315, 0.4994, 0.8343, 0.4605, 0.3698, 0.5051, 0.5327, 0.5626],
  1146. [0.7524, 0.4987, 0.8748, 0.5025, 0.3540, 0.5581, 0.5317, 0.5548],
  1147. [0.5284, 0.3690, 0.6747, 0.3188, 0.3901, 0.2372, 0.5341, 0.5517],
  1148. [0.6771, 0.4421, 0.7946, 0.4589, 0.3322, 0.4516, 0.5312, 0.5619],
  1149. [0.6197, 0.4022, 0.7689, 0.3958, 0.3993, 0.2837, 0.6107, 0.5413],
  1150. [0.7074, 0.4393, 0.8183, 0.4463, 0.3742, 0.3971, 0.5818, 0.5478]],
  1151. device='cuda:0', grad_fn=<AddmmBackward>)
  1152. landmarks are: tensor([[[0.6175, 0.4091, 0.7862, 0.2800, 0.3638, 0.3583, 0.6187, 0.5433],
  1153. [0.6299, 0.4008, 0.8450, 0.5350, 0.4212, 0.5000, 0.6350, 0.5100],
  1154. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  1155. [0.6199, 0.4060, 0.8888, 0.4667, 0.3800, 0.5050, 0.6187, 0.5433],
  1156. [0.6229, 0.4198, 0.7663, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783],
  1157. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  1158. [0.6275, 0.4111, 0.8462, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  1159. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117]]],
  1160. device='cuda:0')
  1161. loss_train_step before backward: tensor(0.0036, device='cuda:0', grad_fn=<MseLossBackward>)
  1162. loss_train_step after backward: tensor(0.0036, device='cuda:0', grad_fn=<MseLossBackward>)
  1163. loss_train: 0.6984764796216041
  1164. step: 34
  1165. running loss: 0.02054342587122365
  1166. Train Steps: 34/90 Loss: 0.0205 torch.Size([8, 600, 800])
  1167. torch.Size([8, 8])
  1168. tensor([[0.6055, 0.4015, 0.7425, 0.2033, 0.4113, 0.1883, 0.5217, 0.4823],
  1169. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  1170. [0.6151, 0.4058, 0.7068, 0.2680, 0.3400, 0.4083, 0.5775, 0.5733],
  1171. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  1172. [ nan, nan, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  1173. [0.6218, 0.4185, 0.7338, 0.2650, 0.4625, 0.1950, 0.5687, 0.5800],
  1174. [0.6321, 0.4048, 0.8738, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  1175. [0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367]],
  1176. device='cuda:0', dtype=torch.float64)
  1177. predictions are: tensor([[0.5280, 0.3390, 0.7156, 0.3190, 0.3913, 0.2567, 0.5322, 0.5371],
  1178. [0.7018, 0.5026, 0.8817, 0.5279, 0.3719, 0.5954, 0.5577, 0.5688],
  1179. [0.6409, 0.4319, 0.7399, 0.3588, 0.3451, 0.4342, 0.5118, 0.5571],
  1180. [0.6910, 0.4941, 0.8952, 0.5139, 0.4234, 0.5669, 0.5235, 0.5712],
  1181. [0.5076, 0.3336, 0.7361, 0.3109, 0.3979, 0.2886, 0.5175, 0.5497],
  1182. [0.5446, 0.3609, 0.7134, 0.3174, 0.3978, 0.2716, 0.5157, 0.5486],
  1183. [0.7687, 0.5217, 0.9188, 0.5964, 0.3390, 0.5564, 0.5295, 0.5449],
  1184. [0.6559, 0.4589, 0.8268, 0.4680, 0.3728, 0.3424, 0.5794, 0.5420]],
  1185. device='cuda:0', grad_fn=<AddmmBackward>)
  1186. landmarks are: tensor([[[0.6055, 0.4015, 0.7425, 0.2033, 0.4112, 0.1883, 0.5217, 0.4823],
  1187. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  1188. [0.6151, 0.4058, 0.7068, 0.2680, 0.3400, 0.4083, 0.5775, 0.5733],
  1189. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  1190. [0.0000, 0.0000, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  1191. [0.6218, 0.4185, 0.7337, 0.2650, 0.4625, 0.1950, 0.5688, 0.5800],
  1192. [0.6321, 0.4048, 0.8737, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  1193. [0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367]]],
  1194. device='cuda:0')
  1195. loss_train_step before backward: tensor(0.0101, device='cuda:0', grad_fn=<MseLossBackward>)
  1196. loss_train_step after backward: tensor(0.0101, device='cuda:0', grad_fn=<MseLossBackward>)
  1197. loss_train: 0.7085485982242972
  1198. step: 35
  1199. running loss: 0.020244245663551347
  1200. Train Steps: 35/90 Loss: 0.0202 torch.Size([8, 600, 800])
  1201. torch.Size([8, 8])
  1202. tensor([[0.6236, 0.3967, 0.8675, 0.5400, 0.3862, 0.4517, 0.5825, 0.5200],
  1203. [0.6068, 0.3963, 0.8650, 0.4317, 0.4037, 0.5083, 0.5253, 0.4999],
  1204. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  1205. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  1206. [0.6357, 0.4097, 0.9038, 0.3883, 0.4213, 0.2950, 0.6686, 0.5390],
  1207. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  1208. [0.6193, 0.4079, 0.7288, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  1209. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690]],
  1210. device='cuda:0', dtype=torch.float64)
  1211. predictions are: tensor([[0.6796, 0.4346, 0.9105, 0.5463, 0.3525, 0.5400, 0.5134, 0.5412],
  1212. [0.6602, 0.4432, 0.9022, 0.4741, 0.3960, 0.5250, 0.5403, 0.5289],
  1213. [0.6180, 0.4283, 0.8339, 0.4508, 0.4131, 0.3817, 0.5499, 0.5397],
  1214. [0.5023, 0.3241, 0.7431, 0.3296, 0.4211, 0.2222, 0.5520, 0.5293],
  1215. [0.5949, 0.3923, 0.8258, 0.4020, 0.4123, 0.3946, 0.5833, 0.5566],
  1216. [0.6230, 0.4350, 0.8654, 0.3879, 0.3692, 0.5040, 0.5558, 0.5466],
  1217. [0.5663, 0.3615, 0.7780, 0.3761, 0.4226, 0.2939, 0.5498, 0.5536],
  1218. [0.6266, 0.4297, 0.8683, 0.4994, 0.3981, 0.5438, 0.5580, 0.5618]],
  1219. device='cuda:0', grad_fn=<AddmmBackward>)
  1220. landmarks are: tensor([[[0.6236, 0.3967, 0.8675, 0.5400, 0.3862, 0.4517, 0.5825, 0.5200],
  1221. [0.6068, 0.3963, 0.8650, 0.4317, 0.4038, 0.5083, 0.5253, 0.4999],
  1222. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  1223. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  1224. [0.6357, 0.4097, 0.9038, 0.3883, 0.4212, 0.2950, 0.6686, 0.5390],
  1225. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  1226. [0.6193, 0.4078, 0.7287, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  1227. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690]]],
  1228. device='cuda:0')
  1229. loss_train_step before backward: tensor(0.0033, device='cuda:0', grad_fn=<MseLossBackward>)
  1230. loss_train_step after backward: tensor(0.0033, device='cuda:0', grad_fn=<MseLossBackward>)
  1231. loss_train: 0.7118957417551428
  1232. step: 36
  1233. running loss: 0.019774881715420634
  1234.  
  1235. Train Steps: 36/90 Loss: 0.0198 torch.Size([8, 600, 800])
  1236. torch.Size([8, 8])
  1237. tensor([[0.6175, 0.4091, 0.7863, 0.2800, 0.3638, 0.3583, 0.6188, 0.5433],
  1238. [0.6308, 0.3990, 0.8688, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  1239. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  1240. [0.6250, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6088, 0.5183],
  1241. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  1242. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  1243. [0.6353, 0.4128, 0.9138, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  1244. [0.6212, 0.4033, 0.8938, 0.4167, 0.3813, 0.4267, 0.5613, 0.5583]],
  1245. device='cuda:0', dtype=torch.float64)
  1246. predictions are: tensor([[0.5292, 0.3872, 0.7569, 0.2700, 0.3864, 0.3128, 0.5665, 0.5585],
  1247. [0.6096, 0.4050, 0.8747, 0.4772, 0.4006, 0.4952, 0.5530, 0.5093],
  1248. [0.5875, 0.3963, 0.8670, 0.4399, 0.4564, 0.4967, 0.5858, 0.5465],
  1249. [0.5625, 0.3597, 0.8807, 0.4257, 0.4473, 0.4756, 0.5719, 0.5578],
  1250. [0.5999, 0.4131, 0.8598, 0.4714, 0.3961, 0.4890, 0.5484, 0.5499],
  1251. [0.5851, 0.3990, 0.8530, 0.4214, 0.3858, 0.4066, 0.5353, 0.5361],
  1252. [0.5423, 0.3711, 0.8811, 0.3812, 0.4638, 0.3094, 0.6062, 0.5717],
  1253. [0.5934, 0.3772, 0.9045, 0.4281, 0.3759, 0.4242, 0.5442, 0.5451]],
  1254. device='cuda:0', grad_fn=<AddmmBackward>)
  1255. landmarks are: tensor([[[0.6175, 0.4091, 0.7862, 0.2800, 0.3638, 0.3583, 0.6187, 0.5433],
  1256. [0.6308, 0.3990, 0.8687, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  1257. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  1258. [0.6251, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6087, 0.5183],
  1259. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  1260. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  1261. [0.6353, 0.4128, 0.9137, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  1262. [0.6212, 0.4033, 0.8938, 0.4167, 0.3812, 0.4267, 0.5612, 0.5583]]],
  1263. device='cuda:0')
  1264. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  1265. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  1266. loss_train: 0.7136210562894121
  1267. step: 37
  1268. running loss: 0.019287055575389515
  1269. Train Steps: 37/90 Loss: 0.0193 torch.Size([8, 600, 800])
  1270. torch.Size([8, 8])
  1271. tensor([[0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  1272. [0.6226, 0.4098, 0.8912, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  1273. [0.6311, 0.4008, 0.7935, 0.5746, 0.3900, 0.5033, 0.6955, 0.5366],
  1274. [0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  1275. [0.6038, 0.3946, 0.8413, 0.4883, 0.3563, 0.4550, 0.5266, 0.4693],
  1276. [0.6175, 0.4091, 0.7863, 0.2800, 0.3638, 0.3583, 0.6188, 0.5433],
  1277. [0.6275, 0.4050, 0.9038, 0.3767, 0.3838, 0.3533, 0.7074, 0.5575],
  1278. [0.6221, 0.4107, 0.7788, 0.3033, 0.3950, 0.2817, 0.6075, 0.5517]],
  1279. device='cuda:0', dtype=torch.float64)
  1280. predictions are: tensor([[0.5222, 0.3515, 0.8506, 0.3014, 0.4592, 0.3052, 0.6338, 0.5594],
  1281. [0.5480, 0.3766, 0.8690, 0.4320, 0.4403, 0.4208, 0.6058, 0.5543],
  1282. [0.5488, 0.3863, 0.8739, 0.4569, 0.4179, 0.4869, 0.5889, 0.5529],
  1283. [0.5554, 0.3790, 0.8614, 0.3570, 0.4100, 0.4137, 0.6209, 0.5165],
  1284. [0.5769, 0.3792, 0.8886, 0.4361, 0.3942, 0.4907, 0.5784, 0.5046],
  1285. [0.5248, 0.3738, 0.7878, 0.2725, 0.3957, 0.3849, 0.6102, 0.5456],
  1286. [0.5949, 0.3894, 0.9011, 0.4044, 0.4239, 0.3835, 0.6411, 0.5413],
  1287. [0.5363, 0.3850, 0.8409, 0.3620, 0.4490, 0.2931, 0.6067, 0.5606]],
  1288. device='cuda:0', grad_fn=<AddmmBackward>)
  1289. landmarks are: tensor([[[0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  1290. [0.6226, 0.4098, 0.8913, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  1291. [0.6311, 0.4008, 0.7935, 0.5746, 0.3900, 0.5033, 0.6955, 0.5366],
  1292. [0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  1293. [0.6038, 0.3946, 0.8413, 0.4883, 0.3562, 0.4550, 0.5266, 0.4693],
  1294. [0.6175, 0.4091, 0.7862, 0.2800, 0.3638, 0.3583, 0.6187, 0.5433],
  1295. [0.6275, 0.4050, 0.9038, 0.3767, 0.3837, 0.3533, 0.7074, 0.5575],
  1296. [0.6221, 0.4107, 0.7788, 0.3033, 0.3950, 0.2817, 0.6075, 0.5517]]],
  1297. device='cuda:0')
  1298. loss_train_step before backward: tensor(0.0029, device='cuda:0', grad_fn=<MseLossBackward>)
  1299. loss_train_step after backward: tensor(0.0029, device='cuda:0', grad_fn=<MseLossBackward>)
  1300. loss_train: 0.7165557543048635
  1301. step: 38
  1302. running loss: 0.018856730376443778
  1303. Train Steps: 38/90 Loss: 0.0189 torch.Size([8, 600, 800])
  1304. torch.Size([8, 8])
  1305. tensor([[0.6215, 0.4119, 0.7688, 0.2300, 0.4200, 0.2283, 0.5925, 0.5317],
  1306. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  1307. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092],
  1308. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  1309. [0.6082, 0.4024, 0.8738, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  1310. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896],
  1311. [0.6125, 0.3974, 0.7725, 0.2517, 0.3538, 0.3317, 0.5887, 0.5500],
  1312. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373]],
  1313. device='cuda:0', dtype=torch.float64)
  1314. predictions are: tensor([[0.4853, 0.3183, 0.7892, 0.2729, 0.4579, 0.1858, 0.6309, 0.5517],
  1315. [0.5639, 0.3632, 0.9041, 0.3791, 0.4606, 0.3558, 0.6677, 0.5591],
  1316. [0.5767, 0.3924, 0.9650, 0.4746, 0.4343, 0.5645, 0.6595, 0.5016],
  1317. [0.5177, 0.3271, 0.7849, 0.2934, 0.3936, 0.3067, 0.6365, 0.5375],
  1318. [0.5677, 0.3806, 0.8894, 0.3982, 0.4131, 0.4094, 0.6496, 0.5093],
  1319. [0.5819, 0.3677, 0.8950, 0.3920, 0.4091, 0.3712, 0.6272, 0.5226],
  1320. [0.5597, 0.3846, 0.8170, 0.2857, 0.4165, 0.3674, 0.6354, 0.5120],
  1321. [0.5893, 0.3831, 0.9125, 0.4968, 0.3887, 0.5205, 0.6712, 0.5350]],
  1322. device='cuda:0', grad_fn=<AddmmBackward>)
  1323. landmarks are: tensor([[[0.6215, 0.4119, 0.7688, 0.2300, 0.4200, 0.2283, 0.5925, 0.5317],
  1324. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  1325. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092],
  1326. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  1327. [0.6082, 0.4024, 0.8737, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  1328. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896],
  1329. [0.6125, 0.3974, 0.7725, 0.2517, 0.3537, 0.3317, 0.5888, 0.5500],
  1330. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373]]],
  1331. device='cuda:0')
  1332. loss_train_step before backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  1333. loss_train_step after backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  1334. loss_train: 0.7197442847536877
  1335. step: 39
  1336. running loss: 0.018454981660350967
  1337. Train Steps: 39/90 Loss: 0.0185 torch.Size([8, 600, 800])
  1338. torch.Size([8, 8])
  1339. tensor([[0.6200, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  1340. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  1341. [0.6203, 0.4096, 0.8862, 0.4267, 0.3538, 0.4117, 0.6025, 0.5650],
  1342. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  1343. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  1344. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309],
  1345. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  1346. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882]],
  1347. device='cuda:0', dtype=torch.float64)
  1348. predictions are: tensor([[0.5669, 0.3682, 0.8350, 0.2945, 0.4606, 0.2369, 0.6672, 0.5693],
  1349. [0.5937, 0.3768, 0.8715, 0.4076, 0.3744, 0.4059, 0.6638, 0.4973],
  1350. [0.5738, 0.3461, 0.9497, 0.3841, 0.3975, 0.4007, 0.6788, 0.5481],
  1351. [0.5936, 0.3624, 0.8194, 0.3669, 0.3962, 0.3466, 0.6540, 0.5133],
  1352. [0.5639, 0.3757, 0.8873, 0.4276, 0.4282, 0.4881, 0.6677, 0.5097],
  1353. [0.6198, 0.4045, 0.8702, 0.4273, 0.4185, 0.3589, 0.6628, 0.5233],
  1354. [0.5214, 0.3388, 0.7650, 0.2842, 0.4229, 0.2296, 0.6664, 0.5563],
  1355. [0.6124, 0.3885, 0.8598, 0.3752, 0.3879, 0.3847, 0.6683, 0.5288]],
  1356. device='cuda:0', grad_fn=<AddmmBackward>)
  1357. landmarks are: tensor([[[0.6199, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  1358. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  1359. [0.6203, 0.4096, 0.8863, 0.4267, 0.3537, 0.4117, 0.6025, 0.5650],
  1360. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  1361. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  1362. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309],
  1363. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  1364. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882]]],
  1365. device='cuda:0')
  1366. loss_train_step before backward: tensor(0.0029, device='cuda:0', grad_fn=<MseLossBackward>)
  1367. loss_train_step after backward: tensor(0.0029, device='cuda:0', grad_fn=<MseLossBackward>)
  1368. loss_train: 0.7226116416277364
  1369. step: 40
  1370. running loss: 0.01806529104069341
  1371.  
  1372. Train Steps: 40/90 Loss: 0.0181 torch.Size([8, 600, 800])
  1373. torch.Size([8, 8])
  1374. tensor([[0.6207, 0.4110, 0.8738, 0.5000, 0.4800, 0.5633, 0.6300, 0.5433],
  1375. [0.6239, 0.4061, 0.8850, 0.4600, 0.4225, 0.5200, 0.6138, 0.5450],
  1376. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783],
  1377. [0.6122, 0.3993, 0.8738, 0.4667, 0.4517, 0.4879, 0.5155, 0.4927],
  1378. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  1379. [0.6200, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913],
  1380. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  1381. [0.6264, 0.3972, 0.8853, 0.4771, 0.3853, 0.4511, 0.6293, 0.5334]],
  1382. device='cuda:0', dtype=torch.float64)
  1383. predictions are: tensor([[0.6079, 0.3684, 0.8767, 0.4069, 0.4390, 0.4066, 0.6747, 0.5424],
  1384. [0.5971, 0.3870, 0.9012, 0.4236, 0.3968, 0.4451, 0.6842, 0.5146],
  1385. [0.6840, 0.4018, 0.8202, 0.3566, 0.3806, 0.2903, 0.6120, 0.5582],
  1386. [0.6218, 0.3943, 0.8632, 0.3988, 0.4154, 0.3771, 0.6226, 0.5125],
  1387. [0.6515, 0.3961, 0.8786, 0.3728, 0.3809, 0.2942, 0.6653, 0.5429],
  1388. [0.6099, 0.3804, 0.8950, 0.4180, 0.4207, 0.4285, 0.6519, 0.5324],
  1389. [0.5717, 0.3496, 0.7514, 0.2635, 0.4039, 0.1859, 0.6216, 0.5170],
  1390. [0.5850, 0.3603, 0.8961, 0.3972, 0.3829, 0.3713, 0.6643, 0.5443]],
  1391. device='cuda:0', grad_fn=<AddmmBackward>)
  1392. landmarks are: tensor([[[0.6207, 0.4110, 0.8737, 0.5000, 0.4800, 0.5633, 0.6300, 0.5433],
  1393. [0.6239, 0.4061, 0.8850, 0.4600, 0.4225, 0.5200, 0.6137, 0.5450],
  1394. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783],
  1395. [0.6122, 0.3993, 0.8737, 0.4667, 0.4517, 0.4879, 0.5155, 0.4927],
  1396. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  1397. [0.6199, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913],
  1398. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  1399. [0.6264, 0.3972, 0.8853, 0.4771, 0.3853, 0.4511, 0.6293, 0.5334]]],
  1400. device='cuda:0')
  1401. loss_train_step before backward: tensor(0.0025, device='cuda:0', grad_fn=<MseLossBackward>)
  1402. loss_train_step after backward: tensor(0.0025, device='cuda:0', grad_fn=<MseLossBackward>)
  1403. loss_train: 0.7251194597920403
  1404. step: 41
  1405. running loss: 0.01768584048273269
  1406. Train Steps: 41/90 Loss: 0.0177 torch.Size([8, 600, 800])
  1407. torch.Size([8, 8])
  1408. tensor([[0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  1409. [0.6336, 0.4191, 0.8938, 0.5167, 0.3937, 0.3517, 0.7343, 0.5748],
  1410. [0.6268, 0.4061, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  1411. [0.6200, 0.4049, 0.8638, 0.5617, 0.4125, 0.5100, 0.6013, 0.5317],
  1412. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  1413. [0.6058, 0.3986, 0.8324, 0.4626, 0.3838, 0.4983, 0.5147, 0.5466],
  1414. [0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431],
  1415. [0.6203, 0.4021, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405]],
  1416. device='cuda:0', dtype=torch.float64)
  1417. predictions are: tensor([[0.6478, 0.3852, 0.8580, 0.4518, 0.3584, 0.3141, 0.5976, 0.5479],
  1418. [0.6389, 0.3808, 0.8447, 0.3671, 0.3886, 0.2617, 0.6178, 0.5360],
  1419. [0.6121, 0.3714, 0.8291, 0.2746, 0.4025, 0.2104, 0.6777, 0.5752],
  1420. [0.6169, 0.3722, 0.8361, 0.4941, 0.3821, 0.4507, 0.6206, 0.5253],
  1421. [0.6501, 0.3987, 0.8182, 0.3100, 0.4028, 0.2576, 0.6209, 0.5517],
  1422. [0.5819, 0.3600, 0.8697, 0.4064, 0.3736, 0.3797, 0.6058, 0.5259],
  1423. [0.6105, 0.3602, 0.8543, 0.4191, 0.4216, 0.3745, 0.6147, 0.5200],
  1424. [0.6841, 0.4008, 0.8740, 0.4576, 0.3594, 0.3333, 0.6435, 0.5299]],
  1425. device='cuda:0', grad_fn=<AddmmBackward>)
  1426. landmarks are: tensor([[[0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  1427. [0.6336, 0.4191, 0.8938, 0.5167, 0.3938, 0.3517, 0.7343, 0.5748],
  1428. [0.6268, 0.4060, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  1429. [0.6199, 0.4049, 0.8637, 0.5617, 0.4125, 0.5100, 0.6012, 0.5317],
  1430. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  1431. [0.6058, 0.3986, 0.8324, 0.4626, 0.3837, 0.4983, 0.5147, 0.5466],
  1432. [0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431],
  1433. [0.6203, 0.4020, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405]]],
  1434. device='cuda:0')
  1435. loss_train_step before backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  1436. loss_train_step after backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  1437. loss_train: 0.7277624051785097
  1438. step: 42
  1439. running loss: 0.017327676313774038
  1440. Train Steps: 42/90 Loss: 0.0173 torch.Size([8, 600, 800])
  1441. torch.Size([8, 8])
  1442. tensor([[0.6214, 0.4112, 0.7838, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  1443. [0.6125, 0.4076, 0.8488, 0.3883, 0.3700, 0.3683, 0.5026, 0.5505],
  1444. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667],
  1445. [0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5837, 0.5317],
  1446. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  1447. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  1448. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  1449. [0.6275, 0.4157, 0.8337, 0.5800, 0.3763, 0.4200, 0.5547, 0.6125]],
  1450. device='cuda:0', dtype=torch.float64)
  1451. predictions are: tensor([[0.6859, 0.4183, 0.8201, 0.3667, 0.3846, 0.2885, 0.6160, 0.5481],
  1452. [0.6493, 0.3851, 0.8564, 0.4168, 0.3520, 0.3721, 0.5779, 0.5315],
  1453. [0.6808, 0.4184, 0.8806, 0.4314, 0.3619, 0.3917, 0.6199, 0.5227],
  1454. [0.6688, 0.3939, 0.8313, 0.3791, 0.3818, 0.2768, 0.5978, 0.5413],
  1455. [0.6302, 0.3876, 0.8173, 0.2933, 0.4438, 0.1975, 0.6074, 0.5589],
  1456. [0.6045, 0.3703, 0.7571, 0.3427, 0.3783, 0.2547, 0.5605, 0.5320],
  1457. [0.6031, 0.4094, 0.8788, 0.5251, 0.4059, 0.4233, 0.5849, 0.5355],
  1458. [0.6389, 0.4025, 0.8686, 0.5786, 0.3450, 0.4725, 0.6304, 0.4850]],
  1459. device='cuda:0', grad_fn=<AddmmBackward>)
  1460. landmarks are: tensor([[[0.6214, 0.4112, 0.7837, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  1461. [0.6125, 0.4076, 0.8487, 0.3883, 0.3700, 0.3683, 0.5026, 0.5505],
  1462. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667],
  1463. [0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5838, 0.5317],
  1464. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  1465. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  1466. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  1467. [0.6275, 0.4157, 0.8338, 0.5800, 0.3762, 0.4200, 0.5547, 0.6125]]],
  1468. device='cuda:0')
  1469. loss_train_step before backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  1470. loss_train_step after backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  1471. loss_train: 0.7299826672533527
  1472. step: 43
  1473. running loss: 0.01697634109891518
  1474. Train Steps: 43/90 Loss: 0.0170 torch.Size([8, 600, 800])
  1475. torch.Size([8, 8])
  1476. tensor([[0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  1477. [0.6250, 0.4103, 0.8950, 0.4400, 0.3912, 0.5650, 0.6050, 0.5133],
  1478. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  1479. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  1480. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  1481. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  1482. [0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510],
  1483. [0.6200, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913]],
  1484. device='cuda:0', dtype=torch.float64)
  1485. predictions are: tensor([[0.6318, 0.4200, 0.8345, 0.3732, 0.4057, 0.3183, 0.5886, 0.5354],
  1486. [0.6617, 0.4213, 0.8959, 0.4752, 0.4035, 0.5062, 0.5819, 0.5257],
  1487. [0.6774, 0.3981, 0.8371, 0.3690, 0.3556, 0.3575, 0.5416, 0.5478],
  1488. [0.6886, 0.4337, 0.8427, 0.3384, 0.4122, 0.2879, 0.5650, 0.5443],
  1489. [0.6776, 0.4242, 0.8516, 0.5299, 0.3947, 0.4612, 0.6092, 0.5451],
  1490. [0.7017, 0.4519, 0.8956, 0.5708, 0.3654, 0.4416, 0.5562, 0.5274],
  1491. [0.6478, 0.4055, 0.7656, 0.3337, 0.3684, 0.2613, 0.5162, 0.5638],
  1492. [0.6874, 0.4252, 0.9035, 0.5301, 0.4162, 0.4708, 0.5566, 0.5378]],
  1493. device='cuda:0', grad_fn=<AddmmBackward>)
  1494. landmarks are: tensor([[[0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  1495. [0.6250, 0.4103, 0.8950, 0.4400, 0.3913, 0.5650, 0.6050, 0.5133],
  1496. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  1497. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  1498. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  1499. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  1500. [0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510],
  1501. [0.6199, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913]]],
  1502. device='cuda:0')
  1503. loss_train_step before backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  1504. loss_train_step after backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  1505. loss_train: 0.7321775729069486
  1506. step: 44
  1507. running loss: 0.01664039938424883
  1508.  
  1509. Train Steps: 44/90 Loss: 0.0166 torch.Size([8, 600, 800])
  1510. torch.Size([8, 8])
  1511. tensor([[ nan, nan, 0.6469, 0.1943, 0.4025, 0.2000, 0.5125, 0.5533],
  1512. [0.6246, 0.4028, 0.8738, 0.4867, 0.4088, 0.5667, 0.6362, 0.5200],
  1513. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  1514. [0.6185, 0.4067, 0.8838, 0.4450, 0.4037, 0.4733, 0.5213, 0.5142],
  1515. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  1516. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5988, 0.5667],
  1517. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116],
  1518. [0.6243, 0.4128, 0.7762, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417]],
  1519. device='cuda:0', dtype=torch.float64)
  1520. predictions are: tensor([[0.5502, 0.3357, 0.6939, 0.2983, 0.3869, 0.2363, 0.4876, 0.5458],
  1521. [0.6379, 0.4245, 0.8473, 0.5687, 0.3954, 0.6164, 0.5743, 0.4970],
  1522. [0.6755, 0.4439, 0.7752, 0.2956, 0.4185, 0.2184, 0.5482, 0.5289],
  1523. [0.6546, 0.4260, 0.8546, 0.4950, 0.3709, 0.4749, 0.5239, 0.5447],
  1524. [0.6950, 0.4629, 0.8671, 0.4698, 0.3533, 0.4084, 0.5627, 0.5403],
  1525. [0.7052, 0.4755, 0.8520, 0.4392, 0.3554, 0.3897, 0.5525, 0.5459],
  1526. [0.6257, 0.4419, 0.8624, 0.4957, 0.3983, 0.5694, 0.5611, 0.5239],
  1527. [0.6736, 0.4606, 0.7992, 0.3360, 0.4000, 0.2437, 0.5425, 0.5643]],
  1528. device='cuda:0', grad_fn=<AddmmBackward>)
  1529. landmarks are: tensor([[[0.0000, 0.0000, 0.6469, 0.1943, 0.4025, 0.2000, 0.5125, 0.5533],
  1530. [0.6246, 0.4028, 0.8737, 0.4867, 0.4087, 0.5667, 0.6363, 0.5200],
  1531. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  1532. [0.6185, 0.4067, 0.8838, 0.4450, 0.4038, 0.4733, 0.5213, 0.5142],
  1533. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  1534. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5987, 0.5667],
  1535. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116],
  1536. [0.6243, 0.4128, 0.7763, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417]]],
  1537. device='cuda:0')
  1538. loss_train_step before backward: tensor(0.0085, device='cuda:0', grad_fn=<MseLossBackward>)
  1539. loss_train_step after backward: tensor(0.0085, device='cuda:0', grad_fn=<MseLossBackward>)
  1540. loss_train: 0.7407123452285305
  1541. step: 45
  1542. running loss: 0.01646027433841179
  1543. Train Steps: 45/90 Loss: 0.0165 torch.Size([8, 600, 800])
  1544. torch.Size([8, 8])
  1545. tensor([[0.6308, 0.3990, 0.8688, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  1546. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  1547. [0.6260, 0.4120, 0.8013, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895],
  1548. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  1549. [0.6199, 0.4093, 0.7913, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700],
  1550. [0.6064, 0.3953, 0.8738, 0.4417, 0.3663, 0.4683, 0.5511, 0.5416],
  1551. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6138, 0.5333],
  1552. [0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050]],
  1553. device='cuda:0', dtype=torch.float64)
  1554. predictions are: tensor([[0.6275, 0.4261, 0.8645, 0.5512, 0.3968, 0.5612, 0.5912, 0.4915],
  1555. [0.5661, 0.3730, 0.7460, 0.2466, 0.4075, 0.2641, 0.5022, 0.5740],
  1556. [0.6225, 0.4330, 0.7935, 0.2694, 0.4527, 0.2330, 0.5499, 0.5673],
  1557. [0.6655, 0.4337, 0.7820, 0.3664, 0.3560, 0.4036, 0.5483, 0.5219],
  1558. [0.6356, 0.4358, 0.8026, 0.3195, 0.4113, 0.3123, 0.5415, 0.5637],
  1559. [0.6347, 0.4366, 0.8743, 0.5053, 0.3707, 0.5440, 0.5326, 0.5076],
  1560. [0.6398, 0.4466, 0.8624, 0.5047, 0.3825, 0.5448, 0.5553, 0.5348],
  1561. [0.6460, 0.4478, 0.9060, 0.5175, 0.3901, 0.5797, 0.5691, 0.5219]],
  1562. device='cuda:0', grad_fn=<AddmmBackward>)
  1563. landmarks are: tensor([[[0.6308, 0.3990, 0.8687, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  1564. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  1565. [0.6259, 0.4120, 0.8012, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895],
  1566. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  1567. [0.6198, 0.4093, 0.7912, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700],
  1568. [0.6064, 0.3952, 0.8737, 0.4417, 0.3663, 0.4683, 0.5511, 0.5416],
  1569. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6137, 0.5333],
  1570. [0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050]]],
  1571. device='cuda:0')
  1572. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  1573. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  1574. loss_train: 0.7426065595354885
  1575. step: 46
  1576. running loss: 0.01614362085946714
  1577. Train Steps: 46/90 Loss: 0.0161 torch.Size([8, 600, 800])
  1578. torch.Size([8, 8])
  1579. tensor([[0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  1580. [0.6267, 0.4080, 0.8438, 0.2633, 0.4763, 0.1800, 0.6259, 0.5240],
  1581. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  1582. [ nan, nan, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600],
  1583. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  1584. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  1585. [0.6124, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  1586. [0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6262, 0.5167]],
  1587. device='cuda:0', dtype=torch.float64)
  1588. predictions are: tensor([[0.5601, 0.4024, 0.8080, 0.4328, 0.3650, 0.5411, 0.5782, 0.5040],
  1589. [0.6328, 0.4254, 0.8367, 0.2879, 0.4258, 0.3150, 0.5593, 0.5436],
  1590. [0.5882, 0.4032, 0.8208, 0.5737, 0.3746, 0.5384, 0.5297, 0.5168],
  1591. [0.5995, 0.4132, 0.7947, 0.2517, 0.4077, 0.3107, 0.5249, 0.5769],
  1592. [0.5679, 0.4346, 0.8613, 0.4552, 0.4261, 0.5152, 0.5542, 0.5333],
  1593. [0.5830, 0.4047, 0.7621, 0.3065, 0.3942, 0.3711, 0.5540, 0.5240],
  1594. [0.5902, 0.4203, 0.7136, 0.2819, 0.3937, 0.3589, 0.5246, 0.5339],
  1595. [0.6296, 0.4507, 0.8701, 0.4186, 0.4044, 0.4399, 0.5818, 0.5159]],
  1596. device='cuda:0', grad_fn=<AddmmBackward>)
  1597. landmarks are: tensor([[[0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  1598. [0.6267, 0.4080, 0.8438, 0.2633, 0.4762, 0.1800, 0.6259, 0.5240],
  1599. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  1600. [0.0000, 0.0000, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600],
  1601. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  1602. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  1603. [0.6123, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  1604. [0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6263, 0.5167]]],
  1605. device='cuda:0')
  1606. loss_train_step before backward: tensor(0.0098, device='cuda:0', grad_fn=<MseLossBackward>)
  1607. loss_train_step after backward: tensor(0.0098, device='cuda:0', grad_fn=<MseLossBackward>)
  1608. loss_train: 0.7524001447018236
  1609. step: 47
  1610. running loss: 0.016008513717060077
  1611. Train Steps: 47/90 Loss: 0.0160 torch.Size([8, 600, 800])
  1612. torch.Size([8, 8])
  1613. tensor([[0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  1614. [0.6210, 0.4164, 0.7202, 0.2930, 0.4025, 0.2483, 0.5687, 0.5567],
  1615. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  1616. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  1617. [0.6286, 0.4060, 0.9188, 0.4333, 0.3675, 0.4167, 0.7034, 0.5528],
  1618. [0.6193, 0.3930, 0.8949, 0.4437, 0.3852, 0.5435, 0.6263, 0.5263],
  1619. [0.6293, 0.3982, 0.8700, 0.5300, 0.3763, 0.4717, 0.7050, 0.5297],
  1620. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650]],
  1621. device='cuda:0', dtype=torch.float64)
  1622. predictions are: tensor([[0.5769, 0.4040, 0.7802, 0.1991, 0.4471, 0.2531, 0.5674, 0.5500],
  1623. [0.5461, 0.3844, 0.7615, 0.2243, 0.4086, 0.2583, 0.5673, 0.5589],
  1624. [0.5276, 0.3678, 0.8232, 0.4573, 0.4060, 0.5099, 0.5400, 0.5218],
  1625. [0.5709, 0.4187, 0.7707, 0.2387, 0.3961, 0.2889, 0.5726, 0.5709],
  1626. [0.5552, 0.3937, 0.8630, 0.3850, 0.4013, 0.4375, 0.5807, 0.5623],
  1627. [0.5481, 0.3639, 0.8349, 0.4752, 0.3836, 0.5852, 0.5817, 0.4908],
  1628. [0.5516, 0.3764, 0.8259, 0.5291, 0.3828, 0.5496, 0.5861, 0.5209],
  1629. [0.5723, 0.4223, 0.8104, 0.3714, 0.3790, 0.3846, 0.5357, 0.5459]],
  1630. device='cuda:0', grad_fn=<AddmmBackward>)
  1631. landmarks are: tensor([[[0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  1632. [0.6210, 0.4164, 0.7202, 0.2930, 0.4025, 0.2483, 0.5688, 0.5567],
  1633. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  1634. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  1635. [0.6286, 0.4060, 0.9187, 0.4333, 0.3675, 0.4167, 0.7034, 0.5528],
  1636. [0.6193, 0.3930, 0.8949, 0.4437, 0.3852, 0.5435, 0.6263, 0.5263],
  1637. [0.6293, 0.3982, 0.8700, 0.5300, 0.3762, 0.4717, 0.7050, 0.5297],
  1638. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650]]],
  1639. device='cuda:0')
  1640. loss_train_step before backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  1641. loss_train_step after backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  1642. loss_train: 0.7547844997607172
  1643. step: 48
  1644. running loss: 0.015724677078348275
  1645.  
  1646. Train Steps: 48/90 Loss: 0.0157 torch.Size([8, 600, 800])
  1647. torch.Size([8, 8])
  1648. tensor([[ nan, nan, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  1649. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  1650. [0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921],
  1651. [0.6203, 0.4078, 0.8800, 0.5083, 0.3900, 0.5000, 0.6100, 0.5583],
  1652. [0.6165, 0.4106, 0.7575, 0.1733, 0.3838, 0.2650, 0.5680, 0.5116],
  1653. [0.6087, 0.3976, 0.8337, 0.3867, 0.3713, 0.3117, 0.5938, 0.5300],
  1654. [0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  1655. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212]],
  1656. device='cuda:0', dtype=torch.float64)
  1657. predictions are: tensor([[0.4047, 0.2704, 0.7542, 0.2308, 0.4533, 0.2697, 0.5537, 0.5551],
  1658. [0.5581, 0.3529, 0.7849, 0.2889, 0.4571, 0.3005, 0.5780, 0.5572],
  1659. [0.6089, 0.4270, 0.8583, 0.2296, 0.4496, 0.3126, 0.6338, 0.5481],
  1660. [0.5649, 0.3624, 0.9076, 0.5557, 0.3788, 0.6005, 0.5847, 0.5724],
  1661. [0.5639, 0.3667, 0.7899, 0.3049, 0.3963, 0.3105, 0.6038, 0.5304],
  1662. [0.5850, 0.4091, 0.8458, 0.3845, 0.3912, 0.4504, 0.5918, 0.5393],
  1663. [0.5680, 0.3801, 0.8733, 0.4001, 0.4094, 0.3984, 0.6094, 0.5167],
  1664. [0.5417, 0.3740, 0.7820, 0.2999, 0.4156, 0.3376, 0.5991, 0.5239]],
  1665. device='cuda:0', grad_fn=<AddmmBackward>)
  1666. landmarks are: tensor([[[0.0000, 0.0000, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  1667. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  1668. [0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921],
  1669. [0.6203, 0.4078, 0.8800, 0.5083, 0.3900, 0.5000, 0.6100, 0.5583],
  1670. [0.6165, 0.4106, 0.7575, 0.1733, 0.3837, 0.2650, 0.5680, 0.5116],
  1671. [0.6087, 0.3976, 0.8338, 0.3867, 0.3713, 0.3117, 0.5938, 0.5300],
  1672. [0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  1673. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212]]],
  1674. device='cuda:0')
  1675. loss_train_step before backward: tensor(0.0062, device='cuda:0', grad_fn=<MseLossBackward>)
  1676. loss_train_step after backward: tensor(0.0062, device='cuda:0', grad_fn=<MseLossBackward>)
  1677. loss_train: 0.760986584238708
  1678. step: 49
  1679. running loss: 0.015530338453851184
  1680. Train Steps: 49/90 Loss: 0.0155 torch.Size([8, 600, 800])
  1681. torch.Size([8, 8])
  1682. tensor([[0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  1683. [0.6308, 0.3990, 0.8688, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  1684. [0.6102, 0.4001, 0.7738, 0.3583, 0.3463, 0.3800, 0.5524, 0.5689],
  1685. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  1686. [0.6125, 0.3974, 0.7725, 0.2517, 0.3538, 0.3317, 0.5887, 0.5500],
  1687. [0.6275, 0.4050, 0.9038, 0.3767, 0.3838, 0.3533, 0.7074, 0.5575],
  1688. [0.6299, 0.4008, 0.8450, 0.5350, 0.4213, 0.5000, 0.6350, 0.5100],
  1689. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817]],
  1690. device='cuda:0', dtype=torch.float64)
  1691. predictions are: tensor([[0.5269, 0.3406, 0.8398, 0.3721, 0.4167, 0.4091, 0.5909, 0.5420],
  1692. [0.5514, 0.3257, 0.8563, 0.4422, 0.3974, 0.4406, 0.6472, 0.5195],
  1693. [0.5518, 0.3459, 0.7828, 0.2487, 0.3847, 0.2868, 0.6127, 0.5394],
  1694. [0.5655, 0.3462, 0.8768, 0.3768, 0.4197, 0.4672, 0.6450, 0.5594],
  1695. [0.5349, 0.3501, 0.7877, 0.2137, 0.3759, 0.2686, 0.5964, 0.5461],
  1696. [0.5685, 0.3744, 0.9011, 0.2612, 0.4193, 0.2459, 0.6702, 0.5458],
  1697. [0.5460, 0.3551, 0.8509, 0.4093, 0.3993, 0.4086, 0.6418, 0.5202],
  1698. [0.5323, 0.3448, 0.8462, 0.3966, 0.4335, 0.3736, 0.5903, 0.5409]],
  1699. device='cuda:0', grad_fn=<AddmmBackward>)
  1700. landmarks are: tensor([[[0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  1701. [0.6308, 0.3990, 0.8687, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  1702. [0.6102, 0.4001, 0.7738, 0.3583, 0.3462, 0.3800, 0.5524, 0.5689],
  1703. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  1704. [0.6125, 0.3974, 0.7725, 0.2517, 0.3537, 0.3317, 0.5888, 0.5500],
  1705. [0.6275, 0.4050, 0.9038, 0.3767, 0.3837, 0.3533, 0.7074, 0.5575],
  1706. [0.6299, 0.4008, 0.8450, 0.5350, 0.4212, 0.5000, 0.6350, 0.5100],
  1707. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817]]],
  1708. device='cuda:0')
  1709. loss_train_step before backward: tensor(0.0044, device='cuda:0', grad_fn=<MseLossBackward>)
  1710. loss_train_step after backward: tensor(0.0044, device='cuda:0', grad_fn=<MseLossBackward>)
  1711. loss_train: 0.765395499765873
  1712. step: 50
  1713. running loss: 0.015307909995317458
  1714. Train Steps: 50/90 Loss: 0.0153 torch.Size([8, 600, 800])
  1715. torch.Size([8, 8])
  1716. tensor([[0.6137, 0.4038, 0.8563, 0.4050, 0.3813, 0.2550, 0.5106, 0.4954],
  1717. [0.6239, 0.4061, 0.8850, 0.4600, 0.4225, 0.5200, 0.6138, 0.5450],
  1718. [0.6159, 0.4085, 0.6900, 0.2283, 0.4088, 0.1950, 0.5123, 0.5397],
  1719. [0.6149, 0.4054, 0.6713, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  1720. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461],
  1721. [0.6126, 0.3954, 0.8538, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  1722. [0.6115, 0.4081, 0.6725, 0.2433, 0.4088, 0.1933, 0.5167, 0.5544],
  1723. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617]],
  1724. device='cuda:0', dtype=torch.float64)
  1725. predictions are: tensor([[0.5544, 0.3628, 0.8616, 0.2909, 0.4169, 0.2274, 0.6133, 0.5377],
  1726. [0.5905, 0.3875, 0.9099, 0.4365, 0.3978, 0.4963, 0.6773, 0.5268],
  1727. [0.4808, 0.2976, 0.7408, 0.1998, 0.4108, 0.1856, 0.6188, 0.5373],
  1728. [0.4906, 0.3093, 0.7173, 0.2025, 0.3813, 0.2083, 0.5842, 0.5237],
  1729. [0.5996, 0.3735, 0.8806, 0.4638, 0.4183, 0.4209, 0.6550, 0.5542],
  1730. [0.5829, 0.3653, 0.9094, 0.4636, 0.4024, 0.4556, 0.6436, 0.5377],
  1731. [0.4791, 0.3142, 0.7376, 0.1940, 0.3909, 0.1909, 0.5938, 0.5389],
  1732. [0.5878, 0.3725, 0.8417, 0.4316, 0.4024, 0.4445, 0.6350, 0.5331]],
  1733. device='cuda:0', grad_fn=<AddmmBackward>)
  1734. landmarks are: tensor([[[0.6137, 0.4038, 0.8562, 0.4050, 0.3812, 0.2550, 0.5106, 0.4954],
  1735. [0.6239, 0.4061, 0.8850, 0.4600, 0.4225, 0.5200, 0.6137, 0.5450],
  1736. [0.6159, 0.4085, 0.6900, 0.2283, 0.4087, 0.1950, 0.5123, 0.5397],
  1737. [0.6149, 0.4054, 0.6712, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  1738. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461],
  1739. [0.6126, 0.3954, 0.8537, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  1740. [0.6115, 0.4081, 0.6725, 0.2433, 0.4087, 0.1933, 0.5167, 0.5544],
  1741. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617]]],
  1742. device='cuda:0')
  1743. loss_train_step before backward: tensor(0.0035, device='cuda:0', grad_fn=<MseLossBackward>)
  1744. loss_train_step after backward: tensor(0.0035, device='cuda:0', grad_fn=<MseLossBackward>)
  1745. loss_train: 0.7689229550305754
  1746. step: 51
  1747. running loss: 0.015076920686874027
  1748. Train Steps: 51/90 Loss: 0.0151 torch.Size([8, 600, 800])
  1749. torch.Size([8, 8])
  1750. tensor([[0.6199, 0.4060, 0.8888, 0.4667, 0.3800, 0.5050, 0.6188, 0.5433],
  1751. [0.6263, 0.4030, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  1752. [0.6371, 0.4092, 0.8337, 0.5850, 0.3950, 0.5117, 0.6559, 0.5262],
  1753. [0.6100, 0.4071, 0.7601, 0.3444, 0.3400, 0.4117, 0.5625, 0.5617],
  1754. [0.6286, 0.4055, 0.9000, 0.4717, 0.3763, 0.4683, 0.7018, 0.5494],
  1755. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250],
  1756. [0.6183, 0.4076, 0.8838, 0.4517, 0.3813, 0.4483, 0.5775, 0.5633],
  1757. [0.6229, 0.4107, 0.8137, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400]],
  1758. device='cuda:0', dtype=torch.float64)
  1759. predictions are: tensor([[0.5841, 0.3481, 0.8627, 0.4467, 0.4061, 0.4344, 0.6089, 0.5218],
  1760. [0.5990, 0.3450, 0.8709, 0.4506, 0.3757, 0.4238, 0.5918, 0.5000],
  1761. [0.5780, 0.3840, 0.8173, 0.4550, 0.3951, 0.3989, 0.6279, 0.5487],
  1762. [0.5716, 0.3701, 0.7407, 0.2549, 0.3653, 0.2957, 0.6200, 0.5060],
  1763. [0.6063, 0.3638, 0.8475, 0.3999, 0.3978, 0.3789, 0.6600, 0.5402],
  1764. [0.6086, 0.3748, 0.8693, 0.3564, 0.3924, 0.2627, 0.6464, 0.5142],
  1765. [0.5666, 0.3266, 0.8481, 0.4245, 0.3842, 0.3965, 0.6189, 0.5329],
  1766. [0.5406, 0.3473, 0.7894, 0.1998, 0.4322, 0.1229, 0.6078, 0.5353]],
  1767. device='cuda:0', grad_fn=<AddmmBackward>)
  1768. landmarks are: tensor([[[0.6199, 0.4060, 0.8888, 0.4667, 0.3800, 0.5050, 0.6187, 0.5433],
  1769. [0.6263, 0.4029, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  1770. [0.6371, 0.4092, 0.8338, 0.5850, 0.3950, 0.5117, 0.6559, 0.5262],
  1771. [0.6100, 0.4071, 0.7601, 0.3444, 0.3400, 0.4117, 0.5625, 0.5617],
  1772. [0.6286, 0.4055, 0.9000, 0.4717, 0.3762, 0.4683, 0.7018, 0.5494],
  1773. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250],
  1774. [0.6183, 0.4076, 0.8838, 0.4517, 0.3812, 0.4483, 0.5775, 0.5633],
  1775. [0.6229, 0.4107, 0.8138, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400]]],
  1776. device='cuda:0')
  1777. loss_train_step before backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  1778. loss_train_step after backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  1779. loss_train: 0.7715486923698336
  1780. step: 52
  1781. running loss: 0.01483747485326603
  1782.  
  1783. Train Steps: 52/90 Loss: 0.0148 torch.Size([8, 600, 800])
  1784. torch.Size([8, 8])
  1785. tensor([[0.6277, 0.4083, 0.8350, 0.2717, 0.4562, 0.1800, 0.5918, 0.4878],
  1786. [0.6275, 0.4050, 0.9038, 0.3767, 0.3838, 0.3533, 0.7074, 0.5575],
  1787. [0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  1788. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  1789. [0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6262, 0.5367],
  1790. [0.6205, 0.4016, 0.8350, 0.2717, 0.3987, 0.2550, 0.5787, 0.5133],
  1791. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  1792. [ nan, nan, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737]],
  1793. device='cuda:0', dtype=torch.float64)
  1794. predictions are: tensor([[0.5546, 0.3509, 0.7689, 0.2416, 0.4326, 0.1638, 0.5948, 0.5102],
  1795. [0.6363, 0.4136, 0.8775, 0.4243, 0.3832, 0.2892, 0.6573, 0.5144],
  1796. [0.6187, 0.3739, 0.8605, 0.6347, 0.3554, 0.5029, 0.5993, 0.5492],
  1797. [0.5874, 0.3769, 0.7346, 0.3428, 0.3623, 0.3021, 0.6076, 0.5091],
  1798. [0.6045, 0.3746, 0.8653, 0.5789, 0.3946, 0.5484, 0.6011, 0.5000],
  1799. [0.6121, 0.3957, 0.7958, 0.3137, 0.3871, 0.2547, 0.6147, 0.5380],
  1800. [0.5961, 0.3638, 0.7663, 0.2782, 0.4365, 0.2224, 0.6491, 0.5104],
  1801. [0.5224, 0.3245, 0.8160, 0.2858, 0.4545, 0.2050, 0.6430, 0.5117]],
  1802. device='cuda:0', grad_fn=<AddmmBackward>)
  1803. landmarks are: tensor([[[0.6277, 0.4083, 0.8350, 0.2717, 0.4563, 0.1800, 0.5918, 0.4878],
  1804. [0.6275, 0.4050, 0.9038, 0.3767, 0.3837, 0.3533, 0.7074, 0.5575],
  1805. [0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  1806. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  1807. [0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6263, 0.5367],
  1808. [0.6205, 0.4015, 0.8350, 0.2717, 0.3988, 0.2550, 0.5788, 0.5133],
  1809. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  1810. [0.0000, 0.0000, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737]]],
  1811. device='cuda:0')
  1812. loss_train_step before backward: tensor(0.0079, device='cuda:0', grad_fn=<MseLossBackward>)
  1813. loss_train_step after backward: tensor(0.0079, device='cuda:0', grad_fn=<MseLossBackward>)
  1814. loss_train: 0.7794073515105993
  1815. step: 53
  1816. running loss: 0.014705799085105647
  1817. Train Steps: 53/90 Loss: 0.0147 torch.Size([8, 600, 800])
  1818. torch.Size([8, 8])
  1819. tensor([[0.6259, 0.4156, 0.8812, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960],
  1820. [0.6115, 0.4081, 0.6725, 0.2433, 0.4088, 0.1933, 0.5167, 0.5544],
  1821. [ nan, nan, 0.6793, 0.2110, 0.4012, 0.2167, 0.5112, 0.5583],
  1822. [0.6151, 0.4125, 0.8738, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  1823. [0.6236, 0.3967, 0.8675, 0.5400, 0.3862, 0.4517, 0.5825, 0.5200],
  1824. [0.6099, 0.4030, 0.8638, 0.5117, 0.4983, 0.4965, 0.5086, 0.5388],
  1825. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  1826. [0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500]],
  1827. device='cuda:0', dtype=torch.float64)
  1828. predictions are: tensor([[0.6415, 0.4262, 0.8866, 0.2842, 0.4599, 0.2407, 0.6511, 0.5367],
  1829. [0.5397, 0.3495, 0.7260, 0.2746, 0.3897, 0.2141, 0.5626, 0.5319],
  1830. [0.4686, 0.3070, 0.6932, 0.2379, 0.4155, 0.1864, 0.5605, 0.5481],
  1831. [0.6670, 0.4144, 0.8815, 0.4926, 0.3525, 0.4148, 0.5968, 0.5435],
  1832. [0.6459, 0.4183, 0.8974, 0.6042, 0.3895, 0.4822, 0.6102, 0.5310],
  1833. [0.6237, 0.3969, 0.8895, 0.5081, 0.4466, 0.4090, 0.6234, 0.5345],
  1834. [0.6468, 0.4130, 0.8978, 0.5317, 0.4216, 0.4910, 0.6532, 0.5334],
  1835. [0.6487, 0.4075, 0.7803, 0.3463, 0.3606, 0.3783, 0.6201, 0.5232]],
  1836. device='cuda:0', grad_fn=<AddmmBackward>)
  1837. landmarks are: tensor([[[0.6259, 0.4156, 0.8813, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960],
  1838. [0.6115, 0.4081, 0.6725, 0.2433, 0.4087, 0.1933, 0.5167, 0.5544],
  1839. [0.0000, 0.0000, 0.6793, 0.2110, 0.4013, 0.2167, 0.5113, 0.5583],
  1840. [0.6151, 0.4125, 0.8737, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  1841. [0.6236, 0.3967, 0.8675, 0.5400, 0.3862, 0.4517, 0.5825, 0.5200],
  1842. [0.6098, 0.4030, 0.8637, 0.5117, 0.4983, 0.4965, 0.5086, 0.5388],
  1843. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  1844. [0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500]]],
  1845. device='cuda:0')
  1846. loss_train_step before backward: tensor(0.0063, device='cuda:0', grad_fn=<MseLossBackward>)
  1847. loss_train_step after backward: tensor(0.0063, device='cuda:0', grad_fn=<MseLossBackward>)
  1848. loss_train: 0.7857216421980411
  1849. step: 54
  1850. running loss: 0.014550400781445205
  1851. Train Steps: 54/90 Loss: 0.0146 torch.Size([8, 600, 800])
  1852. torch.Size([8, 8])
  1853. tensor([[0.6228, 0.4004, 0.8750, 0.5250, 0.3825, 0.5233, 0.6362, 0.5000],
  1854. [0.6307, 0.4029, 0.8988, 0.4817, 0.3937, 0.3500, 0.7311, 0.5378],
  1855. [0.6251, 0.4163, 0.8662, 0.4467, 0.3625, 0.3567, 0.6038, 0.5533],
  1856. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  1857. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864],
  1858. [0.6263, 0.4065, 0.9038, 0.4317, 0.3588, 0.4550, 0.6325, 0.5250],
  1859. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  1860. [0.6273, 0.4100, 0.7137, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633]],
  1861. device='cuda:0', dtype=torch.float64)
  1862. predictions are: tensor([[0.6358, 0.4032, 0.8800, 0.5148, 0.4160, 0.5419, 0.5985, 0.5544],
  1863. [0.6271, 0.4309, 0.8779, 0.4180, 0.4278, 0.2989, 0.6445, 0.5209],
  1864. [0.6354, 0.4176, 0.8796, 0.4386, 0.3965, 0.3395, 0.5999, 0.5642],
  1865. [0.5834, 0.3736, 0.8202, 0.4424, 0.3896, 0.4175, 0.5881, 0.5508],
  1866. [0.6109, 0.4081, 0.8717, 0.4947, 0.4118, 0.4414, 0.6274, 0.5529],
  1867. [0.6070, 0.4307, 0.8827, 0.4406, 0.4242, 0.4110, 0.6265, 0.5525],
  1868. [0.6422, 0.4216, 0.8492, 0.5282, 0.4020, 0.4422, 0.6189, 0.5158],
  1869. [0.5524, 0.3625, 0.7314, 0.2355, 0.4092, 0.1922, 0.5738, 0.5495]],
  1870. device='cuda:0', grad_fn=<AddmmBackward>)
  1871. landmarks are: tensor([[[0.6228, 0.4004, 0.8750, 0.5250, 0.3825, 0.5233, 0.6363, 0.5000],
  1872. [0.6307, 0.4029, 0.8988, 0.4817, 0.3938, 0.3500, 0.7311, 0.5378],
  1873. [0.6252, 0.4162, 0.8662, 0.4467, 0.3625, 0.3567, 0.6037, 0.5533],
  1874. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  1875. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864],
  1876. [0.6263, 0.4065, 0.9038, 0.4317, 0.3587, 0.4550, 0.6325, 0.5250],
  1877. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  1878. [0.6273, 0.4099, 0.7138, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633]]],
  1879. device='cuda:0')
  1880. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  1881. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  1882. loss_train: 0.7869276243727654
  1883. step: 55
  1884. running loss: 0.014307774988595735
  1885. Train Steps: 55/90 Loss: 0.0143 torch.Size([8, 600, 800])
  1886. torch.Size([8, 8])
  1887. tensor([[0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  1888. [0.6057, 0.4011, 0.8750, 0.4267, 0.4400, 0.5800, 0.5845, 0.5585],
  1889. [0.6286, 0.4078, 0.8063, 0.2267, 0.4788, 0.1533, 0.5953, 0.4913],
  1890. [0.6241, 0.4143, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  1891. [0.6200, 0.4118, 0.8287, 0.4017, 0.3775, 0.2833, 0.5391, 0.5799],
  1892. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  1893. [0.6272, 0.4071, 0.8738, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  1894. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791]],
  1895. device='cuda:0', dtype=torch.float64)
  1896. predictions are: tensor([[0.5301, 0.3866, 0.7485, 0.3359, 0.3693, 0.2995, 0.5671, 0.5649],
  1897. [0.6037, 0.4085, 0.8429, 0.4616, 0.4316, 0.4904, 0.6046, 0.5223],
  1898. [0.5098, 0.3556, 0.8043, 0.2597, 0.4651, 0.1535, 0.5834, 0.5238],
  1899. [0.6246, 0.4276, 0.9140, 0.5233, 0.4322, 0.5490, 0.6235, 0.5671],
  1900. [0.5730, 0.4027, 0.8279, 0.3740, 0.3846, 0.2692, 0.5577, 0.5516],
  1901. [0.6058, 0.3992, 0.8399, 0.3735, 0.3873, 0.4737, 0.6092, 0.5630],
  1902. [0.6692, 0.4559, 0.9170, 0.5879, 0.3788, 0.4056, 0.6201, 0.5233],
  1903. [0.6449, 0.4282, 0.8855, 0.5277, 0.4079, 0.5283, 0.6165, 0.5521]],
  1904. device='cuda:0', grad_fn=<AddmmBackward>)
  1905. landmarks are: tensor([[[0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  1906. [0.6057, 0.4011, 0.8750, 0.4267, 0.4400, 0.5800, 0.5845, 0.5585],
  1907. [0.6286, 0.4078, 0.8062, 0.2267, 0.4787, 0.1533, 0.5953, 0.4913],
  1908. [0.6241, 0.4142, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  1909. [0.6200, 0.4118, 0.8288, 0.4017, 0.3775, 0.2833, 0.5391, 0.5799],
  1910. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  1911. [0.6272, 0.4071, 0.8737, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  1912. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791]]],
  1913. device='cuda:0')
  1914. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  1915. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  1916. loss_train: 0.7881838389439508
  1917. step: 56
  1918. running loss: 0.014074711409713407
  1919.  
  1920. Train Steps: 56/90 Loss: 0.0141 torch.Size([8, 600, 800])
  1921. torch.Size([8, 8])
  1922. tensor([[0.6117, 0.4019, 0.8538, 0.4067, 0.3513, 0.3583, 0.5663, 0.5133],
  1923. [0.6075, 0.4007, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280],
  1924. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5837, 0.5500],
  1925. [0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  1926. [0.6202, 0.4079, 0.8025, 0.2500, 0.3763, 0.3217, 0.6125, 0.5533],
  1927. [0.6270, 0.4267, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  1928. [0.6132, 0.4066, 0.7259, 0.2402, 0.3588, 0.3300, 0.6000, 0.5600],
  1929. [0.6179, 0.4082, 0.6688, 0.2667, 0.3588, 0.3317, 0.5750, 0.5783]],
  1930. device='cuda:0', dtype=torch.float64)
  1931. predictions are: tensor([[0.6256, 0.4479, 0.9064, 0.4920, 0.3685, 0.3892, 0.6073, 0.5520],
  1932. [0.6175, 0.4390, 0.8795, 0.5274, 0.4236, 0.5354, 0.5843, 0.5512],
  1933. [0.6787, 0.4470, 0.9259, 0.5724, 0.4245, 0.5948, 0.5997, 0.5572],
  1934. [0.6151, 0.4132, 0.8721, 0.4313, 0.4208, 0.4392, 0.6534, 0.5397],
  1935. [0.6369, 0.4252, 0.8739, 0.3726, 0.4443, 0.3527, 0.6142, 0.5384],
  1936. [0.5044, 0.3321, 0.7214, 0.3021, 0.4216, 0.3138, 0.5538, 0.5532],
  1937. [0.5753, 0.4049, 0.7476, 0.3612, 0.3829, 0.3842, 0.6129, 0.5504],
  1938. [0.5940, 0.4013, 0.7566, 0.3702, 0.3730, 0.3629, 0.5568, 0.5448]],
  1939. device='cuda:0', grad_fn=<AddmmBackward>)
  1940. landmarks are: tensor([[[0.6116, 0.4019, 0.8537, 0.4067, 0.3512, 0.3583, 0.5663, 0.5133],
  1941. [0.6075, 0.4006, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280],
  1942. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5838, 0.5500],
  1943. [0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  1944. [0.6202, 0.4079, 0.8025, 0.2500, 0.3762, 0.3217, 0.6125, 0.5533],
  1945. [0.6270, 0.4266, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  1946. [0.6132, 0.4066, 0.7259, 0.2402, 0.3587, 0.3300, 0.6000, 0.5600],
  1947. [0.6179, 0.4082, 0.6687, 0.2667, 0.3587, 0.3317, 0.5750, 0.5783]]],
  1948. device='cuda:0')
  1949. loss_train_step before backward: tensor(0.0031, device='cuda:0', grad_fn=<MseLossBackward>)
  1950. loss_train_step after backward: tensor(0.0031, device='cuda:0', grad_fn=<MseLossBackward>)
  1951. loss_train: 0.7912858905037865
  1952. step: 57
  1953. running loss: 0.013882208605329589
  1954. Train Steps: 57/90 Loss: 0.0139 torch.Size([8, 600, 800])
  1955. torch.Size([8, 8])
  1956. tensor([[ nan, nan, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  1957. [0.6162, 0.4134, 0.6700, 0.2467, 0.3962, 0.2533, 0.5737, 0.5467],
  1958. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  1959. [0.6305, 0.3983, 0.8950, 0.4833, 0.3688, 0.4683, 0.6375, 0.5117],
  1960. [0.6179, 0.4008, 0.8600, 0.4015, 0.3932, 0.2515, 0.5711, 0.5438],
  1961. [ nan, nan, 0.7725, 0.2611, 0.3675, 0.2733, 0.5413, 0.5167],
  1962. [0.6259, 0.4133, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  1963. [0.6118, 0.4052, 0.8463, 0.3917, 0.3538, 0.3450, 0.5053, 0.5593]],
  1964. device='cuda:0', dtype=torch.float64)
  1965. predictions are: tensor([[0.5422, 0.3630, 0.8784, 0.3457, 0.4616, 0.3802, 0.6294, 0.5449],
  1966. [0.5507, 0.3724, 0.7493, 0.3501, 0.3832, 0.3287, 0.5473, 0.5549],
  1967. [0.6227, 0.4195, 0.7417, 0.3306, 0.3928, 0.3698, 0.5464, 0.5404],
  1968. [0.7243, 0.4714, 0.9390, 0.5854, 0.3803, 0.6338, 0.6083, 0.5526],
  1969. [0.5849, 0.4123, 0.8004, 0.3820, 0.4038, 0.3513, 0.5566, 0.5422],
  1970. [0.5595, 0.3575, 0.7537, 0.3040, 0.3828, 0.3033, 0.5416, 0.5295],
  1971. [0.5844, 0.3820, 0.8241, 0.2918, 0.4535, 0.3019, 0.6126, 0.5243],
  1972. [0.6423, 0.4446, 0.8735, 0.4902, 0.3554, 0.5000, 0.5660, 0.5697]],
  1973. device='cuda:0', grad_fn=<AddmmBackward>)
  1974. landmarks are: tensor([[[0.0000, 0.0000, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  1975. [0.6162, 0.4134, 0.6700, 0.2467, 0.3963, 0.2533, 0.5738, 0.5467],
  1976. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  1977. [0.6305, 0.3983, 0.8950, 0.4833, 0.3688, 0.4683, 0.6375, 0.5117],
  1978. [0.6179, 0.4008, 0.8600, 0.4015, 0.3932, 0.2515, 0.5711, 0.5438],
  1979. [0.0000, 0.0000, 0.7725, 0.2611, 0.3675, 0.2733, 0.5412, 0.5167],
  1980. [0.6259, 0.4132, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  1981. [0.6118, 0.4052, 0.8462, 0.3917, 0.3537, 0.3450, 0.5053, 0.5593]]],
  1982. device='cuda:0')
  1983. loss_train_step before backward: tensor(0.0170, device='cuda:0', grad_fn=<MseLossBackward>)
  1984. loss_train_step after backward: tensor(0.0170, device='cuda:0', grad_fn=<MseLossBackward>)
  1985. loss_train: 0.8083162078401074
  1986. step: 58
  1987. running loss: 0.013936486342070818
  1988. Train Steps: 58/90 Loss: 0.0139 torch.Size([8, 600, 800])
  1989. torch.Size([8, 8])
  1990. tensor([[0.6115, 0.4005, 0.8838, 0.3867, 0.3763, 0.4700, 0.5800, 0.5550],
  1991. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  1992. [0.6133, 0.4094, 0.8495, 0.4028, 0.3588, 0.3200, 0.5003, 0.5407],
  1993. [ nan, nan, 0.6900, 0.1917, 0.3937, 0.2367, 0.5240, 0.5246],
  1994. [0.6134, 0.4090, 0.6926, 0.2819, 0.3538, 0.3233, 0.5563, 0.5667],
  1995. [0.6142, 0.3982, 0.8650, 0.4883, 0.3912, 0.4317, 0.5315, 0.5350],
  1996. [0.6182, 0.4058, 0.8738, 0.4350, 0.3563, 0.3400, 0.5290, 0.5822],
  1997. [0.6229, 0.4066, 0.8513, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350]],
  1998. device='cuda:0', dtype=torch.float64)
  1999. predictions are: tensor([[0.6941, 0.4152, 0.8931, 0.4134, 0.3910, 0.5324, 0.5766, 0.5431],
  2000. [0.5465, 0.3455, 0.7323, 0.2617, 0.4144, 0.2806, 0.5498, 0.5433],
  2001. [0.5984, 0.3990, 0.8667, 0.3813, 0.4116, 0.3690, 0.5511, 0.5340],
  2002. [0.4598, 0.3009, 0.7508, 0.1999, 0.4238, 0.2400, 0.5251, 0.5418],
  2003. [0.6028, 0.3708, 0.7092, 0.2750, 0.3748, 0.3487, 0.5502, 0.5231],
  2004. [0.6448, 0.4068, 0.8794, 0.5131, 0.4047, 0.5059, 0.5803, 0.5123],
  2005. [0.6070, 0.3968, 0.8930, 0.4225, 0.3829, 0.4504, 0.5513, 0.5751],
  2006. [0.6185, 0.3836, 0.8760, 0.4788, 0.4527, 0.4987, 0.5625, 0.5476]],
  2007. device='cuda:0', grad_fn=<AddmmBackward>)
  2008. landmarks are: tensor([[[0.6115, 0.4005, 0.8838, 0.3867, 0.3762, 0.4700, 0.5800, 0.5550],
  2009. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  2010. [0.6133, 0.4094, 0.8495, 0.4028, 0.3587, 0.3200, 0.5003, 0.5407],
  2011. [0.0000, 0.0000, 0.6900, 0.1917, 0.3938, 0.2367, 0.5240, 0.5246],
  2012. [0.6134, 0.4090, 0.6926, 0.2819, 0.3537, 0.3233, 0.5562, 0.5667],
  2013. [0.6143, 0.3982, 0.8650, 0.4883, 0.3913, 0.4317, 0.5315, 0.5350],
  2014. [0.6182, 0.4058, 0.8737, 0.4350, 0.3562, 0.3400, 0.5290, 0.5822],
  2015. [0.6229, 0.4066, 0.8512, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350]]],
  2016. device='cuda:0')
  2017. loss_train_step before backward: tensor(0.0060, device='cuda:0', grad_fn=<MseLossBackward>)
  2018. loss_train_step after backward: tensor(0.0060, device='cuda:0', grad_fn=<MseLossBackward>)
  2019. loss_train: 0.8143562333425507
  2020. step: 59
  2021. running loss: 0.013802648022755097
  2022. Train Steps: 59/90 Loss: 0.0138 torch.Size([8, 600, 800])
  2023. torch.Size([8, 8])
  2024. tensor([[0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  2025. [0.6204, 0.4049, 0.7975, 0.2700, 0.3937, 0.2567, 0.5700, 0.5183],
  2026. [0.6236, 0.3966, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  2027. [0.6112, 0.4029, 0.8638, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  2028. [0.6109, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  2029. [0.6068, 0.3963, 0.8650, 0.4317, 0.4037, 0.5083, 0.5253, 0.4999],
  2030. [0.6038, 0.3946, 0.8413, 0.4883, 0.3563, 0.4550, 0.5266, 0.4693],
  2031. [0.6132, 0.4037, 0.6963, 0.2217, 0.4100, 0.1950, 0.5395, 0.5175]],
  2032. device='cuda:0', dtype=torch.float64)
  2033. predictions are: tensor([[0.6058, 0.3606, 0.8605, 0.3885, 0.4073, 0.3083, 0.5800, 0.5317],
  2034. [0.5016, 0.3565, 0.7446, 0.2448, 0.4139, 0.2543, 0.5300, 0.5583],
  2035. [0.6240, 0.3779, 0.8711, 0.4626, 0.3618, 0.4597, 0.5244, 0.5394],
  2036. [0.5254, 0.3303, 0.8582, 0.3840, 0.4548, 0.4209, 0.5505, 0.5425],
  2037. [0.5991, 0.3819, 0.8540, 0.4017, 0.3868, 0.4029, 0.5255, 0.5292],
  2038. [0.6212, 0.3578, 0.8351, 0.3708, 0.3966, 0.4988, 0.5554, 0.5446],
  2039. [0.6145, 0.3990, 0.8441, 0.4090, 0.3736, 0.4491, 0.5455, 0.5312],
  2040. [0.4360, 0.2643, 0.6835, 0.1788, 0.4104, 0.1723, 0.5302, 0.5247]],
  2041. device='cuda:0', grad_fn=<AddmmBackward>)
  2042. landmarks are: tensor([[[0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  2043. [0.6204, 0.4049, 0.7975, 0.2700, 0.3938, 0.2567, 0.5700, 0.5183],
  2044. [0.6236, 0.3965, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  2045. [0.6112, 0.4029, 0.8637, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  2046. [0.6108, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  2047. [0.6068, 0.3963, 0.8650, 0.4317, 0.4038, 0.5083, 0.5253, 0.4999],
  2048. [0.6038, 0.3946, 0.8413, 0.4883, 0.3562, 0.4550, 0.5266, 0.4693],
  2049. [0.6132, 0.4037, 0.6963, 0.2217, 0.4100, 0.1950, 0.5395, 0.5175]]],
  2050. device='cuda:0')
  2051. loss_train_step before backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  2052. loss_train_step after backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  2053. loss_train: 0.8169467503903434
  2054. step: 60
  2055. running loss: 0.01361577917317239
  2056.  
  2057. Train Steps: 60/90 Loss: 0.0136 torch.Size([8, 600, 800])
  2058. torch.Size([8, 8])
  2059. tensor([[0.6234, 0.4179, 0.7825, 0.3450, 0.3813, 0.2867, 0.5675, 0.5617],
  2060. [0.6090, 0.4010, 0.7838, 0.3483, 0.3538, 0.3783, 0.5462, 0.5077],
  2061. [0.6109, 0.4015, 0.7668, 0.3639, 0.3513, 0.3667, 0.5200, 0.5641],
  2062. [0.6236, 0.3966, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  2063. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6038, 0.6167],
  2064. [0.6200, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  2065. [0.6261, 0.4045, 0.8865, 0.5369, 0.3895, 0.4859, 0.6683, 0.5249],
  2066. [0.6265, 0.4091, 0.8950, 0.3533, 0.3600, 0.3967, 0.6295, 0.4901]],
  2067. device='cuda:0', dtype=torch.float64)
  2068. predictions are: tensor([[0.4491, 0.2910, 0.7879, 0.2603, 0.4166, 0.1751, 0.5204, 0.5405],
  2069. [0.5714, 0.3382, 0.7695, 0.2836, 0.3397, 0.3514, 0.5478, 0.5101],
  2070. [0.5415, 0.3494, 0.7559, 0.2859, 0.3479, 0.3300, 0.5355, 0.5410],
  2071. [0.5705, 0.3587, 0.8571, 0.4829, 0.3522, 0.4165, 0.5035, 0.5268],
  2072. [0.5447, 0.3654, 0.8386, 0.2982, 0.4155, 0.3437, 0.5604, 0.5473],
  2073. [0.5558, 0.3338, 0.8648, 0.4201, 0.3983, 0.3573, 0.5427, 0.5397],
  2074. [0.6101, 0.3714, 0.8259, 0.4802, 0.3784, 0.4516, 0.5814, 0.5289],
  2075. [0.5437, 0.3351, 0.8664, 0.3020, 0.3978, 0.2938, 0.5660, 0.5327]],
  2076. device='cuda:0', grad_fn=<AddmmBackward>)
  2077. landmarks are: tensor([[[0.6234, 0.4179, 0.7825, 0.3450, 0.3812, 0.2867, 0.5675, 0.5617],
  2078. [0.6090, 0.4010, 0.7837, 0.3483, 0.3537, 0.3783, 0.5462, 0.5077],
  2079. [0.6109, 0.4015, 0.7668, 0.3639, 0.3512, 0.3667, 0.5200, 0.5641],
  2080. [0.6236, 0.3965, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  2081. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6037, 0.6167],
  2082. [0.6201, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  2083. [0.6261, 0.4045, 0.8865, 0.5369, 0.3895, 0.4859, 0.6683, 0.5249],
  2084. [0.6265, 0.4091, 0.8950, 0.3533, 0.3600, 0.3967, 0.6295, 0.4901]]],
  2085. device='cuda:0')
  2086. loss_train_step before backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  2087. loss_train_step after backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  2088. loss_train: 0.8201047728070989
  2089. step: 61
  2090. running loss: 0.013444340537821294
  2091. Train Steps: 61/90 Loss: 0.0134 torch.Size([8, 600, 800])
  2092. torch.Size([8, 8])
  2093. tensor([[0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167],
  2094. [ nan, nan, 0.7268, 0.2333, 0.4125, 0.1933, 0.5112, 0.5383],
  2095. [0.6236, 0.3966, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  2096. [0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767],
  2097. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  2098. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  2099. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  2100. [0.6218, 0.4098, 0.7238, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350]],
  2101. device='cuda:0', dtype=torch.float64)
  2102. predictions are: tensor([[0.5500, 0.3635, 0.7813, 0.2577, 0.3925, 0.3167, 0.5751, 0.5632],
  2103. [0.3665, 0.2547, 0.7070, 0.2034, 0.3935, 0.1840, 0.5407, 0.5374],
  2104. [0.6265, 0.3912, 0.8983, 0.4880, 0.3373, 0.4178, 0.5388, 0.5015],
  2105. [0.5959, 0.3870, 0.8860, 0.4874, 0.3806, 0.4561, 0.5588, 0.5384],
  2106. [0.5975, 0.3805, 0.8572, 0.4717, 0.3760, 0.3946, 0.5776, 0.5321],
  2107. [0.6113, 0.3528, 0.8609, 0.4085, 0.3538, 0.4293, 0.5591, 0.5227],
  2108. [0.5645, 0.3679, 0.8989, 0.3874, 0.3639, 0.3078, 0.5889, 0.5240],
  2109. [0.4978, 0.3402, 0.7412, 0.1953, 0.4422, 0.2193, 0.5693, 0.5261]],
  2110. device='cuda:0', grad_fn=<AddmmBackward>)
  2111. landmarks are: tensor([[[0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167],
  2112. [0.0000, 0.0000, 0.7268, 0.2333, 0.4125, 0.1933, 0.5113, 0.5383],
  2113. [0.6236, 0.3965, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  2114. [0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767],
  2115. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  2116. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  2117. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  2118. [0.6218, 0.4098, 0.7237, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350]]],
  2119. device='cuda:0')
  2120. loss_train_step before backward: tensor(0.0045, device='cuda:0', grad_fn=<MseLossBackward>)
  2121. loss_train_step after backward: tensor(0.0045, device='cuda:0', grad_fn=<MseLossBackward>)
  2122. loss_train: 0.824571170960553
  2123. step: 62
  2124. running loss: 0.01329953501549279
  2125. Train Steps: 62/90 Loss: 0.0133 torch.Size([8, 600, 800])
  2126. torch.Size([8, 8])
  2127. tensor([[0.6128, 0.4118, 0.8638, 0.5333, 0.4625, 0.5267, 0.5193, 0.5475],
  2128. [0.6197, 0.4118, 0.8688, 0.5517, 0.4037, 0.5233, 0.5875, 0.5600],
  2129. [0.6109, 0.4036, 0.7188, 0.1750, 0.3850, 0.2550, 0.5863, 0.5567],
  2130. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5787, 0.5117],
  2131. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167],
  2132. [0.6336, 0.4086, 0.8900, 0.3950, 0.3900, 0.2950, 0.6504, 0.5066],
  2133. [0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915],
  2134. [0.6276, 0.4095, 0.8237, 0.2250, 0.4662, 0.1783, 0.6171, 0.4869]],
  2135. device='cuda:0', dtype=torch.float64)
  2136. predictions are: tensor([[0.5856, 0.3832, 0.8667, 0.5038, 0.3929, 0.4306, 0.5876, 0.5351],
  2137. [0.6264, 0.3945, 0.8378, 0.5135, 0.3571, 0.5204, 0.6011, 0.5409],
  2138. [0.4624, 0.3228, 0.7371, 0.2446, 0.3710, 0.2221, 0.5621, 0.5435],
  2139. [0.4359, 0.2948, 0.7194, 0.2421, 0.3845, 0.2105, 0.5463, 0.5387],
  2140. [0.5081, 0.3626, 0.7719, 0.2733, 0.3700, 0.3125, 0.5887, 0.5739],
  2141. [0.5906, 0.3802, 0.8621, 0.3707, 0.3631, 0.3314, 0.6340, 0.4989],
  2142. [0.5955, 0.3969, 0.8825, 0.4975, 0.3163, 0.2913, 0.5767, 0.5016],
  2143. [0.4488, 0.3219, 0.8165, 0.2282, 0.4266, 0.1937, 0.6143, 0.5187]],
  2144. device='cuda:0', grad_fn=<AddmmBackward>)
  2145. landmarks are: tensor([[[0.6128, 0.4118, 0.8637, 0.5333, 0.4625, 0.5267, 0.5193, 0.5475],
  2146. [0.6197, 0.4118, 0.8687, 0.5517, 0.4038, 0.5233, 0.5875, 0.5600],
  2147. [0.6108, 0.4036, 0.7188, 0.1750, 0.3850, 0.2550, 0.5863, 0.5567],
  2148. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5788, 0.5117],
  2149. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167],
  2150. [0.6336, 0.4086, 0.8900, 0.3950, 0.3900, 0.2950, 0.6504, 0.5066],
  2151. [0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915],
  2152. [0.6276, 0.4095, 0.8238, 0.2250, 0.4663, 0.1783, 0.6171, 0.4869]]],
  2153. device='cuda:0')
  2154. loss_train_step before backward: tensor(0.0030, device='cuda:0', grad_fn=<MseLossBackward>)
  2155. loss_train_step after backward: tensor(0.0030, device='cuda:0', grad_fn=<MseLossBackward>)
  2156. loss_train: 0.8275592027930543
  2157. step: 63
  2158. running loss: 0.013135860361794512
  2159. Train Steps: 63/90 Loss: 0.0131 torch.Size([8, 600, 800])
  2160. torch.Size([8, 8])
  2161. tensor([[0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  2162. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  2163. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  2164. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  2165. [0.6223, 0.4028, 0.8988, 0.4200, 0.3763, 0.5733, 0.6375, 0.5167],
  2166. [0.6224, 0.4061, 0.8988, 0.4300, 0.3838, 0.4750, 0.6112, 0.5483],
  2167. [0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  2168. [0.6083, 0.3957, 0.8638, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980]],
  2169. device='cuda:0', dtype=torch.float64)
  2170. predictions are: tensor([[0.4981, 0.3343, 0.8217, 0.2960, 0.3723, 0.1908, 0.6148, 0.5354],
  2171. [0.4779, 0.3498, 0.6806, 0.3148, 0.3164, 0.2243, 0.5684, 0.5736],
  2172. [0.5703, 0.4017, 0.8371, 0.4836, 0.3770, 0.4784, 0.6296, 0.5814],
  2173. [0.4624, 0.3337, 0.7966, 0.2725, 0.4403, 0.1343, 0.6080, 0.5010],
  2174. [0.6624, 0.4184, 0.8434, 0.4282, 0.3695, 0.5177, 0.6353, 0.5406],
  2175. [0.6169, 0.4147, 0.8715, 0.4719, 0.3306, 0.4530, 0.6164, 0.5171],
  2176. [0.5956, 0.3959, 0.8104, 0.3552, 0.3451, 0.2360, 0.6289, 0.5024],
  2177. [0.5393, 0.3685, 0.8544, 0.4550, 0.3720, 0.3635, 0.5987, 0.5363]],
  2178. device='cuda:0', grad_fn=<AddmmBackward>)
  2179. landmarks are: tensor([[[0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  2180. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  2181. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  2182. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  2183. [0.6223, 0.4028, 0.8988, 0.4200, 0.3762, 0.5733, 0.6375, 0.5167],
  2184. [0.6224, 0.4061, 0.8988, 0.4300, 0.3837, 0.4750, 0.6112, 0.5483],
  2185. [0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  2186. [0.6083, 0.3957, 0.8637, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980]]],
  2187. device='cuda:0')
  2188. loss_train_step before backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  2189. loss_train_step after backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  2190. loss_train: 0.830406641936861
  2191. step: 64
  2192. running loss: 0.012975103780263453
  2193.  
  2194. Train Steps: 64/90 Loss: 0.0130 torch.Size([8, 600, 800])
  2195. torch.Size([8, 8])
  2196. tensor([[0.6229, 0.4066, 0.7612, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  2197. [0.6185, 0.4080, 0.8625, 0.3483, 0.3788, 0.2650, 0.5320, 0.5272],
  2198. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  2199. [0.6218, 0.4098, 0.7238, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  2200. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  2201. [0.6135, 0.3994, 0.7913, 0.3050, 0.3625, 0.3050, 0.5837, 0.5050],
  2202. [0.6159, 0.4085, 0.6900, 0.2283, 0.4088, 0.1950, 0.5123, 0.5397],
  2203. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6362, 0.5200]],
  2204. device='cuda:0', dtype=torch.float64)
  2205. predictions are: tensor([[0.5603, 0.4120, 0.7911, 0.3092, 0.4101, 0.2705, 0.6473, 0.5467],
  2206. [0.5694, 0.3968, 0.8457, 0.4254, 0.3745, 0.2732, 0.5703, 0.5492],
  2207. [0.6512, 0.4446, 0.9230, 0.5599, 0.3967, 0.4851, 0.6733, 0.5388],
  2208. [0.5368, 0.4146, 0.7461, 0.2814, 0.4314, 0.2485, 0.6166, 0.5398],
  2209. [0.5517, 0.3704, 0.7644, 0.3018, 0.4405, 0.2130, 0.5926, 0.5472],
  2210. [0.6188, 0.4373, 0.8200, 0.3571, 0.3411, 0.3280, 0.6295, 0.5389],
  2211. [0.4397, 0.3159, 0.7076, 0.2952, 0.4046, 0.2003, 0.5565, 0.5434],
  2212. [0.6508, 0.4446, 0.8944, 0.4353, 0.3638, 0.5242, 0.6494, 0.5329]],
  2213. device='cuda:0', grad_fn=<AddmmBackward>)
  2214. landmarks are: tensor([[[0.6229, 0.4066, 0.7613, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  2215. [0.6186, 0.4080, 0.8625, 0.3483, 0.3787, 0.2650, 0.5320, 0.5272],
  2216. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  2217. [0.6218, 0.4098, 0.7237, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  2218. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  2219. [0.6135, 0.3994, 0.7912, 0.3050, 0.3625, 0.3050, 0.5838, 0.5050],
  2220. [0.6159, 0.4085, 0.6900, 0.2283, 0.4087, 0.1950, 0.5123, 0.5397],
  2221. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6363, 0.5200]]],
  2222. device='cuda:0')
  2223. loss_train_step before backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  2224. loss_train_step after backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  2225. loss_train: 0.832502335193567
  2226. step: 65
  2227. running loss: 0.012807728233747184
  2228. Train Steps: 65/90 Loss: 0.0128 torch.Size([8, 600, 800])
  2229. torch.Size([8, 8])
  2230. tensor([[0.6286, 0.4060, 0.9188, 0.4333, 0.3675, 0.4167, 0.7034, 0.5528],
  2231. [0.6261, 0.4045, 0.8865, 0.5369, 0.3895, 0.4859, 0.6683, 0.5249],
  2232. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  2233. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  2234. [0.6346, 0.4086, 0.7938, 0.5500, 0.3962, 0.4867, 0.7343, 0.5702],
  2235. [ nan, nan, 0.7612, 0.3250, 0.4037, 0.2533, 0.5438, 0.5767],
  2236. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395],
  2237. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050]],
  2238. device='cuda:0', dtype=torch.float64)
  2239. predictions are: tensor([[0.6186, 0.4025, 0.9083, 0.4060, 0.3845, 0.3154, 0.6452, 0.5331],
  2240. [0.7243, 0.4844, 0.8551, 0.5121, 0.3872, 0.4598, 0.6464, 0.5225],
  2241. [0.6315, 0.4299, 0.7698, 0.2274, 0.3831, 0.2722, 0.6215, 0.5394],
  2242. [0.6787, 0.4502, 0.8737, 0.4614, 0.4429, 0.4977, 0.6371, 0.5566],
  2243. [0.6768, 0.4684, 0.8210, 0.4514, 0.4072, 0.4349, 0.6284, 0.5537],
  2244. [0.5218, 0.3466, 0.7645, 0.2660, 0.4297, 0.2230, 0.5633, 0.5546],
  2245. [0.5632, 0.3770, 0.7442, 0.2615, 0.4154, 0.1907, 0.6021, 0.5459],
  2246. [0.6338, 0.4305, 0.8497, 0.4663, 0.4445, 0.4249, 0.6135, 0.5289]],
  2247. device='cuda:0', grad_fn=<AddmmBackward>)
  2248. landmarks are: tensor([[[0.6286, 0.4060, 0.9187, 0.4333, 0.3675, 0.4167, 0.7034, 0.5528],
  2249. [0.6261, 0.4045, 0.8865, 0.5369, 0.3895, 0.4859, 0.6683, 0.5249],
  2250. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  2251. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  2252. [0.6346, 0.4086, 0.7937, 0.5500, 0.3963, 0.4867, 0.7343, 0.5702],
  2253. [0.0000, 0.0000, 0.7613, 0.3250, 0.4038, 0.2533, 0.5437, 0.5767],
  2254. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395],
  2255. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050]]],
  2256. device='cuda:0')
  2257. loss_train_step before backward: tensor(0.0078, device='cuda:0', grad_fn=<MseLossBackward>)
  2258. loss_train_step after backward: tensor(0.0078, device='cuda:0', grad_fn=<MseLossBackward>)
  2259. loss_train: 0.8403434584615752
  2260. step: 66
  2261. running loss: 0.0127324766433572
  2262. Train Steps: 66/90 Loss: 0.0127 torch.Size([8, 600, 800])
  2263. torch.Size([8, 8])
  2264. tensor([[0.6144, 0.4032, 0.8563, 0.3283, 0.3525, 0.4200, 0.5775, 0.5583],
  2265. [0.6085, 0.4005, 0.8400, 0.4317, 0.3763, 0.4750, 0.5476, 0.5058],
  2266. [0.6201, 0.4082, 0.8827, 0.3715, 0.3825, 0.2712, 0.5845, 0.5412],
  2267. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  2268. [0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767],
  2269. [0.6248, 0.4032, 0.7738, 0.1900, 0.4813, 0.1400, 0.5941, 0.4904],
  2270. [0.6202, 0.4053, 0.8638, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  2271. [0.6275, 0.4024, 0.8500, 0.5383, 0.3912, 0.4883, 0.6288, 0.5100]],
  2272. device='cuda:0', dtype=torch.float64)
  2273. predictions are: tensor([[0.6932, 0.4450, 0.8531, 0.3368, 0.3961, 0.3903, 0.6321, 0.5604],
  2274. [0.7086, 0.4471, 0.8215, 0.3704, 0.4030, 0.4471, 0.6338, 0.5231],
  2275. [0.6424, 0.4138, 0.8391, 0.3307, 0.4225, 0.2711, 0.6367, 0.5248],
  2276. [0.6202, 0.4266, 0.8456, 0.4140, 0.4342, 0.4385, 0.6412, 0.5650],
  2277. [0.6707, 0.4512, 0.8489, 0.4675, 0.4233, 0.4563, 0.6271, 0.5644],
  2278. [0.5194, 0.3450, 0.6983, 0.1942, 0.4850, 0.1519, 0.6323, 0.5231],
  2279. [0.6732, 0.4427, 0.8219, 0.4614, 0.4517, 0.4596, 0.6375, 0.5391],
  2280. [0.7076, 0.4632, 0.8452, 0.4561, 0.4193, 0.4764, 0.6555, 0.5310]],
  2281. device='cuda:0', grad_fn=<AddmmBackward>)
  2282. landmarks are: tensor([[[0.6144, 0.4032, 0.8562, 0.3283, 0.3525, 0.4200, 0.5775, 0.5583],
  2283. [0.6084, 0.4005, 0.8400, 0.4317, 0.3762, 0.4750, 0.5476, 0.5058],
  2284. [0.6201, 0.4082, 0.8827, 0.3715, 0.3825, 0.2712, 0.5845, 0.5412],
  2285. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  2286. [0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767],
  2287. [0.6248, 0.4032, 0.7738, 0.1900, 0.4812, 0.1400, 0.5941, 0.4904],
  2288. [0.6202, 0.4053, 0.8637, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  2289. [0.6275, 0.4024, 0.8500, 0.5383, 0.3913, 0.4883, 0.6288, 0.5100]]],
  2290. device='cuda:0')
  2291. loss_train_step before backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  2292. loss_train_step after backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  2293. loss_train: 0.84257042047102
  2294. step: 67
  2295. running loss: 0.012575677917477912
  2296. Train Steps: 67/90 Loss: 0.0126 torch.Size([8, 600, 800])
  2297. torch.Size([8, 8])
  2298. tensor([[0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  2299. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  2300. [0.6275, 0.4111, 0.8463, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  2301. [0.6202, 0.4053, 0.8638, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  2302. [0.6263, 0.4065, 0.9038, 0.4317, 0.3588, 0.4550, 0.6325, 0.5250],
  2303. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  2304. [0.6102, 0.4001, 0.7738, 0.3583, 0.3463, 0.3800, 0.5524, 0.5689],
  2305. [0.6229, 0.4066, 0.7612, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350]],
  2306. device='cuda:0', dtype=torch.float64)
  2307. predictions are: tensor([[0.6659, 0.4144, 0.8573, 0.4683, 0.4495, 0.4916, 0.6621, 0.5355],
  2308. [0.6986, 0.4502, 0.8708, 0.4760, 0.4388, 0.5419, 0.6633, 0.5493],
  2309. [0.6271, 0.3910, 0.8422, 0.2759, 0.4727, 0.2775, 0.6521, 0.5047],
  2310. [0.6675, 0.4245, 0.8207, 0.4953, 0.4582, 0.4860, 0.6060, 0.5264],
  2311. [0.6498, 0.4277, 0.8808, 0.4407, 0.4198, 0.4427, 0.6638, 0.5178],
  2312. [0.6913, 0.4296, 0.8735, 0.4553, 0.4264, 0.4849, 0.6353, 0.5129],
  2313. [0.6791, 0.4282, 0.7736, 0.3109, 0.3759, 0.3660, 0.6060, 0.5501],
  2314. [0.6112, 0.3996, 0.7485, 0.2576, 0.4500, 0.2823, 0.6342, 0.5452]],
  2315. device='cuda:0', grad_fn=<AddmmBackward>)
  2316. landmarks are: tensor([[[0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  2317. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  2318. [0.6275, 0.4111, 0.8462, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  2319. [0.6202, 0.4053, 0.8637, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  2320. [0.6263, 0.4065, 0.9038, 0.4317, 0.3587, 0.4550, 0.6325, 0.5250],
  2321. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  2322. [0.6102, 0.4001, 0.7738, 0.3583, 0.3462, 0.3800, 0.5524, 0.5689],
  2323. [0.6229, 0.4066, 0.7613, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350]]],
  2324. device='cuda:0')
  2325. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  2326. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  2327. loss_train: 0.8437589161330834
  2328. step: 68
  2329. running loss: 0.012408219354898286
  2330.  
  2331. Train Steps: 68/90 Loss: 0.0124 torch.Size([8, 600, 800])
  2332. torch.Size([8, 8])
  2333. tensor([[0.6230, 0.4152, 0.7588, 0.2283, 0.4012, 0.2883, 0.6200, 0.5767],
  2334. [0.6243, 0.4128, 0.7762, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417],
  2335. [0.6143, 0.4040, 0.8237, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  2336. [0.6139, 0.4019, 0.7137, 0.2150, 0.4375, 0.1533, 0.5293, 0.5006],
  2337. [0.6201, 0.3970, 0.8413, 0.4950, 0.4413, 0.5183, 0.6088, 0.5400],
  2338. [ nan, nan, 0.7097, 0.2346, 0.4250, 0.1850, 0.5175, 0.5583],
  2339. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  2340. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413]],
  2341. device='cuda:0', dtype=torch.float64)
  2342. predictions are: tensor([[0.6574, 0.4143, 0.8031, 0.3755, 0.4134, 0.3475, 0.6050, 0.5221],
  2343. [0.6055, 0.3822, 0.7905, 0.3181, 0.4092, 0.3100, 0.6078, 0.5171],
  2344. [0.6002, 0.3704, 0.7617, 0.3438, 0.4273, 0.2910, 0.5583, 0.5095],
  2345. [0.5412, 0.3202, 0.6806, 0.2306, 0.4217, 0.2505, 0.5640, 0.5077],
  2346. [0.6804, 0.4248, 0.9077, 0.5563, 0.4179, 0.5659, 0.5997, 0.5211],
  2347. [0.4818, 0.3003, 0.6873, 0.2609, 0.4478, 0.2683, 0.5393, 0.5116],
  2348. [0.7298, 0.4601, 0.8969, 0.5499, 0.4313, 0.6434, 0.6491, 0.5297],
  2349. [0.6359, 0.4006, 0.8891, 0.3423, 0.4818, 0.3901, 0.6773, 0.4832]],
  2350. device='cuda:0', grad_fn=<AddmmBackward>)
  2351. landmarks are: tensor([[[0.6230, 0.4152, 0.7588, 0.2283, 0.4013, 0.2883, 0.6200, 0.5767],
  2352. [0.6243, 0.4128, 0.7763, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417],
  2353. [0.6143, 0.4040, 0.8238, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  2354. [0.6139, 0.4019, 0.7138, 0.2150, 0.4375, 0.1533, 0.5293, 0.5006],
  2355. [0.6201, 0.3970, 0.8413, 0.4950, 0.4412, 0.5183, 0.6087, 0.5400],
  2356. [0.0000, 0.0000, 0.7097, 0.2346, 0.4250, 0.1850, 0.5175, 0.5583],
  2357. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  2358. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413]]],
  2359. device='cuda:0')
  2360. loss_train_step before backward: tensor(0.0075, device='cuda:0', grad_fn=<MseLossBackward>)
  2361. loss_train_step after backward: tensor(0.0075, device='cuda:0', grad_fn=<MseLossBackward>)
  2362. loss_train: 0.8512897320324555
  2363. step: 69
  2364. running loss: 0.012337532348296456
  2365. Train Steps: 69/90 Loss: 0.0123 torch.Size([8, 600, 800])
  2366. torch.Size([8, 8])
  2367. tensor([[0.6175, 0.4093, 0.8800, 0.4433, 0.4075, 0.4367, 0.5128, 0.4901],
  2368. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  2369. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  2370. [0.6109, 0.4003, 0.8650, 0.4883, 0.4775, 0.4867, 0.5175, 0.5683],
  2371. [0.6111, 0.3995, 0.8788, 0.4567, 0.3813, 0.4833, 0.5450, 0.5700],
  2372. [0.6043, 0.4022, 0.6887, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136],
  2373. [0.6346, 0.4165, 0.9138, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471],
  2374. [0.6262, 0.4163, 0.8850, 0.5183, 0.3763, 0.4150, 0.6025, 0.5500]],
  2375. device='cuda:0', dtype=torch.float64)
  2376. predictions are: tensor([[0.6429, 0.3846, 0.9090, 0.4461, 0.4379, 0.4578, 0.5886, 0.5220],
  2377. [0.6341, 0.3974, 0.8621, 0.5092, 0.4944, 0.4995, 0.6058, 0.5295],
  2378. [0.6139, 0.4076, 0.7800, 0.3218, 0.3755, 0.4063, 0.5946, 0.5400],
  2379. [0.6022, 0.3759, 0.8596, 0.4588, 0.5010, 0.5012, 0.6091, 0.5032],
  2380. [0.6845, 0.4206, 0.8694, 0.4623, 0.3940, 0.5087, 0.6284, 0.5446],
  2381. [0.5806, 0.3420, 0.6818, 0.2290, 0.4298, 0.2716, 0.5601, 0.5289],
  2382. [0.6133, 0.3850, 0.9095, 0.4131, 0.4331, 0.4100, 0.6622, 0.5053],
  2383. [0.6524, 0.3739, 0.8804, 0.4687, 0.3983, 0.4143, 0.6038, 0.5301]],
  2384. device='cuda:0', grad_fn=<AddmmBackward>)
  2385. landmarks are: tensor([[[0.6175, 0.4093, 0.8800, 0.4433, 0.4075, 0.4367, 0.5128, 0.4901],
  2386. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  2387. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  2388. [0.6109, 0.4003, 0.8650, 0.4883, 0.4775, 0.4867, 0.5175, 0.5683],
  2389. [0.6111, 0.3995, 0.8788, 0.4567, 0.3812, 0.4833, 0.5450, 0.5700],
  2390. [0.6043, 0.4022, 0.6888, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136],
  2391. [0.6346, 0.4165, 0.9137, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471],
  2392. [0.6262, 0.4163, 0.8850, 0.5183, 0.3762, 0.4150, 0.6025, 0.5500]]],
  2393. device='cuda:0')
  2394. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  2395. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  2396. loss_train: 0.8528305467916653
  2397. step: 70
  2398. running loss: 0.012183293525595218
  2399. Train Steps: 70/90 Loss: 0.0122 torch.Size([8, 600, 800])
  2400. torch.Size([8, 8])
  2401. tensor([[0.6026, 0.3979, 0.8550, 0.4233, 0.3613, 0.5233, 0.5582, 0.4967],
  2402. [0.6286, 0.4078, 0.8063, 0.2267, 0.4788, 0.1533, 0.5953, 0.4913],
  2403. [0.6102, 0.4001, 0.7738, 0.3583, 0.3463, 0.3800, 0.5524, 0.5689],
  2404. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  2405. [0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  2406. [0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  2407. [0.6286, 0.4055, 0.9000, 0.4717, 0.3763, 0.4683, 0.7018, 0.5494],
  2408. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683]],
  2409. device='cuda:0', dtype=torch.float64)
  2410. predictions are: tensor([[0.6255, 0.3748, 0.8938, 0.4418, 0.3911, 0.5406, 0.5918, 0.5268],
  2411. [0.5163, 0.3047, 0.7975, 0.2625, 0.4969, 0.1968, 0.5712, 0.5116],
  2412. [0.6273, 0.3768, 0.8290, 0.3421, 0.3700, 0.3842, 0.5794, 0.5481],
  2413. [0.5764, 0.3479, 0.7528, 0.3221, 0.3897, 0.3325, 0.5494, 0.5493],
  2414. [0.5990, 0.3626, 0.8662, 0.4235, 0.3871, 0.5108, 0.6057, 0.5491],
  2415. [0.6179, 0.3711, 0.8998, 0.5116, 0.4923, 0.4869, 0.5835, 0.5161],
  2416. [0.6275, 0.3973, 0.9284, 0.4910, 0.4084, 0.5210, 0.6650, 0.5278],
  2417. [0.6413, 0.3900, 0.8882, 0.5357, 0.3857, 0.4287, 0.5707, 0.5402]],
  2418. device='cuda:0', grad_fn=<AddmmBackward>)
  2419. landmarks are: tensor([[[0.6026, 0.3979, 0.8550, 0.4233, 0.3613, 0.5233, 0.5582, 0.4967],
  2420. [0.6286, 0.4078, 0.8062, 0.2267, 0.4787, 0.1533, 0.5953, 0.4913],
  2421. [0.6102, 0.4001, 0.7738, 0.3583, 0.3462, 0.3800, 0.5524, 0.5689],
  2422. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  2423. [0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  2424. [0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  2425. [0.6286, 0.4055, 0.9000, 0.4717, 0.3762, 0.4683, 0.7018, 0.5494],
  2426. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683]]],
  2427. device='cuda:0')
  2428. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  2429. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  2430. loss_train: 0.8542323596775532
  2431. step: 71
  2432. running loss: 0.01203144168559934
  2433. Train Steps: 71/90 Loss: 0.0120 torch.Size([8, 600, 800])
  2434. torch.Size([8, 8])
  2435. tensor([[0.6339, 0.4102, 0.8588, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089],
  2436. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  2437. [0.6118, 0.4052, 0.8463, 0.3917, 0.3538, 0.3450, 0.5053, 0.5593],
  2438. [0.6268, 0.4052, 0.8175, 0.2250, 0.4688, 0.1917, 0.6375, 0.5267],
  2439. [0.6157, 0.4102, 0.8513, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  2440. [0.6107, 0.4013, 0.8700, 0.4650, 0.5049, 0.5176, 0.5850, 0.5550],
  2441. [0.6276, 0.4095, 0.8237, 0.2250, 0.4662, 0.1783, 0.6171, 0.4869],
  2442. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986]],
  2443. device='cuda:0', dtype=torch.float64)
  2444. predictions are: tensor([[0.5770, 0.3702, 0.8246, 0.3224, 0.4389, 0.3208, 0.6044, 0.5446],
  2445. [0.5811, 0.3755, 0.8506, 0.5164, 0.3648, 0.5003, 0.5496, 0.5302],
  2446. [0.5792, 0.3675, 0.8441, 0.4449, 0.3259, 0.4114, 0.5041, 0.5739],
  2447. [0.5408, 0.3342, 0.7698, 0.2914, 0.4380, 0.2589, 0.5881, 0.5285],
  2448. [0.5530, 0.3469, 0.8460, 0.4321, 0.3400, 0.4086, 0.5180, 0.5602],
  2449. [0.5483, 0.3793, 0.8731, 0.4913, 0.4134, 0.4735, 0.5696, 0.5521],
  2450. [0.5468, 0.3407, 0.8052, 0.2672, 0.4373, 0.2658, 0.5853, 0.5293],
  2451. [0.5349, 0.3487, 0.8739, 0.4018, 0.4091, 0.3381, 0.5687, 0.5299]],
  2452. device='cuda:0', grad_fn=<AddmmBackward>)
  2453. landmarks are: tensor([[[0.6339, 0.4102, 0.8587, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089],
  2454. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  2455. [0.6118, 0.4052, 0.8462, 0.3917, 0.3537, 0.3450, 0.5053, 0.5593],
  2456. [0.6268, 0.4052, 0.8175, 0.2250, 0.4688, 0.1917, 0.6375, 0.5267],
  2457. [0.6157, 0.4102, 0.8512, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  2458. [0.6107, 0.4013, 0.8700, 0.4650, 0.5049, 0.5176, 0.5850, 0.5550],
  2459. [0.6276, 0.4095, 0.8238, 0.2250, 0.4663, 0.1783, 0.6171, 0.4869],
  2460. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986]]],
  2461. device='cuda:0')
  2462. loss_train_step before backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  2463. loss_train_step after backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  2464. loss_train: 0.856674063950777
  2465. step: 72
  2466. running loss: 0.011898250888205238
  2467.  
  2468. Train Steps: 72/90 Loss: 0.0119 torch.Size([8, 600, 800])
  2469. torch.Size([8, 8])
  2470. tensor([[0.6198, 0.4115, 0.7762, 0.2717, 0.3713, 0.3200, 0.5837, 0.5683],
  2471. [0.6202, 0.4079, 0.8025, 0.2500, 0.3763, 0.3217, 0.6125, 0.5533],
  2472. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5988, 0.5667],
  2473. [0.6198, 0.4164, 0.8700, 0.5067, 0.4625, 0.5650, 0.5464, 0.5197],
  2474. [0.6135, 0.3994, 0.7913, 0.3050, 0.3625, 0.3050, 0.5837, 0.5050],
  2475. [0.6200, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  2476. [0.6201, 0.3970, 0.8413, 0.4950, 0.4413, 0.5183, 0.6088, 0.5400],
  2477. [0.6161, 0.4055, 0.8675, 0.3867, 0.3713, 0.4033, 0.5195, 0.5162]],
  2478. device='cuda:0', dtype=torch.float64)
  2479. predictions are: tensor([[0.5721, 0.3648, 0.8202, 0.3269, 0.3815, 0.3023, 0.5253, 0.5557],
  2480. [0.6015, 0.3963, 0.8259, 0.3027, 0.3946, 0.2854, 0.5421, 0.5332],
  2481. [0.6075, 0.4009, 0.8630, 0.3853, 0.3760, 0.3307, 0.5422, 0.5464],
  2482. [0.5815, 0.3768, 0.8523, 0.5206, 0.4387, 0.4850, 0.5653, 0.5391],
  2483. [0.5945, 0.3826, 0.7995, 0.3192, 0.3579, 0.3361, 0.5497, 0.5543],
  2484. [0.5935, 0.3802, 0.8935, 0.5007, 0.3762, 0.4188, 0.5464, 0.5353],
  2485. [0.5296, 0.3543, 0.8798, 0.5008, 0.4285, 0.4748, 0.5198, 0.5715],
  2486. [0.5407, 0.3619, 0.8395, 0.4189, 0.3694, 0.3604, 0.4980, 0.5526]],
  2487. device='cuda:0', grad_fn=<AddmmBackward>)
  2488. landmarks are: tensor([[[0.6198, 0.4115, 0.7763, 0.2717, 0.3713, 0.3200, 0.5838, 0.5683],
  2489. [0.6202, 0.4079, 0.8025, 0.2500, 0.3762, 0.3217, 0.6125, 0.5533],
  2490. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5987, 0.5667],
  2491. [0.6198, 0.4164, 0.8700, 0.5067, 0.4625, 0.5650, 0.5464, 0.5197],
  2492. [0.6135, 0.3994, 0.7912, 0.3050, 0.3625, 0.3050, 0.5838, 0.5050],
  2493. [0.6201, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  2494. [0.6201, 0.3970, 0.8413, 0.4950, 0.4412, 0.5183, 0.6087, 0.5400],
  2495. [0.6161, 0.4055, 0.8675, 0.3867, 0.3713, 0.4033, 0.5195, 0.5162]]],
  2496. device='cuda:0')
  2497. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  2498. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  2499. loss_train: 0.8580122668063268
  2500. step: 73
  2501. running loss: 0.01175359269597708
  2502. Train Steps: 73/90 Loss: 0.0118 torch.Size([8, 600, 800])
  2503. torch.Size([8, 8])
  2504. tensor([[0.6277, 0.4083, 0.8350, 0.2717, 0.4562, 0.1800, 0.5918, 0.4878],
  2505. [0.6199, 0.4065, 0.7598, 0.2385, 0.4317, 0.1981, 0.5933, 0.5221],
  2506. [0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  2507. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  2508. [0.6204, 0.4049, 0.7975, 0.2700, 0.3937, 0.2567, 0.5700, 0.5183],
  2509. [0.6111, 0.3995, 0.8788, 0.4567, 0.3813, 0.4833, 0.5450, 0.5700],
  2510. [0.6132, 0.4066, 0.7259, 0.2402, 0.3588, 0.3300, 0.6000, 0.5600],
  2511. [0.6124, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485]],
  2512. device='cuda:0', dtype=torch.float64)
  2513. predictions are: tensor([[0.5556, 0.3674, 0.8377, 0.2737, 0.4335, 0.2239, 0.5279, 0.5071],
  2514. [0.5495, 0.3718, 0.7725, 0.2917, 0.4186, 0.2792, 0.5280, 0.5383],
  2515. [0.5975, 0.4284, 0.8890, 0.5499, 0.3918, 0.4092, 0.5249, 0.5415],
  2516. [0.5203, 0.3569, 0.8910, 0.2834, 0.4844, 0.2743, 0.6489, 0.5207],
  2517. [0.5784, 0.4225, 0.8261, 0.3489, 0.3764, 0.2834, 0.5185, 0.5526],
  2518. [0.6175, 0.4279, 0.9348, 0.5584, 0.3562, 0.5445, 0.5273, 0.5581],
  2519. [0.6138, 0.4256, 0.7771, 0.3004, 0.3516, 0.3386, 0.5488, 0.5665],
  2520. [0.6027, 0.4076, 0.7594, 0.3404, 0.3540, 0.3074, 0.4966, 0.5514]],
  2521. device='cuda:0', grad_fn=<AddmmBackward>)
  2522. landmarks are: tensor([[[0.6277, 0.4083, 0.8350, 0.2717, 0.4563, 0.1800, 0.5918, 0.4878],
  2523. [0.6199, 0.4065, 0.7598, 0.2385, 0.4317, 0.1981, 0.5933, 0.5221],
  2524. [0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  2525. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  2526. [0.6204, 0.4049, 0.7975, 0.2700, 0.3938, 0.2567, 0.5700, 0.5183],
  2527. [0.6111, 0.3995, 0.8788, 0.4567, 0.3812, 0.4833, 0.5450, 0.5700],
  2528. [0.6132, 0.4066, 0.7259, 0.2402, 0.3587, 0.3300, 0.6000, 0.5600],
  2529. [0.6123, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485]]],
  2530. device='cuda:0')
  2531. loss_train_step before backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  2532. loss_train_step after backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  2533. loss_train: 0.8606450074585155
  2534. step: 74
  2535. running loss: 0.011630337938628587
  2536. Train Steps: 74/90 Loss: 0.0116 torch.Size([8, 600, 800])
  2537. torch.Size([8, 8])
  2538. tensor([[0.6357, 0.4097, 0.9038, 0.3883, 0.4213, 0.2950, 0.6686, 0.5390],
  2539. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  2540. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  2541. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6038, 0.4833],
  2542. [0.6163, 0.4114, 0.7650, 0.2017, 0.3763, 0.2867, 0.5631, 0.5071],
  2543. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  2544. [0.6176, 0.3911, 0.8738, 0.4217, 0.3488, 0.4033, 0.6025, 0.4817],
  2545. [0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494]],
  2546. device='cuda:0', dtype=torch.float64)
  2547. predictions are: tensor([[0.6290, 0.4395, 0.8563, 0.3367, 0.4081, 0.2847, 0.6071, 0.5613],
  2548. [0.6373, 0.4309, 0.8770, 0.4503, 0.4436, 0.4187, 0.5412, 0.5536],
  2549. [0.6073, 0.4401, 0.8787, 0.4466, 0.4488, 0.4279, 0.5620, 0.5581],
  2550. [0.6246, 0.4447, 0.8888, 0.4558, 0.3932, 0.4469, 0.5294, 0.5504],
  2551. [0.6540, 0.4370, 0.7701, 0.2456, 0.3673, 0.2334, 0.5448, 0.5332],
  2552. [0.6194, 0.4169, 0.8621, 0.4388, 0.4443, 0.5149, 0.5497, 0.5537],
  2553. [0.6443, 0.4170, 0.8732, 0.3948, 0.3623, 0.3728, 0.5195, 0.5393],
  2554. [0.6201, 0.4285, 0.8864, 0.3037, 0.4047, 0.3044, 0.6332, 0.5474]],
  2555. device='cuda:0', grad_fn=<AddmmBackward>)
  2556. landmarks are: tensor([[[0.6357, 0.4097, 0.9038, 0.3883, 0.4212, 0.2950, 0.6686, 0.5390],
  2557. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  2558. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  2559. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6037, 0.4833],
  2560. [0.6163, 0.4114, 0.7650, 0.2017, 0.3762, 0.2867, 0.5631, 0.5071],
  2561. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  2562. [0.6176, 0.3911, 0.8737, 0.4217, 0.3487, 0.4033, 0.6025, 0.4817],
  2563. [0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494]]],
  2564. device='cuda:0')
  2565. loss_train_step before backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  2566. loss_train_step after backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  2567. loss_train: 0.8626611140789464
  2568. step: 75
  2569. running loss: 0.011502148187719285
  2570. Train Steps: 75/90 Loss: 0.0115 torch.Size([8, 600, 800])
  2571. torch.Size([8, 8])
  2572. tensor([[0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950],
  2573. [0.6222, 0.4169, 0.8638, 0.5650, 0.4313, 0.4783, 0.5637, 0.5633],
  2574. [0.6268, 0.4061, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  2575. [0.6275, 0.4003, 0.9100, 0.3783, 0.4388, 0.3133, 0.7058, 0.5343],
  2576. [0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  2577. [0.6296, 0.4045, 0.9138, 0.4100, 0.4232, 0.4242, 0.7422, 0.5297],
  2578. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  2579. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5413, 0.5717]],
  2580. device='cuda:0', dtype=torch.float64)
  2581. predictions are: tensor([[0.6245, 0.4360, 0.8635, 0.4273, 0.4271, 0.4352, 0.5655, 0.5258],
  2582. [0.6267, 0.4584, 0.8634, 0.4847, 0.3942, 0.4329, 0.5814, 0.5642],
  2583. [0.6581, 0.4601, 0.8488, 0.2100, 0.4235, 0.1943, 0.6559, 0.5183],
  2584. [0.6071, 0.4238, 0.8191, 0.2969, 0.3818, 0.2718, 0.6269, 0.5375],
  2585. [0.6349, 0.4543, 0.8456, 0.4415, 0.4136, 0.4470, 0.6092, 0.5495],
  2586. [0.6615, 0.4462, 0.8442, 0.3378, 0.3510, 0.3641, 0.6180, 0.5369],
  2587. [0.6187, 0.4142, 0.8872, 0.4145, 0.3976, 0.4471, 0.5936, 0.5266],
  2588. [0.6192, 0.4561, 0.8630, 0.4248, 0.4287, 0.4223, 0.5841, 0.5305]],
  2589. device='cuda:0', grad_fn=<AddmmBackward>)
  2590. landmarks are: tensor([[[0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950],
  2591. [0.6222, 0.4169, 0.8637, 0.5650, 0.4313, 0.4783, 0.5638, 0.5633],
  2592. [0.6268, 0.4060, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  2593. [0.6275, 0.4003, 0.9100, 0.3783, 0.4387, 0.3133, 0.7058, 0.5343],
  2594. [0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  2595. [0.6296, 0.4045, 0.9137, 0.4100, 0.4232, 0.4242, 0.7422, 0.5297],
  2596. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  2597. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5412, 0.5717]]],
  2598. device='cuda:0')
  2599. loss_train_step before backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  2600. loss_train_step after backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  2601. loss_train: 0.86508232972119
  2602. step: 76
  2603. running loss: 0.011382662233173553
  2604.  
  2605. Train Steps: 76/90 Loss: 0.0114 torch.Size([8, 600, 800])
  2606. torch.Size([8, 8])
  2607. tensor([[0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  2608. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  2609. [0.6204, 0.4055, 0.8438, 0.5733, 0.4574, 0.4801, 0.5487, 0.5617],
  2610. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  2611. [0.6250, 0.4131, 0.8688, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  2612. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  2613. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  2614. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938]],
  2615. device='cuda:0', dtype=torch.float64)
  2616. predictions are: tensor([[0.5771, 0.3978, 0.8029, 0.2722, 0.4141, 0.3073, 0.6815, 0.5198],
  2617. [0.6085, 0.4261, 0.8393, 0.4270, 0.3635, 0.4012, 0.5760, 0.5183],
  2618. [0.5968, 0.4245, 0.8060, 0.4617, 0.4207, 0.4515, 0.6187, 0.5316],
  2619. [0.6251, 0.4489, 0.8296, 0.3746, 0.3768, 0.3473, 0.5871, 0.5281],
  2620. [0.6556, 0.4485, 0.8583, 0.2192, 0.4339, 0.2290, 0.6672, 0.5177],
  2621. [0.6243, 0.4550, 0.7873, 0.1887, 0.4288, 0.1917, 0.6569, 0.5013],
  2622. [0.5891, 0.4459, 0.8296, 0.4930, 0.4048, 0.5645, 0.6507, 0.5437],
  2623. [0.6506, 0.4640, 0.8796, 0.4551, 0.3506, 0.4018, 0.6355, 0.4990]],
  2624. device='cuda:0', grad_fn=<AddmmBackward>)
  2625. landmarks are: tensor([[[0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  2626. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  2627. [0.6204, 0.4055, 0.8438, 0.5733, 0.4574, 0.4801, 0.5487, 0.5617],
  2628. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  2629. [0.6250, 0.4131, 0.8687, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  2630. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  2631. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  2632. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938]]],
  2633. device='cuda:0')
  2634. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  2635. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  2636. loss_train: 0.8669920227257535
  2637. step: 77
  2638. running loss: 0.01125963665877602
  2639. Train Steps: 77/90 Loss: 0.0113 torch.Size([8, 600, 800])
  2640. torch.Size([8, 8])
  2641. tensor([[0.6109, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  2642. [0.6205, 0.4062, 0.8337, 0.2683, 0.3675, 0.4283, 0.6338, 0.5250],
  2643. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  2644. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  2645. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  2646. [0.6125, 0.4035, 0.7825, 0.3100, 0.3463, 0.4900, 0.5832, 0.5637],
  2647. [0.6107, 0.4013, 0.8700, 0.4650, 0.5049, 0.5176, 0.5850, 0.5550],
  2648. [0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297]],
  2649. device='cuda:0', dtype=torch.float64)
  2650. predictions are: tensor([[0.6650, 0.4347, 0.8744, 0.4900, 0.3861, 0.4386, 0.6273, 0.5055],
  2651. [0.6044, 0.4057, 0.8026, 0.2726, 0.3897, 0.3852, 0.6577, 0.5237],
  2652. [0.7325, 0.5001, 0.8779, 0.5122, 0.4095, 0.3320, 0.6974, 0.5019],
  2653. [0.6298, 0.4239, 0.7692, 0.2225, 0.4536, 0.1487, 0.6777, 0.4890],
  2654. [0.6142, 0.3649, 0.8179, 0.2979, 0.4253, 0.2464, 0.6475, 0.5156],
  2655. [0.5920, 0.3943, 0.7995, 0.3386, 0.3788, 0.4432, 0.6688, 0.5305],
  2656. [0.6075, 0.4290, 0.8975, 0.4775, 0.4835, 0.4502, 0.6819, 0.5282],
  2657. [0.6853, 0.4624, 0.8745, 0.5447, 0.3783, 0.4823, 0.6775, 0.5141]],
  2658. device='cuda:0', grad_fn=<AddmmBackward>)
  2659. landmarks are: tensor([[[0.6108, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  2660. [0.6205, 0.4062, 0.8338, 0.2683, 0.3675, 0.4283, 0.6338, 0.5250],
  2661. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  2662. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  2663. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  2664. [0.6125, 0.4035, 0.7825, 0.3100, 0.3462, 0.4900, 0.5832, 0.5637],
  2665. [0.6107, 0.4013, 0.8700, 0.4650, 0.5049, 0.5176, 0.5850, 0.5550],
  2666. [0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297]]],
  2667. device='cuda:0')
  2668. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  2669. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  2670. loss_train: 0.8684362914646044
  2671. step: 78
  2672. running loss: 0.01113379860852057
  2673. Train Steps: 78/90 Loss: 0.0111 torch.Size([8, 600, 800])
  2674. torch.Size([8, 8])
  2675. tensor([[0.6216, 0.4099, 0.7225, 0.2033, 0.4188, 0.2217, 0.5975, 0.5283],
  2676. [0.6252, 0.4158, 0.8988, 0.4083, 0.3788, 0.4783, 0.6225, 0.5633],
  2677. [0.6115, 0.3998, 0.7063, 0.2383, 0.4037, 0.1950, 0.5320, 0.4993],
  2678. [0.6275, 0.4157, 0.8337, 0.5800, 0.3763, 0.4200, 0.5547, 0.6125],
  2679. [0.6113, 0.4088, 0.6859, 0.2208, 0.4363, 0.1700, 0.5188, 0.5533],
  2680. [0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166],
  2681. [0.6218, 0.4185, 0.7338, 0.2650, 0.4625, 0.1950, 0.5687, 0.5800],
  2682. [0.6222, 0.3937, 0.8350, 0.5617, 0.4138, 0.4600, 0.5800, 0.5233]],
  2683. device='cuda:0', dtype=torch.float64)
  2684. predictions are: tensor([[0.6333, 0.4204, 0.7791, 0.2819, 0.4128, 0.2189, 0.6612, 0.5254],
  2685. [0.7018, 0.4661, 0.9411, 0.4837, 0.3963, 0.5429, 0.7287, 0.5176],
  2686. [0.6514, 0.4204, 0.7638, 0.2597, 0.4110, 0.1882, 0.6433, 0.4838],
  2687. [0.7044, 0.4533, 0.8991, 0.6170, 0.3925, 0.5476, 0.6761, 0.5157],
  2688. [0.5008, 0.2885, 0.7548, 0.2519, 0.4445, 0.1971, 0.6265, 0.5172],
  2689. [0.6706, 0.4102, 0.9236, 0.5346, 0.4214, 0.6095, 0.6859, 0.4901],
  2690. [0.6292, 0.3884, 0.8051, 0.2587, 0.4408, 0.2057, 0.6468, 0.5499],
  2691. [0.6336, 0.3999, 0.9124, 0.6172, 0.4317, 0.5350, 0.6373, 0.5216]],
  2692. device='cuda:0', grad_fn=<AddmmBackward>)
  2693. landmarks are: tensor([[[0.6216, 0.4099, 0.7225, 0.2033, 0.4187, 0.2217, 0.5975, 0.5283],
  2694. [0.6252, 0.4158, 0.8988, 0.4083, 0.3787, 0.4783, 0.6225, 0.5633],
  2695. [0.6115, 0.3998, 0.7063, 0.2383, 0.4038, 0.1950, 0.5320, 0.4993],
  2696. [0.6275, 0.4157, 0.8338, 0.5800, 0.3762, 0.4200, 0.5547, 0.6125],
  2697. [0.6113, 0.4088, 0.6859, 0.2208, 0.4363, 0.1700, 0.5188, 0.5533],
  2698. [0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166],
  2699. [0.6218, 0.4185, 0.7337, 0.2650, 0.4625, 0.1950, 0.5688, 0.5800],
  2700. [0.6222, 0.3937, 0.8350, 0.5617, 0.4137, 0.4600, 0.5800, 0.5233]]],
  2701. device='cuda:0')
  2702. loss_train_step before backward: tensor(0.0033, device='cuda:0', grad_fn=<MseLossBackward>)
  2703. loss_train_step after backward: tensor(0.0033, device='cuda:0', grad_fn=<MseLossBackward>)
  2704. loss_train: 0.8717839835444465
  2705. step: 79
  2706. running loss: 0.011035240298030968
  2707. Train Steps: 79/90 Loss: 0.0110 torch.Size([8, 600, 800])
  2708. torch.Size([8, 8])
  2709. tensor([[0.6102, 0.4005, 0.8688, 0.5100, 0.4813, 0.5400, 0.5404, 0.5064],
  2710. [0.6353, 0.4128, 0.8488, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  2711. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  2712. [0.6124, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  2713. [0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297],
  2714. [0.6040, 0.4002, 0.7338, 0.2267, 0.3975, 0.2100, 0.5231, 0.4778],
  2715. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  2716. [0.6307, 0.4029, 0.8988, 0.4817, 0.3937, 0.3500, 0.7311, 0.5378]],
  2717. device='cuda:0', dtype=torch.float64)
  2718. predictions are: tensor([[0.6030, 0.3594, 0.8412, 0.5356, 0.4528, 0.4940, 0.6149, 0.5121],
  2719. [0.5933, 0.3381, 0.8520, 0.2899, 0.5123, 0.2131, 0.6733, 0.5340],
  2720. [0.6143, 0.3798, 0.8628, 0.5381, 0.4447, 0.5084, 0.6310, 0.5308],
  2721. [0.6493, 0.3699, 0.8399, 0.5038, 0.3668, 0.4863, 0.6120, 0.5405],
  2722. [0.6789, 0.4246, 0.8697, 0.5666, 0.3557, 0.4841, 0.6516, 0.5291],
  2723. [0.6382, 0.3728, 0.7099, 0.2474, 0.4178, 0.1808, 0.5965, 0.5131],
  2724. [0.6297, 0.3438, 0.8159, 0.3174, 0.4052, 0.2538, 0.6249, 0.5244],
  2725. [0.7083, 0.4381, 0.9010, 0.4947, 0.4144, 0.3560, 0.7052, 0.5052]],
  2726. device='cuda:0', grad_fn=<AddmmBackward>)
  2727. landmarks are: tensor([[[0.6102, 0.4005, 0.8687, 0.5100, 0.4812, 0.5400, 0.5404, 0.5064],
  2728. [0.6353, 0.4128, 0.8487, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  2729. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  2730. [0.6123, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  2731. [0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297],
  2732. [0.6040, 0.4002, 0.7337, 0.2267, 0.3975, 0.2100, 0.5231, 0.4778],
  2733. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  2734. [0.6307, 0.4029, 0.8988, 0.4817, 0.3938, 0.3500, 0.7311, 0.5378]]],
  2735. device='cuda:0')
  2736. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  2737. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  2738. loss_train: 0.8729040948674083
  2739. step: 80
  2740. running loss: 0.010911301185842603
  2741.  
  2742. Train Steps: 80/90 Loss: 0.0109 torch.Size([8, 600, 800])
  2743. torch.Size([8, 8])
  2744. tensor([[ nan, nan, 0.7515, 0.2708, 0.3987, 0.2267, 0.5162, 0.5567],
  2745. [0.6134, 0.4090, 0.6926, 0.2819, 0.3538, 0.3233, 0.5563, 0.5667],
  2746. [0.6200, 0.4049, 0.8638, 0.5617, 0.4125, 0.5100, 0.6013, 0.5317],
  2747. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  2748. [0.6246, 0.4028, 0.8738, 0.4867, 0.4088, 0.5667, 0.6362, 0.5200],
  2749. [0.6198, 0.4076, 0.7945, 0.2453, 0.4475, 0.1871, 0.5897, 0.5350],
  2750. [0.6277, 0.4103, 0.8087, 0.5717, 0.4188, 0.4750, 0.5663, 0.6083],
  2751. [0.6275, 0.4024, 0.8600, 0.2283, 0.5350, 0.1800, 0.7074, 0.5413]],
  2752. device='cuda:0', dtype=torch.float64)
  2753. predictions are: tensor([[0.5762, 0.3329, 0.7517, 0.2781, 0.4145, 0.1953, 0.5292, 0.5349],
  2754. [0.6681, 0.4076, 0.7123, 0.3287, 0.3506, 0.2726, 0.5663, 0.5310],
  2755. [0.6413, 0.3766, 0.8545, 0.6113, 0.4202, 0.5100, 0.5876, 0.5184],
  2756. [0.5882, 0.3514, 0.9092, 0.5452, 0.4597, 0.5661, 0.6020, 0.5366],
  2757. [0.6590, 0.3704, 0.8900, 0.5376, 0.4571, 0.5833, 0.6199, 0.5511],
  2758. [0.6882, 0.4035, 0.7938, 0.2930, 0.4340, 0.1886, 0.6237, 0.5297],
  2759. [0.6247, 0.3833, 0.8365, 0.5761, 0.4203, 0.4921, 0.5985, 0.5927],
  2760. [0.6403, 0.3778, 0.8570, 0.3007, 0.5033, 0.2290, 0.6974, 0.5138]],
  2761. device='cuda:0', grad_fn=<AddmmBackward>)
  2762. landmarks are: tensor([[[0.0000, 0.0000, 0.7515, 0.2708, 0.3988, 0.2267, 0.5163, 0.5567],
  2763. [0.6134, 0.4090, 0.6926, 0.2819, 0.3537, 0.3233, 0.5562, 0.5667],
  2764. [0.6199, 0.4049, 0.8637, 0.5617, 0.4125, 0.5100, 0.6012, 0.5317],
  2765. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  2766. [0.6246, 0.4028, 0.8737, 0.4867, 0.4087, 0.5667, 0.6363, 0.5200],
  2767. [0.6198, 0.4076, 0.7945, 0.2453, 0.4475, 0.1871, 0.5897, 0.5350],
  2768. [0.6277, 0.4103, 0.8087, 0.5717, 0.4187, 0.4750, 0.5663, 0.6083],
  2769. [0.6275, 0.4024, 0.8600, 0.2283, 0.5350, 0.1800, 0.7074, 0.5413]]],
  2770. device='cuda:0')
  2771. loss_train_step before backward: tensor(0.0079, device='cuda:0', grad_fn=<MseLossBackward>)
  2772. loss_train_step after backward: tensor(0.0079, device='cuda:0', grad_fn=<MseLossBackward>)
  2773. loss_train: 0.880808724090457
  2774. step: 81
  2775. running loss: 0.01087418177889453
  2776. Train Steps: 81/90 Loss: 0.0109 torch.Size([8, 600, 800])
  2777. torch.Size([8, 8])
  2778. tensor([[0.6261, 0.4045, 0.8865, 0.5369, 0.3895, 0.4859, 0.6683, 0.5249],
  2779. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  2780. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  2781. [0.6090, 0.4010, 0.7838, 0.3483, 0.3538, 0.3783, 0.5462, 0.5077],
  2782. [0.6170, 0.4102, 0.7468, 0.3695, 0.3463, 0.3767, 0.5238, 0.5823],
  2783. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  2784. [0.6133, 0.4066, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358],
  2785. [0.6329, 0.4055, 0.9050, 0.4783, 0.3613, 0.3917, 0.6464, 0.5019]],
  2786. device='cuda:0', dtype=torch.float64)
  2787. predictions are: tensor([[0.6216, 0.3691, 0.8633, 0.5882, 0.4457, 0.5147, 0.6355, 0.5467],
  2788. [0.6850, 0.3830, 0.8209, 0.3345, 0.4442, 0.2901, 0.6099, 0.5626],
  2789. [0.6290, 0.3722, 0.8811, 0.4910, 0.4291, 0.4807, 0.6123, 0.5476],
  2790. [0.6498, 0.3795, 0.8206, 0.3536, 0.3999, 0.3637, 0.5786, 0.5534],
  2791. [0.6327, 0.3948, 0.7703, 0.3668, 0.4033, 0.3676, 0.5708, 0.5905],
  2792. [0.6211, 0.3289, 0.8803, 0.5134, 0.4360, 0.4698, 0.5691, 0.5627],
  2793. [0.5953, 0.3649, 0.7168, 0.2784, 0.4429, 0.1925, 0.5494, 0.5407],
  2794. [0.6489, 0.3648, 0.8822, 0.5375, 0.4523, 0.4195, 0.5986, 0.5419]],
  2795. device='cuda:0', grad_fn=<AddmmBackward>)
  2796. landmarks are: tensor([[[0.6261, 0.4045, 0.8865, 0.5369, 0.3895, 0.4859, 0.6683, 0.5249],
  2797. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  2798. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  2799. [0.6090, 0.4010, 0.7837, 0.3483, 0.3537, 0.3783, 0.5462, 0.5077],
  2800. [0.6170, 0.4102, 0.7468, 0.3695, 0.3462, 0.3767, 0.5238, 0.5823],
  2801. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  2802. [0.6133, 0.4065, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358],
  2803. [0.6329, 0.4055, 0.9050, 0.4783, 0.3613, 0.3917, 0.6464, 0.5019]]],
  2804. device='cuda:0')
  2805. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  2806. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  2807. loss_train: 0.882431311882101
  2808. step: 82
  2809. running loss: 0.010761357461976841
  2810. Train Steps: 82/90 Loss: 0.0108 torch.Size([8, 600, 800])
  2811. torch.Size([8, 8])
  2812. tensor([[0.6147, 0.4026, 0.6600, 0.2467, 0.4088, 0.2150, 0.5489, 0.5773],
  2813. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791],
  2814. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  2815. [0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  2816. [0.6339, 0.4118, 0.7988, 0.5800, 0.3912, 0.4583, 0.7343, 0.5760],
  2817. [ nan, nan, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600],
  2818. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  2819. [ nan, nan, 0.8363, 0.3317, 0.3563, 0.3367, 0.5329, 0.5142]],
  2820. device='cuda:0', dtype=torch.float64)
  2821. predictions are: tensor([[0.6214, 0.3727, 0.7104, 0.2819, 0.4215, 0.2213, 0.5095, 0.5704],
  2822. [0.6013, 0.3810, 0.8906, 0.5344, 0.4157, 0.5617, 0.6100, 0.5443],
  2823. [0.6315, 0.3857, 0.8651, 0.4446, 0.3643, 0.4125, 0.5376, 0.5502],
  2824. [0.6312, 0.3986, 0.8583, 0.5682, 0.5022, 0.4889, 0.5536, 0.5491],
  2825. [0.5955, 0.4035, 0.8144, 0.5087, 0.3951, 0.4339, 0.5913, 0.5845],
  2826. [0.5218, 0.3318, 0.7971, 0.2683, 0.4273, 0.2634, 0.5459, 0.5848],
  2827. [0.5299, 0.3395, 0.7230, 0.2444, 0.4156, 0.1974, 0.5371, 0.5694],
  2828. [0.5515, 0.3686, 0.8005, 0.3105, 0.3675, 0.3014, 0.5491, 0.5271]],
  2829. device='cuda:0', grad_fn=<AddmmBackward>)
  2830. landmarks are: tensor([[[0.6147, 0.4026, 0.6600, 0.2467, 0.4087, 0.2150, 0.5489, 0.5773],
  2831. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791],
  2832. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  2833. [0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  2834. [0.6339, 0.4118, 0.7987, 0.5800, 0.3913, 0.4583, 0.7343, 0.5760],
  2835. [0.0000, 0.0000, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600],
  2836. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  2837. [0.0000, 0.0000, 0.8363, 0.3317, 0.3562, 0.3367, 0.5329, 0.5142]]],
  2838. device='cuda:0')
  2839. loss_train_step before backward: tensor(0.0140, device='cuda:0', grad_fn=<MseLossBackward>)
  2840. loss_train_step after backward: tensor(0.0140, device='cuda:0', grad_fn=<MseLossBackward>)
  2841. loss_train: 0.8964379815151915
  2842. step: 83
  2843. running loss: 0.010800457608616766
  2844. Train Steps: 83/90 Loss: 0.0108 torch.Size([8, 600, 800])
  2845. torch.Size([8, 8])
  2846. tensor([[0.6140, 0.4070, 0.8700, 0.5000, 0.4612, 0.4900, 0.5260, 0.5852],
  2847. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  2848. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667],
  2849. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5963, 0.6217],
  2850. [0.6293, 0.3982, 0.8700, 0.5300, 0.3763, 0.4717, 0.7050, 0.5297],
  2851. [0.6257, 0.4024, 0.8672, 0.5422, 0.4196, 0.5198, 0.6694, 0.5204],
  2852. [0.6135, 0.3994, 0.7913, 0.3050, 0.3625, 0.3050, 0.5837, 0.5050],
  2853. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400]],
  2854. device='cuda:0', dtype=torch.float64)
  2855. predictions are: tensor([[0.5326, 0.3346, 0.8373, 0.5127, 0.4292, 0.4920, 0.5383, 0.5489],
  2856. [0.5526, 0.3637, 0.8333, 0.2463, 0.4774, 0.2220, 0.6235, 0.5523],
  2857. [0.5676, 0.3812, 0.8502, 0.3600, 0.3812, 0.3840, 0.5246, 0.5630],
  2858. [0.5029, 0.3379, 0.7094, 0.2572, 0.4360, 0.2681, 0.5220, 0.5629],
  2859. [0.5869, 0.3627, 0.8255, 0.5369, 0.3665, 0.4870, 0.5639, 0.5548],
  2860. [0.5658, 0.3800, 0.8173, 0.5128, 0.4016, 0.4779, 0.5786, 0.5503],
  2861. [0.5854, 0.3794, 0.7980, 0.2851, 0.3610, 0.3052, 0.5562, 0.5387],
  2862. [0.5243, 0.3401, 0.8351, 0.3306, 0.3540, 0.4555, 0.5691, 0.5675]],
  2863. device='cuda:0', grad_fn=<AddmmBackward>)
  2864. landmarks are: tensor([[[0.6140, 0.4070, 0.8700, 0.5000, 0.4613, 0.4900, 0.5260, 0.5852],
  2865. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  2866. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667],
  2867. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5962, 0.6217],
  2868. [0.6293, 0.3982, 0.8700, 0.5300, 0.3762, 0.4717, 0.7050, 0.5297],
  2869. [0.6257, 0.4024, 0.8672, 0.5422, 0.4196, 0.5198, 0.6694, 0.5204],
  2870. [0.6135, 0.3994, 0.7912, 0.3050, 0.3625, 0.3050, 0.5838, 0.5050],
  2871. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400]]],
  2872. device='cuda:0')
  2873. loss_train_step before backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  2874. loss_train_step after backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  2875. loss_train: 0.8985337674384937
  2876. step: 84
  2877. running loss: 0.010696830564743973
  2878.  
  2879. Train Steps: 84/90 Loss: 0.0107 torch.Size([8, 600, 800])
  2880. torch.Size([8, 8])
  2881. tensor([[0.6085, 0.4005, 0.8400, 0.4317, 0.3763, 0.4750, 0.5476, 0.5058],
  2882. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5963, 0.6217],
  2883. [0.6333, 0.4037, 0.8638, 0.5733, 0.4012, 0.4717, 0.6369, 0.4938],
  2884. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  2885. [0.6214, 0.4112, 0.7838, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  2886. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  2887. [0.6293, 0.3982, 0.8700, 0.5300, 0.3763, 0.4717, 0.7050, 0.5297],
  2888. [0.6277, 0.4103, 0.8087, 0.5717, 0.4188, 0.4750, 0.5663, 0.6083]],
  2889. device='cuda:0', dtype=torch.float64)
  2890. predictions are: tensor([[0.5225, 0.3469, 0.8365, 0.3800, 0.3507, 0.4460, 0.5555, 0.5378],
  2891. [0.4763, 0.3448, 0.7153, 0.2141, 0.4209, 0.2689, 0.5513, 0.5684],
  2892. [0.5475, 0.3803, 0.8438, 0.5036, 0.3565, 0.4401, 0.5760, 0.5350],
  2893. [0.4865, 0.3568, 0.8518, 0.3545, 0.3320, 0.3578, 0.5361, 0.5598],
  2894. [0.5865, 0.3979, 0.8120, 0.2146, 0.3638, 0.2559, 0.5730, 0.5537],
  2895. [0.4913, 0.3482, 0.8402, 0.4484, 0.4285, 0.4935, 0.5410, 0.5497],
  2896. [0.5621, 0.3714, 0.8258, 0.4831, 0.3521, 0.4723, 0.6025, 0.5578],
  2897. [0.5258, 0.3645, 0.8049, 0.4832, 0.3943, 0.4806, 0.5652, 0.6209]],
  2898. device='cuda:0', grad_fn=<AddmmBackward>)
  2899. landmarks are: tensor([[[0.6084, 0.4005, 0.8400, 0.4317, 0.3762, 0.4750, 0.5476, 0.5058],
  2900. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5962, 0.6217],
  2901. [0.6334, 0.4037, 0.8637, 0.5733, 0.4013, 0.4717, 0.6369, 0.4938],
  2902. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  2903. [0.6214, 0.4112, 0.7837, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  2904. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  2905. [0.6293, 0.3982, 0.8700, 0.5300, 0.3762, 0.4717, 0.7050, 0.5297],
  2906. [0.6277, 0.4103, 0.8087, 0.5717, 0.4187, 0.4750, 0.5663, 0.6083]]],
  2907. device='cuda:0')
  2908. loss_train_step before backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  2909. loss_train_step after backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  2910. loss_train: 0.901197498315014
  2911. step: 85
  2912. running loss: 0.0106023235095884
  2913. Train Steps: 85/90 Loss: 0.0106 torch.Size([8, 600, 800])
  2914. torch.Size([8, 8])
  2915. tensor([[0.6170, 0.4102, 0.7468, 0.3695, 0.3463, 0.3767, 0.5238, 0.5823],
  2916. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  2917. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  2918. [ nan, nan, 0.8363, 0.3317, 0.3563, 0.3367, 0.5329, 0.5142],
  2919. [0.6133, 0.4066, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358],
  2920. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  2921. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  2922. [0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915]],
  2923. device='cuda:0', dtype=torch.float64)
  2924. predictions are: tensor([[0.5604, 0.4111, 0.7505, 0.3297, 0.3333, 0.3955, 0.5836, 0.5738],
  2925. [0.4741, 0.3754, 0.8639, 0.3444, 0.4385, 0.3285, 0.6015, 0.5387],
  2926. [0.4714, 0.3799, 0.8131, 0.3051, 0.3737, 0.3540, 0.5557, 0.5612],
  2927. [0.3484, 0.2857, 0.7763, 0.2619, 0.3515, 0.3478, 0.5347, 0.5393],
  2928. [0.4544, 0.3571, 0.6789, 0.2221, 0.3817, 0.2357, 0.5271, 0.5358],
  2929. [0.5848, 0.3920, 0.8319, 0.3195, 0.3314, 0.4011, 0.6398, 0.5342],
  2930. [0.5762, 0.3898, 0.8072, 0.5178, 0.3728, 0.5168, 0.5860, 0.5971],
  2931. [0.5460, 0.3911, 0.8495, 0.4813, 0.3454, 0.3946, 0.5857, 0.5254]],
  2932. device='cuda:0', grad_fn=<AddmmBackward>)
  2933. landmarks are: tensor([[[0.6170, 0.4102, 0.7468, 0.3695, 0.3462, 0.3767, 0.5238, 0.5823],
  2934. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  2935. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  2936. [0.0000, 0.0000, 0.8363, 0.3317, 0.3562, 0.3367, 0.5329, 0.5142],
  2937. [0.6133, 0.4065, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358],
  2938. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  2939. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  2940. [0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915]]],
  2941. device='cuda:0')
  2942. loss_train_step before backward: tensor(0.0056, device='cuda:0', grad_fn=<MseLossBackward>)
  2943. loss_train_step after backward: tensor(0.0056, device='cuda:0', grad_fn=<MseLossBackward>)
  2944. loss_train: 0.9067885667318478
  2945. step: 86
  2946. running loss: 0.010544053101533115
  2947. Train Steps: 86/90 Loss: 0.0105 torch.Size([8, 600, 800])
  2948. torch.Size([8, 8])
  2949. tensor([[0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364],
  2950. [0.6058, 0.3978, 0.8287, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461],
  2951. [0.6059, 0.4002, 0.7562, 0.2767, 0.3538, 0.3033, 0.5529, 0.5455],
  2952. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  2953. [0.6203, 0.4021, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  2954. [ nan, nan, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  2955. [0.6129, 0.4063, 0.8738, 0.5250, 0.4313, 0.4733, 0.5230, 0.5874],
  2956. [0.6263, 0.4065, 0.9038, 0.4317, 0.3588, 0.4550, 0.6325, 0.5250]],
  2957. device='cuda:0', dtype=torch.float64)
  2958. predictions are: tensor([[0.4804, 0.3765, 0.7629, 0.2209, 0.4160, 0.2672, 0.5957, 0.5630],
  2959. [0.5317, 0.4018, 0.8057, 0.3704, 0.3130, 0.4027, 0.5884, 0.5213],
  2960. [0.5202, 0.4078, 0.7146, 0.2645, 0.3400, 0.3162, 0.5732, 0.5277],
  2961. [0.5567, 0.4072, 0.7966, 0.2697, 0.3687, 0.3272, 0.5992, 0.5057],
  2962. [0.5670, 0.3871, 0.8749, 0.5211, 0.3079, 0.4514, 0.5793, 0.5269],
  2963. [0.3928, 0.3086, 0.8130, 0.2346, 0.4734, 0.2574, 0.5827, 0.5535],
  2964. [0.5530, 0.3793, 0.8436, 0.5372, 0.3630, 0.5155, 0.5582, 0.5575],
  2965. [0.5016, 0.3815, 0.8580, 0.4418, 0.3176, 0.5024, 0.6462, 0.5298]],
  2966. device='cuda:0', grad_fn=<AddmmBackward>)
  2967. landmarks are: tensor([[[0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364],
  2968. [0.6058, 0.3978, 0.8288, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461],
  2969. [0.6059, 0.4002, 0.7563, 0.2767, 0.3537, 0.3033, 0.5529, 0.5455],
  2970. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  2971. [0.6203, 0.4020, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  2972. [0.0000, 0.0000, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  2973. [0.6130, 0.4063, 0.8737, 0.5250, 0.4313, 0.4733, 0.5230, 0.5874],
  2974. [0.6263, 0.4065, 0.9038, 0.4317, 0.3587, 0.4550, 0.6325, 0.5250]]],
  2975. device='cuda:0')
  2976. loss_train_step before backward: tensor(0.0060, device='cuda:0', grad_fn=<MseLossBackward>)
  2977. loss_train_step after backward: tensor(0.0060, device='cuda:0', grad_fn=<MseLossBackward>)
  2978. loss_train: 0.9128216054523364
  2979. step: 87
  2980. running loss: 0.010492202361521107
  2981. Train Steps: 87/90 Loss: 0.0105 torch.Size([8, 600, 800])
  2982. torch.Size([8, 8])
  2983. tensor([[0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  2984. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467],
  2985. [0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  2986. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650],
  2987. [0.6210, 0.4164, 0.7202, 0.2930, 0.4025, 0.2483, 0.5687, 0.5567],
  2988. [0.6196, 0.4094, 0.7562, 0.2817, 0.3937, 0.3183, 0.6013, 0.6183],
  2989. [ nan, nan, 0.7725, 0.2611, 0.3675, 0.2733, 0.5413, 0.5167],
  2990. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317]],
  2991. device='cuda:0', dtype=torch.float64)
  2992. predictions are: tensor([[0.6309, 0.4411, 0.8566, 0.5458, 0.3440, 0.4899, 0.6248, 0.5228],
  2993. [0.6560, 0.4673, 0.8795, 0.3399, 0.3029, 0.3756, 0.6669, 0.4935],
  2994. [0.5384, 0.3722, 0.7807, 0.1819, 0.4088, 0.2115, 0.6427, 0.5097],
  2995. [0.4743, 0.3617, 0.8475, 0.3988, 0.3414, 0.3437, 0.5752, 0.5261],
  2996. [0.4291, 0.3059, 0.7685, 0.2349, 0.4137, 0.2631, 0.5626, 0.5433],
  2997. [0.5035, 0.3525, 0.8084, 0.2813, 0.3893, 0.3058, 0.5725, 0.5391],
  2998. [0.3030, 0.2367, 0.7332, 0.2236, 0.3690, 0.2707, 0.5061, 0.5414],
  2999. [0.5988, 0.4033, 0.8804, 0.5317, 0.3392, 0.4191, 0.6056, 0.5135]],
  3000. device='cuda:0', grad_fn=<AddmmBackward>)
  3001. landmarks are: tensor([[[0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  3002. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467],
  3003. [0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  3004. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650],
  3005. [0.6210, 0.4164, 0.7202, 0.2930, 0.4025, 0.2483, 0.5688, 0.5567],
  3006. [0.6196, 0.4094, 0.7563, 0.2817, 0.3938, 0.3183, 0.6012, 0.6183],
  3007. [0.0000, 0.0000, 0.7725, 0.2611, 0.3675, 0.2733, 0.5412, 0.5167],
  3008. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317]]],
  3009. device='cuda:0')
  3010. loss_train_step before backward: tensor(0.0050, device='cuda:0', grad_fn=<MseLossBackward>)
  3011. loss_train_step after backward: tensor(0.0050, device='cuda:0', grad_fn=<MseLossBackward>)
  3012. loss_train: 0.9177949238801375
  3013. step: 88
  3014. running loss: 0.010429487771365199
  3015.  
  3016. Train Steps: 88/90 Loss: 0.0104 torch.Size([8, 600, 800])
  3017. torch.Size([8, 8])
  3018. tensor([[0.6026, 0.3979, 0.8550, 0.4233, 0.3613, 0.5233, 0.5582, 0.4967],
  3019. [0.6202, 0.4066, 0.8746, 0.3376, 0.3717, 0.3090, 0.5842, 0.5165],
  3020. [0.6216, 0.4099, 0.7225, 0.2033, 0.4188, 0.2217, 0.5975, 0.5283],
  3021. [0.6064, 0.4019, 0.8650, 0.4517, 0.4037, 0.5367, 0.5703, 0.5609],
  3022. [0.6200, 0.4059, 0.8700, 0.4900, 0.4163, 0.5000, 0.6162, 0.5467],
  3023. [0.6142, 0.3982, 0.8650, 0.4883, 0.3912, 0.4317, 0.5315, 0.5350],
  3024. [0.6263, 0.4065, 0.9038, 0.4317, 0.3588, 0.4550, 0.6325, 0.5250],
  3025. [0.6203, 0.4078, 0.8800, 0.5083, 0.3900, 0.5000, 0.6100, 0.5583]],
  3026. device='cuda:0', dtype=torch.float64)
  3027. predictions are: tensor([[0.5929, 0.3731, 0.8705, 0.4031, 0.3528, 0.4151, 0.6100, 0.5097],
  3028. [0.4927, 0.3507, 0.8621, 0.2883, 0.3869, 0.2458, 0.5743, 0.5135],
  3029. [0.5013, 0.3601, 0.7256, 0.2112, 0.4145, 0.1674, 0.5532, 0.5421],
  3030. [0.5520, 0.3886, 0.8470, 0.3874, 0.4040, 0.4425, 0.6123, 0.5298],
  3031. [0.6129, 0.4044, 0.8738, 0.4380, 0.4148, 0.4499, 0.6192, 0.5410],
  3032. [0.5775, 0.3733, 0.8324, 0.4880, 0.3656, 0.3719, 0.5710, 0.5103],
  3033. [0.5449, 0.3729, 0.8941, 0.4075, 0.3529, 0.3875, 0.6507, 0.5150],
  3034. [0.5946, 0.3648, 0.8880, 0.5087, 0.3705, 0.4399, 0.5809, 0.5327]],
  3035. device='cuda:0', grad_fn=<AddmmBackward>)
  3036. landmarks are: tensor([[[0.6026, 0.3979, 0.8550, 0.4233, 0.3613, 0.5233, 0.5582, 0.4967],
  3037. [0.6202, 0.4066, 0.8746, 0.3376, 0.3717, 0.3090, 0.5842, 0.5165],
  3038. [0.6216, 0.4099, 0.7225, 0.2033, 0.4187, 0.2217, 0.5975, 0.5283],
  3039. [0.6064, 0.4019, 0.8650, 0.4517, 0.4038, 0.5367, 0.5703, 0.5609],
  3040. [0.6199, 0.4059, 0.8700, 0.4900, 0.4162, 0.5000, 0.6162, 0.5467],
  3041. [0.6143, 0.3982, 0.8650, 0.4883, 0.3913, 0.4317, 0.5315, 0.5350],
  3042. [0.6263, 0.4065, 0.9038, 0.4317, 0.3587, 0.4550, 0.6325, 0.5250],
  3043. [0.6203, 0.4078, 0.8800, 0.5083, 0.3900, 0.5000, 0.6100, 0.5583]]],
  3044. device='cuda:0')
  3045. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  3046. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  3047. loss_train: 0.9196687457151711
  3048. step: 89
  3049. running loss: 0.010333356693428889
  3050. Train Steps: 89/90 Loss: 0.0103 torch.Size([8, 600, 800])
  3051. torch.Size([8, 8])
  3052. tensor([[0.6213, 0.4131, 0.8438, 0.3550, 0.3513, 0.4400, 0.5716, 0.5123],
  3053. [0.6128, 0.4115, 0.7934, 0.3778, 0.3450, 0.4033, 0.5337, 0.5456],
  3054. [0.6212, 0.4171, 0.7875, 0.3633, 0.3813, 0.2933, 0.5675, 0.5700],
  3055. [0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6088, 0.5583],
  3056. [ nan, nan, 0.8938, 0.2850, 0.4662, 0.3117, 0.7406, 0.5528],
  3057. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  3058. [0.6277, 0.4057, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  3059. [0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033]],
  3060. device='cuda:0', dtype=torch.float64)
  3061. predictions are: tensor([[0.5435, 0.3552, 0.8985, 0.4497, 0.3395, 0.4142, 0.5730, 0.5355],
  3062. [0.5902, 0.3901, 0.8323, 0.3892, 0.3376, 0.3772, 0.5743, 0.5076],
  3063. [0.6234, 0.4027, 0.7875, 0.3771, 0.3801, 0.3209, 0.5971, 0.5437],
  3064. [0.5622, 0.3665, 0.7203, 0.2761, 0.3826, 0.2560, 0.5562, 0.5354],
  3065. [0.5153, 0.3298, 0.9352, 0.3171, 0.4653, 0.2956, 0.6762, 0.5267],
  3066. [0.5273, 0.3673, 0.7741, 0.2380, 0.4547, 0.1751, 0.5787, 0.5063],
  3067. [0.6221, 0.4118, 0.8282, 0.2833, 0.4419, 0.2312, 0.6158, 0.5239],
  3068. [0.6492, 0.3897, 0.8705, 0.5609, 0.4380, 0.5152, 0.5819, 0.5223]],
  3069. device='cuda:0', grad_fn=<AddmmBackward>)
  3070. landmarks are: tensor([[[0.6213, 0.4131, 0.8438, 0.3550, 0.3512, 0.4400, 0.5716, 0.5123],
  3071. [0.6128, 0.4115, 0.7934, 0.3778, 0.3450, 0.4033, 0.5337, 0.5456],
  3072. [0.6212, 0.4171, 0.7875, 0.3633, 0.3812, 0.2933, 0.5675, 0.5700],
  3073. [0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6087, 0.5583],
  3074. [0.0000, 0.0000, 0.8938, 0.2850, 0.4663, 0.3117, 0.7406, 0.5528],
  3075. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  3076. [0.6277, 0.4056, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  3077. [0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033]]],
  3078. device='cuda:0')
  3079. loss_train_step before backward: tensor(0.0071, device='cuda:0', grad_fn=<MseLossBackward>)
  3080. loss_train_step after backward: tensor(0.0071, device='cuda:0', grad_fn=<MseLossBackward>)
  3081. loss_train: 0.9267268031835556
  3082. step: 90
  3083. running loss: 0.010296964479817285
  3084. Valid Steps: 10/10 Loss: nan 9.9663
  3085. --------------------------------------------------
  3086. Epoch: 1 Train Loss: 0.0103 Valid Loss: nan
  3087. --------------------------------------------------
  3088. size of train loader is: 90
  3089. torch.Size([8, 600, 800])
  3090. torch.Size([8, 8])
  3091. tensor([[0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  3092. [0.6090, 0.4010, 0.7838, 0.3483, 0.3538, 0.3783, 0.5462, 0.5077],
  3093. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  3094. [0.6299, 0.4303, 0.7963, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  3095. [0.6151, 0.4058, 0.7068, 0.2680, 0.3400, 0.4083, 0.5775, 0.5733],
  3096. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6038, 0.6167],
  3097. [0.6350, 0.4043, 0.8738, 0.5650, 0.3850, 0.4750, 0.6401, 0.4950],
  3098. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389]],
  3099. device='cuda:0', dtype=torch.float64)
  3100. predictions are: tensor([[0.6872, 0.4337, 0.8972, 0.4201, 0.3770, 0.3324, 0.6289, 0.5134],
  3101. [0.6172, 0.3803, 0.8485, 0.3648, 0.3583, 0.3490, 0.5906, 0.5130],
  3102. [0.6515, 0.3853, 0.8185, 0.2864, 0.4441, 0.2439, 0.6147, 0.5116],
  3103. [0.4792, 0.2720, 0.8376, 0.3685, 0.4626, 0.3081, 0.5410, 0.5550],
  3104. [0.5997, 0.3969, 0.7568, 0.2729, 0.3546, 0.3386, 0.5979, 0.5413],
  3105. [0.5900, 0.3652, 0.8555, 0.3659, 0.4145, 0.3966, 0.5892, 0.5485],
  3106. [0.6974, 0.4176, 0.8878, 0.5917, 0.4094, 0.4353, 0.6341, 0.5390],
  3107. [0.4887, 0.3089, 0.8156, 0.2632, 0.4195, 0.2369, 0.5856, 0.5529]],
  3108. device='cuda:0', grad_fn=<AddmmBackward>)
  3109. landmarks are: tensor([[[0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  3110. [0.6090, 0.4010, 0.7837, 0.3483, 0.3537, 0.3783, 0.5462, 0.5077],
  3111. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  3112. [0.6299, 0.4303, 0.7962, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  3113. [0.6151, 0.4058, 0.7068, 0.2680, 0.3400, 0.4083, 0.5775, 0.5733],
  3114. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6037, 0.6167],
  3115. [0.6350, 0.4043, 0.8737, 0.5650, 0.3850, 0.4750, 0.6401, 0.4950],
  3116. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389]]],
  3117. device='cuda:0')
  3118. loss_train_step before backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  3119. loss_train_step after backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  3120. loss_train: 0.002330109942704439
  3121. step: 1
  3122. running loss: 0.002330109942704439
  3123. Train Steps: 1/90 Loss: 0.0023 torch.Size([8, 600, 800])
  3124. torch.Size([8, 8])
  3125. tensor([[0.6109, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  3126. [0.6091, 0.3997, 0.8314, 0.4334, 0.3788, 0.4550, 0.5213, 0.5656],
  3127. [0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153],
  3128. [0.6200, 0.4071, 0.7338, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  3129. [0.6339, 0.4123, 0.8638, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  3130. [0.6245, 0.4115, 0.8700, 0.4883, 0.4625, 0.5517, 0.6100, 0.5217],
  3131. [0.6204, 0.4049, 0.7975, 0.2700, 0.3937, 0.2567, 0.5700, 0.5183],
  3132. [0.6148, 0.3996, 0.8488, 0.3867, 0.3488, 0.4067, 0.5863, 0.5000]],
  3133. device='cuda:0', dtype=torch.float64)
  3134. predictions are: tensor([[0.6145, 0.3738, 0.8607, 0.4208, 0.3780, 0.3628, 0.5549, 0.5305],
  3135. [0.6738, 0.3928, 0.8229, 0.4044, 0.3819, 0.3683, 0.5741, 0.5606],
  3136. [0.6270, 0.3677, 0.8551, 0.4638, 0.4560, 0.4233, 0.5913, 0.5175],
  3137. [0.5668, 0.3573, 0.7547, 0.2084, 0.4347, 0.1876, 0.6145, 0.5623],
  3138. [0.7260, 0.4346, 0.8786, 0.5015, 0.4410, 0.4891, 0.6635, 0.5420],
  3139. [0.6284, 0.3754, 0.8509, 0.4435, 0.4574, 0.4805, 0.6267, 0.5466],
  3140. [0.5074, 0.3131, 0.7886, 0.2444, 0.4170, 0.2448, 0.5512, 0.5679],
  3141. [0.6932, 0.4020, 0.8751, 0.3573, 0.3753, 0.3645, 0.6225, 0.5110]],
  3142. device='cuda:0', grad_fn=<AddmmBackward>)
  3143. landmarks are: tensor([[[0.6108, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  3144. [0.6091, 0.3997, 0.8314, 0.4334, 0.3787, 0.4550, 0.5213, 0.5656],
  3145. [0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153],
  3146. [0.6200, 0.4071, 0.7337, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  3147. [0.6339, 0.4123, 0.8637, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  3148. [0.6245, 0.4115, 0.8700, 0.4883, 0.4625, 0.5517, 0.6100, 0.5217],
  3149. [0.6204, 0.4049, 0.7975, 0.2700, 0.3938, 0.2567, 0.5700, 0.5183],
  3150. [0.6148, 0.3996, 0.8487, 0.3867, 0.3487, 0.4067, 0.5863, 0.5000]]],
  3151. device='cuda:0')
  3152. loss_train_step before backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  3153. loss_train_step after backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  3154. loss_train: 0.004169229883700609
  3155. step: 2
  3156. running loss: 0.0020846149418503046
  3157.  
  3158. Train Steps: 2/90 Loss: 0.0021 torch.Size([8, 600, 800])
  3159. torch.Size([8, 8])
  3160. tensor([[0.6072, 0.4029, 0.7037, 0.2150, 0.3912, 0.2267, 0.5516, 0.5507],
  3161. [0.6264, 0.4248, 0.8938, 0.4183, 0.3875, 0.4100, 0.5400, 0.5350],
  3162. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683],
  3163. [0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166],
  3164. [0.6129, 0.4069, 0.8750, 0.5067, 0.3875, 0.4233, 0.5235, 0.5881],
  3165. [0.6064, 0.4019, 0.8650, 0.4517, 0.4037, 0.5367, 0.5703, 0.5609],
  3166. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  3167. [0.6204, 0.4110, 0.7913, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367]],
  3168. device='cuda:0', dtype=torch.float64)
  3169. predictions are: tensor([[0.6210, 0.3733, 0.7070, 0.1875, 0.4298, 0.2392, 0.6003, 0.5556],
  3170. [0.5637, 0.3689, 0.8830, 0.3828, 0.3889, 0.4043, 0.5980, 0.5772],
  3171. [0.6553, 0.3891, 0.8563, 0.4888, 0.3935, 0.3760, 0.5745, 0.5657],
  3172. [0.7391, 0.4401, 0.8648, 0.4355, 0.4207, 0.4890, 0.6368, 0.5273],
  3173. [0.7080, 0.4109, 0.8534, 0.5227, 0.4033, 0.4343, 0.5789, 0.5379],
  3174. [0.6257, 0.4029, 0.8454, 0.4019, 0.4475, 0.5099, 0.6265, 0.5559],
  3175. [0.5795, 0.3841, 0.8735, 0.4313, 0.3932, 0.4179, 0.6239, 0.5749],
  3176. [0.7210, 0.4312, 0.7837, 0.2215, 0.4313, 0.2648, 0.6441, 0.5431]],
  3177. device='cuda:0', grad_fn=<AddmmBackward>)
  3178. landmarks are: tensor([[[0.6072, 0.4029, 0.7038, 0.2150, 0.3913, 0.2267, 0.5516, 0.5507],
  3179. [0.6264, 0.4248, 0.8938, 0.4183, 0.3875, 0.4100, 0.5400, 0.5350],
  3180. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683],
  3181. [0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166],
  3182. [0.6129, 0.4069, 0.8750, 0.5067, 0.3875, 0.4233, 0.5235, 0.5881],
  3183. [0.6064, 0.4019, 0.8650, 0.4517, 0.4038, 0.5367, 0.5703, 0.5609],
  3184. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  3185. [0.6204, 0.4110, 0.7912, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367]]],
  3186. device='cuda:0')
  3187. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  3188. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  3189. loss_train: 0.005690339137800038
  3190. step: 3
  3191. running loss: 0.0018967797126000125
  3192. Train Steps: 3/90 Loss: 0.0019 torch.Size([8, 600, 800])
  3193. torch.Size([8, 8])
  3194. tensor([[0.6196, 0.4094, 0.7562, 0.2817, 0.3937, 0.3183, 0.6013, 0.6183],
  3195. [0.6264, 0.3972, 0.8853, 0.4771, 0.3853, 0.4511, 0.6293, 0.5334],
  3196. [0.6095, 0.3970, 0.8688, 0.4767, 0.4860, 0.4879, 0.5191, 0.4940],
  3197. [0.6339, 0.4118, 0.7988, 0.5800, 0.3912, 0.4583, 0.7343, 0.5760],
  3198. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  3199. [0.6346, 0.4092, 0.7712, 0.5917, 0.4037, 0.4767, 0.7343, 0.5725],
  3200. [ nan, nan, 0.7268, 0.2333, 0.4125, 0.1933, 0.5112, 0.5383],
  3201. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131]],
  3202. device='cuda:0', dtype=torch.float64)
  3203. predictions are: tensor([[0.6077, 0.3637, 0.8153, 0.3016, 0.4283, 0.3415, 0.6094, 0.5728],
  3204. [0.7280, 0.4462, 0.8907, 0.4716, 0.3838, 0.4796, 0.6378, 0.5437],
  3205. [0.6187, 0.3866, 0.8951, 0.4905, 0.4692, 0.4792, 0.5636, 0.5441],
  3206. [0.6859, 0.4542, 0.8208, 0.4771, 0.3889, 0.4240, 0.6505, 0.5725],
  3207. [0.6441, 0.4278, 0.7551, 0.2768, 0.4004, 0.3053, 0.5613, 0.5257],
  3208. [0.6740, 0.4174, 0.8163, 0.4472, 0.3975, 0.4796, 0.6239, 0.5814],
  3209. [0.4253, 0.2775, 0.7268, 0.2199, 0.4191, 0.2303, 0.5546, 0.5585],
  3210. [0.7390, 0.4758, 0.8779, 0.4258, 0.3620, 0.4260, 0.6183, 0.5347]],
  3211. device='cuda:0', grad_fn=<AddmmBackward>)
  3212. landmarks are: tensor([[[0.6196, 0.4094, 0.7563, 0.2817, 0.3938, 0.3183, 0.6012, 0.6183],
  3213. [0.6264, 0.3972, 0.8853, 0.4771, 0.3853, 0.4511, 0.6293, 0.5334],
  3214. [0.6095, 0.3970, 0.8687, 0.4767, 0.4860, 0.4879, 0.5191, 0.4940],
  3215. [0.6339, 0.4118, 0.7987, 0.5800, 0.3913, 0.4583, 0.7343, 0.5760],
  3216. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  3217. [0.6346, 0.4092, 0.7713, 0.5917, 0.4038, 0.4767, 0.7343, 0.5725],
  3218. [0.0000, 0.0000, 0.7268, 0.2333, 0.4125, 0.1933, 0.5113, 0.5383],
  3219. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131]]],
  3220. device='cuda:0')
  3221. loss_train_step before backward: tensor(0.0060, device='cuda:0', grad_fn=<MseLossBackward>)
  3222. loss_train_step after backward: tensor(0.0060, device='cuda:0', grad_fn=<MseLossBackward>)
  3223. loss_train: 0.011675780289806426
  3224. step: 4
  3225. running loss: 0.0029189450724516064
  3226. Train Steps: 4/90 Loss: 0.0029 torch.Size([8, 600, 800])
  3227. torch.Size([8, 8])
  3228. tensor([[0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  3229. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5787, 0.5550],
  3230. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  3231. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  3232. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  3233. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  3234. [0.6339, 0.4118, 0.7988, 0.5800, 0.3912, 0.4583, 0.7343, 0.5760],
  3235. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533]],
  3236. device='cuda:0', dtype=torch.float64)
  3237. predictions are: tensor([[0.6264, 0.3890, 0.8849, 0.4577, 0.3787, 0.3752, 0.6042, 0.5390],
  3238. [0.6357, 0.4101, 0.8017, 0.3282, 0.3719, 0.4082, 0.6085, 0.5483],
  3239. [0.5796, 0.3818, 0.7151, 0.2136, 0.4244, 0.2336, 0.6140, 0.5432],
  3240. [0.5836, 0.3846, 0.8652, 0.4589, 0.3859, 0.4856, 0.5789, 0.5566],
  3241. [0.4955, 0.3177, 0.7559, 0.3055, 0.3664, 0.3517, 0.5298, 0.5826],
  3242. [0.6555, 0.4345, 0.8346, 0.5506, 0.4640, 0.4997, 0.5712, 0.5504],
  3243. [0.6729, 0.4611, 0.7886, 0.4845, 0.3807, 0.4276, 0.6603, 0.5793],
  3244. [0.6554, 0.4426, 0.8285, 0.4267, 0.4289, 0.4932, 0.5877, 0.5431]],
  3245. device='cuda:0', grad_fn=<AddmmBackward>)
  3246. landmarks are: tensor([[[0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  3247. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5788, 0.5550],
  3248. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  3249. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  3250. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  3251. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  3252. [0.6339, 0.4118, 0.7987, 0.5800, 0.3913, 0.4583, 0.7343, 0.5760],
  3253. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533]]],
  3254. device='cuda:0')
  3255. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  3256. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  3257. loss_train: 0.013066976680420339
  3258. step: 5
  3259. running loss: 0.002613395336084068
  3260. Train Steps: 5/90 Loss: 0.0026 torch.Size([8, 600, 800])
  3261. torch.Size([8, 8])
  3262. tensor([[0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  3263. [0.6299, 0.4008, 0.8450, 0.5350, 0.4213, 0.5000, 0.6350, 0.5100],
  3264. [ nan, nan, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  3265. [0.6034, 0.4011, 0.7350, 0.2533, 0.3438, 0.3367, 0.5516, 0.5084],
  3266. [0.6136, 0.4085, 0.6688, 0.2317, 0.3862, 0.2367, 0.5517, 0.5783],
  3267. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  3268. [0.6205, 0.4062, 0.8337, 0.2683, 0.3675, 0.4283, 0.6338, 0.5250],
  3269. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750]],
  3270. device='cuda:0', dtype=torch.float64)
  3271. predictions are: tensor([[0.5977, 0.4350, 0.8050, 0.4319, 0.3415, 0.4584, 0.5779, 0.5418],
  3272. [0.6764, 0.4721, 0.8811, 0.6016, 0.4422, 0.5708, 0.6735, 0.5176],
  3273. [0.3770, 0.2507, 0.8024, 0.3486, 0.4141, 0.3534, 0.5272, 0.5591],
  3274. [0.6396, 0.4189, 0.7438, 0.3510, 0.3652, 0.4136, 0.6224, 0.5179],
  3275. [0.5993, 0.3926, 0.7307, 0.3044, 0.3918, 0.3173, 0.5537, 0.5438],
  3276. [0.5945, 0.3974, 0.8848, 0.5304, 0.3780, 0.4062, 0.5683, 0.5607],
  3277. [0.6597, 0.4484, 0.8276, 0.3433, 0.3842, 0.4623, 0.6476, 0.5271],
  3278. [0.6816, 0.4689, 0.7731, 0.3694, 0.3935, 0.3450, 0.6229, 0.5515]],
  3279. device='cuda:0', grad_fn=<AddmmBackward>)
  3280. landmarks are: tensor([[[0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  3281. [0.6299, 0.4008, 0.8450, 0.5350, 0.4212, 0.5000, 0.6350, 0.5100],
  3282. [0.0000, 0.0000, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  3283. [0.6033, 0.4011, 0.7350, 0.2533, 0.3438, 0.3367, 0.5516, 0.5084],
  3284. [0.6136, 0.4085, 0.6687, 0.2317, 0.3862, 0.2367, 0.5517, 0.5783],
  3285. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  3286. [0.6205, 0.4062, 0.8338, 0.2683, 0.3675, 0.4283, 0.6338, 0.5250],
  3287. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750]]],
  3288. device='cuda:0')
  3289. loss_train_step before backward: tensor(0.0054, device='cuda:0', grad_fn=<MseLossBackward>)
  3290. loss_train_step after backward: tensor(0.0054, device='cuda:0', grad_fn=<MseLossBackward>)
  3291.  
  3292. loss_train: 0.018506109598092735
  3293. step: 6
  3294. running loss: 0.0030843515996821225
  3295. Train Steps: 6/90 Loss: 0.0031 torch.Size([8, 600, 800])
  3296. torch.Size([8, 8])
  3297. tensor([[0.6128, 0.4115, 0.7934, 0.3778, 0.3450, 0.4033, 0.5337, 0.5456],
  3298. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  3299. [0.6200, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  3300. [0.6177, 0.4086, 0.8738, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700],
  3301. [0.6040, 0.4002, 0.7338, 0.2267, 0.3975, 0.2100, 0.5231, 0.4778],
  3302. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  3303. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  3304. [0.6161, 0.4024, 0.8838, 0.4583, 0.3688, 0.3733, 0.5311, 0.5344]],
  3305. device='cuda:0', dtype=torch.float64)
  3306. predictions are: tensor([[0.5708, 0.3968, 0.7693, 0.4015, 0.3199, 0.4287, 0.5562, 0.5247],
  3307. [0.5546, 0.3773, 0.8702, 0.5485, 0.3906, 0.5924, 0.6215, 0.5399],
  3308. [0.6222, 0.4316, 0.7642, 0.3521, 0.3556, 0.3273, 0.5974, 0.5201],
  3309. [0.6468, 0.4305, 0.8375, 0.4463, 0.3929, 0.5610, 0.6378, 0.5235],
  3310. [0.5255, 0.3410, 0.6833, 0.2793, 0.3942, 0.2555, 0.5385, 0.5084],
  3311. [0.5840, 0.3774, 0.7915, 0.2801, 0.4503, 0.2402, 0.6608, 0.5368],
  3312. [0.5709, 0.4004, 0.8241, 0.6080, 0.4465, 0.5133, 0.5526, 0.5187],
  3313. [0.5726, 0.3613, 0.8635, 0.5536, 0.3483, 0.4468, 0.5813, 0.5152]],
  3314. device='cuda:0', grad_fn=<AddmmBackward>)
  3315. landmarks are: tensor([[[0.6128, 0.4115, 0.7934, 0.3778, 0.3450, 0.4033, 0.5337, 0.5456],
  3316. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  3317. [0.6201, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  3318. [0.6177, 0.4085, 0.8737, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700],
  3319. [0.6040, 0.4002, 0.7337, 0.2267, 0.3975, 0.2100, 0.5231, 0.4778],
  3320. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  3321. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  3322. [0.6161, 0.4024, 0.8838, 0.4583, 0.3688, 0.3733, 0.5311, 0.5344]]],
  3323. device='cuda:0')
  3324. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  3325. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  3326. loss_train: 0.01996891770977527
  3327. step: 7
  3328. running loss: 0.0028527025299678955
  3329. Train Steps: 7/90 Loss: 0.0029 torch.Size([8, 600, 800])
  3330. torch.Size([8, 8])
  3331. tensor([[0.6205, 0.4016, 0.8350, 0.2717, 0.3987, 0.2550, 0.5787, 0.5133],
  3332. [0.6055, 0.4015, 0.7425, 0.2033, 0.4113, 0.1883, 0.5217, 0.4823],
  3333. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  3334. [0.6273, 0.4110, 0.8900, 0.3817, 0.4188, 0.2167, 0.5858, 0.4835],
  3335. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  3336. [0.6200, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913],
  3337. [0.6273, 0.4105, 0.8988, 0.4517, 0.3912, 0.2550, 0.5894, 0.4811],
  3338. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795]],
  3339. device='cuda:0', dtype=torch.float64)
  3340. predictions are: tensor([[0.5045, 0.3187, 0.7871, 0.2918, 0.3870, 0.3102, 0.5536, 0.5498],
  3341. [0.4373, 0.3065, 0.6487, 0.2241, 0.3569, 0.2030, 0.5153, 0.5458],
  3342. [0.6749, 0.4379, 0.8354, 0.4962, 0.3637, 0.5110, 0.5942, 0.4977],
  3343. [0.5190, 0.3392, 0.8178, 0.3587, 0.4088, 0.2443, 0.5731, 0.5093],
  3344. [0.5491, 0.3799, 0.8266, 0.4311, 0.3344, 0.3934, 0.5794, 0.5571],
  3345. [0.6324, 0.3970, 0.8339, 0.5412, 0.4179, 0.5375, 0.5340, 0.5230],
  3346. [0.5478, 0.3553, 0.8102, 0.4492, 0.3866, 0.3094, 0.5413, 0.5078],
  3347. [0.6430, 0.4296, 0.8219, 0.5368, 0.3574, 0.5627, 0.6212, 0.5529]],
  3348. device='cuda:0', grad_fn=<AddmmBackward>)
  3349. landmarks are: tensor([[[0.6205, 0.4015, 0.8350, 0.2717, 0.3988, 0.2550, 0.5788, 0.5133],
  3350. [0.6055, 0.4015, 0.7425, 0.2033, 0.4112, 0.1883, 0.5217, 0.4823],
  3351. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  3352. [0.6273, 0.4110, 0.8900, 0.3817, 0.4187, 0.2167, 0.5858, 0.4835],
  3353. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  3354. [0.6199, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913],
  3355. [0.6273, 0.4105, 0.8988, 0.4517, 0.3913, 0.2550, 0.5894, 0.4811],
  3356. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795]]],
  3357. device='cuda:0')
  3358. loss_train_step before backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  3359. loss_train_step after backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  3360. loss_train: 0.022683476679958403
  3361. step: 8
  3362. running loss: 0.0028354345849948004
  3363. Train Steps: 8/90 Loss: 0.0028 torch.Size([8, 600, 800])
  3364. torch.Size([8, 8])
  3365. tensor([[0.6148, 0.4076, 0.8666, 0.4820, 0.4138, 0.5067, 0.5250, 0.5767],
  3366. [0.6142, 0.4127, 0.7575, 0.3067, 0.3438, 0.4383, 0.5778, 0.5207],
  3367. [0.6161, 0.4024, 0.8662, 0.4683, 0.4935, 0.5364, 0.6063, 0.5567],
  3368. [0.6197, 0.4050, 0.7527, 0.2000, 0.4042, 0.2249, 0.5895, 0.4995],
  3369. [0.6202, 0.4066, 0.8746, 0.3376, 0.3717, 0.3090, 0.5842, 0.5165],
  3370. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317],
  3371. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373],
  3372. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540]],
  3373. device='cuda:0', dtype=torch.float64)
  3374. predictions are: tensor([[0.6031, 0.4043, 0.8679, 0.4927, 0.4254, 0.4905, 0.5511, 0.5095],
  3375. [0.5522, 0.3762, 0.7357, 0.2780, 0.3510, 0.3617, 0.5283, 0.5241],
  3376. [0.5653, 0.3769, 0.8385, 0.4342, 0.4193, 0.4412, 0.5588, 0.5099],
  3377. [0.5708, 0.3708, 0.7378, 0.2137, 0.3824, 0.2108, 0.5804, 0.4831],
  3378. [0.5330, 0.3495, 0.8532, 0.3130, 0.3553, 0.2720, 0.5710, 0.4873],
  3379. [0.6152, 0.3664, 0.8556, 0.5128, 0.3753, 0.3823, 0.5417, 0.4911],
  3380. [0.6507, 0.4053, 0.8211, 0.5239, 0.3740, 0.4510, 0.5287, 0.5169],
  3381. [0.5049, 0.3249, 0.8977, 0.4204, 0.3946, 0.3056, 0.6048, 0.5078]],
  3382. device='cuda:0', grad_fn=<AddmmBackward>)
  3383. landmarks are: tensor([[[0.6148, 0.4076, 0.8666, 0.4820, 0.4137, 0.5067, 0.5250, 0.5767],
  3384. [0.6142, 0.4127, 0.7575, 0.3067, 0.3438, 0.4383, 0.5778, 0.5207],
  3385. [0.6161, 0.4024, 0.8662, 0.4683, 0.4935, 0.5364, 0.6062, 0.5567],
  3386. [0.6197, 0.4050, 0.7527, 0.2000, 0.4042, 0.2249, 0.5895, 0.4995],
  3387. [0.6202, 0.4066, 0.8746, 0.3376, 0.3717, 0.3090, 0.5842, 0.5165],
  3388. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317],
  3389. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373],
  3390. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540]]],
  3391. device='cuda:0')
  3392. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  3393. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  3394. loss_train: 0.02442885807249695
  3395. step: 9
  3396. running loss: 0.002714317563610772
  3397. Train Steps: 9/90 Loss: 0.0027 torch.Size([8, 600, 800])
  3398. torch.Size([8, 8])
  3399. tensor([[0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  3400. [ nan, nan, 0.7525, 0.2291, 0.3838, 0.3017, 0.6050, 0.5667],
  3401. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  3402. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  3403. [0.6198, 0.4076, 0.7945, 0.2453, 0.4475, 0.1871, 0.5897, 0.5350],
  3404. [0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6262, 0.5367],
  3405. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  3406. [0.6117, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103]],
  3407. device='cuda:0', dtype=torch.float64)
  3408. predictions are: tensor([[0.6161, 0.3976, 0.8913, 0.5049, 0.4627, 0.4538, 0.5329, 0.5048],
  3409. [0.4613, 0.3151, 0.7858, 0.2366, 0.3750, 0.2504, 0.5672, 0.5301],
  3410. [0.6097, 0.4158, 0.8656, 0.5438, 0.3685, 0.4619, 0.5735, 0.4866],
  3411. [0.5733, 0.3802, 0.7500, 0.2107, 0.4054, 0.1448, 0.5669, 0.5125],
  3412. [0.5220, 0.3303, 0.7986, 0.2464, 0.3953, 0.2086, 0.5738, 0.5130],
  3413. [0.5901, 0.4047, 0.9199, 0.4660, 0.4396, 0.5506, 0.5753, 0.4923],
  3414. [0.6348, 0.4031, 0.9029, 0.5513, 0.3419, 0.3958, 0.5533, 0.5214],
  3415. [0.5432, 0.3565, 0.7056, 0.2154, 0.3556, 0.2192, 0.5092, 0.5026]],
  3416. device='cuda:0', grad_fn=<AddmmBackward>)
  3417. landmarks are: tensor([[[0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  3418. [0.0000, 0.0000, 0.7525, 0.2291, 0.3837, 0.3017, 0.6050, 0.5667],
  3419. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  3420. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  3421. [0.6198, 0.4076, 0.7945, 0.2453, 0.4475, 0.1871, 0.5897, 0.5350],
  3422. [0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6263, 0.5367],
  3423. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  3424. [0.6116, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103]]],
  3425. device='cuda:0')
  3426. loss_train_step before backward: tensor(0.0062, device='cuda:0', grad_fn=<MseLossBackward>)
  3427. loss_train_step after backward: tensor(0.0062, device='cuda:0', grad_fn=<MseLossBackward>)
  3428. loss_train: 0.030593233765102923
  3429. step: 10
  3430. running loss: 0.003059323376510292
  3431.  
  3432. Train Steps: 10/90 Loss: 0.0031 torch.Size([8, 600, 800])
  3433. torch.Size([8, 8])
  3434. tensor([[0.6267, 0.4065, 0.8313, 0.2467, 0.4788, 0.1733, 0.6312, 0.5133],
  3435. [0.6248, 0.4032, 0.7738, 0.1900, 0.4813, 0.1400, 0.5941, 0.4904],
  3436. [0.6219, 0.4114, 0.8175, 0.2817, 0.3925, 0.2783, 0.5900, 0.5350],
  3437. [0.6311, 0.3998, 0.7975, 0.5767, 0.3838, 0.4850, 0.7327, 0.5343],
  3438. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  3439. [0.6189, 0.4049, 0.8888, 0.4417, 0.4213, 0.5200, 0.5988, 0.5633],
  3440. [0.6236, 0.4084, 0.7738, 0.2133, 0.3663, 0.3233, 0.5813, 0.5567],
  3441. [0.6124, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485]],
  3442. device='cuda:0', dtype=torch.float64)
  3443. predictions are: tensor([[0.6524, 0.4194, 0.8302, 0.2453, 0.4642, 0.2088, 0.6127, 0.5133],
  3444. [0.5333, 0.3385, 0.7581, 0.2202, 0.4625, 0.1170, 0.5673, 0.4804],
  3445. [0.4795, 0.3257, 0.8051, 0.2669, 0.3902, 0.2321, 0.5726, 0.5239],
  3446. [0.5878, 0.4107, 0.8545, 0.5105, 0.3563, 0.4354, 0.5768, 0.5078],
  3447. [0.4419, 0.3261, 0.7948, 0.2944, 0.3893, 0.2562, 0.5276, 0.5387],
  3448. [0.6046, 0.4057, 0.9291, 0.5008, 0.4122, 0.5362, 0.5740, 0.5424],
  3449. [0.5551, 0.3804, 0.8104, 0.3170, 0.3766, 0.2811, 0.5634, 0.5270],
  3450. [0.6001, 0.3880, 0.7420, 0.3093, 0.3546, 0.2835, 0.4877, 0.5316]],
  3451. device='cuda:0', grad_fn=<AddmmBackward>)
  3452. landmarks are: tensor([[[0.6266, 0.4065, 0.8313, 0.2467, 0.4787, 0.1733, 0.6313, 0.5133],
  3453. [0.6248, 0.4032, 0.7738, 0.1900, 0.4812, 0.1400, 0.5941, 0.4904],
  3454. [0.6219, 0.4114, 0.8175, 0.2817, 0.3925, 0.2783, 0.5900, 0.5350],
  3455. [0.6311, 0.3998, 0.7975, 0.5767, 0.3837, 0.4850, 0.7327, 0.5343],
  3456. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  3457. [0.6189, 0.4049, 0.8888, 0.4417, 0.4212, 0.5200, 0.5987, 0.5633],
  3458. [0.6236, 0.4084, 0.7738, 0.2133, 0.3663, 0.3233, 0.5813, 0.5567],
  3459. [0.6123, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485]]],
  3460. device='cuda:0')
  3461. loss_train_step before backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  3462. loss_train_step after backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  3463. loss_train: 0.033024250413291156
  3464. step: 11
  3465. running loss: 0.0030022045830264688
  3466. Train Steps: 11/90 Loss: 0.0030 torch.Size([8, 600, 800])
  3467. torch.Size([8, 8])
  3468. tensor([[0.6336, 0.4154, 0.8900, 0.2767, 0.4988, 0.2867, 0.7422, 0.5540],
  3469. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  3470. [0.6296, 0.4076, 0.8400, 0.5583, 0.3700, 0.4367, 0.6876, 0.5494],
  3471. [0.6266, 0.4070, 0.8712, 0.5600, 0.3713, 0.4783, 0.5775, 0.6100],
  3472. [0.6325, 0.4165, 0.9000, 0.4617, 0.3813, 0.4900, 0.7485, 0.5447],
  3473. [0.6198, 0.4130, 0.8762, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  3474. [0.6250, 0.4236, 0.8638, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  3475. [0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633]],
  3476. device='cuda:0', dtype=torch.float64)
  3477. predictions are: tensor([[0.5612, 0.3811, 0.8602, 0.2296, 0.4916, 0.2306, 0.6434, 0.5336],
  3478. [0.6425, 0.4218, 0.8356, 0.3556, 0.4207, 0.3121, 0.6238, 0.5244],
  3479. [0.6146, 0.4007, 0.8371, 0.4343, 0.4082, 0.3477, 0.6174, 0.5276],
  3480. [0.6623, 0.4279, 0.8380, 0.4641, 0.4136, 0.4596, 0.5665, 0.5384],
  3481. [0.5593, 0.3786, 0.9084, 0.4347, 0.4017, 0.4581, 0.6508, 0.5335],
  3482. [0.6136, 0.4184, 0.9224, 0.3770, 0.3960, 0.4299, 0.5676, 0.5565],
  3483. [0.5552, 0.3466, 0.8568, 0.3470, 0.4307, 0.2913, 0.5624, 0.5741],
  3484. [0.6698, 0.4377, 0.8489, 0.4851, 0.4061, 0.4395, 0.5585, 0.5461]],
  3485. device='cuda:0', grad_fn=<AddmmBackward>)
  3486. landmarks are: tensor([[[0.6336, 0.4154, 0.8900, 0.2767, 0.4988, 0.2867, 0.7422, 0.5540],
  3487. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  3488. [0.6296, 0.4076, 0.8400, 0.5583, 0.3700, 0.4367, 0.6876, 0.5494],
  3489. [0.6266, 0.4070, 0.8712, 0.5600, 0.3713, 0.4783, 0.5775, 0.6100],
  3490. [0.6325, 0.4165, 0.9000, 0.4617, 0.3812, 0.4900, 0.7485, 0.5447],
  3491. [0.6198, 0.4130, 0.8763, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  3492. [0.6250, 0.4236, 0.8637, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  3493. [0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633]]],
  3494. device='cuda:0')
  3495. loss_train_step before backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  3496. loss_train_step after backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  3497. loss_train: 0.03542143327649683
  3498. step: 12
  3499. running loss: 0.0029517861063747355
  3500. Train Steps: 12/90 Loss: 0.0030 torch.Size([8, 600, 800])
  3501. torch.Size([8, 8])
  3502. tensor([[0.6102, 0.4020, 0.8638, 0.3717, 0.3625, 0.5017, 0.6038, 0.5500],
  3503. [0.6207, 0.4081, 0.7662, 0.2067, 0.3962, 0.3200, 0.6312, 0.5300],
  3504. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  3505. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385],
  3506. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  3507. [ nan, nan, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600],
  3508. [0.6203, 0.4073, 0.8189, 0.2398, 0.4400, 0.2054, 0.5929, 0.5501],
  3509. [0.6204, 0.4049, 0.7975, 0.2700, 0.3937, 0.2567, 0.5700, 0.5183]],
  3510. device='cuda:0', dtype=torch.float64)
  3511. predictions are: tensor([[0.5890, 0.4046, 0.8961, 0.4043, 0.4132, 0.5221, 0.6450, 0.5752],
  3512. [0.6591, 0.4443, 0.7609, 0.2467, 0.4018, 0.2730, 0.6092, 0.5504],
  3513. [0.6208, 0.4260, 0.8892, 0.4191, 0.3758, 0.3963, 0.5760, 0.5460],
  3514. [0.6432, 0.4296, 0.8513, 0.2810, 0.4636, 0.2191, 0.6936, 0.5478],
  3515. [0.6852, 0.4710, 0.8878, 0.5743, 0.4098, 0.4560, 0.5968, 0.5572],
  3516. [0.3552, 0.2292, 0.8169, 0.2640, 0.4351, 0.2463, 0.5953, 0.5789],
  3517. [0.6208, 0.4168, 0.8051, 0.2661, 0.4353, 0.2119, 0.6514, 0.5384],
  3518. [0.5933, 0.4019, 0.8066, 0.2821, 0.4158, 0.2314, 0.5887, 0.5460]],
  3519. device='cuda:0', grad_fn=<AddmmBackward>)
  3520. landmarks are: tensor([[[0.6102, 0.4020, 0.8637, 0.3717, 0.3625, 0.5017, 0.6037, 0.5500],
  3521. [0.6207, 0.4081, 0.7663, 0.2067, 0.3963, 0.3200, 0.6313, 0.5300],
  3522. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  3523. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385],
  3524. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  3525. [0.0000, 0.0000, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600],
  3526. [0.6203, 0.4073, 0.8189, 0.2398, 0.4400, 0.2054, 0.5929, 0.5501],
  3527. [0.6204, 0.4049, 0.7975, 0.2700, 0.3938, 0.2567, 0.5700, 0.5183]]],
  3528. device='cuda:0')
  3529. loss_train_step before backward: tensor(0.0035, device='cuda:0', grad_fn=<MseLossBackward>)
  3530. loss_train_step after backward: tensor(0.0035, device='cuda:0', grad_fn=<MseLossBackward>)
  3531. loss_train: 0.038919162820093334
  3532. step: 13
  3533. running loss: 0.0029937817553917947
  3534. Train Steps: 13/90 Loss: 0.0030 torch.Size([8, 600, 800])
  3535. torch.Size([8, 8])
  3536. tensor([[0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  3537. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  3538. [0.6189, 0.4029, 0.8375, 0.5767, 0.4745, 0.4829, 0.5551, 0.5598],
  3539. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6038, 0.4833],
  3540. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  3541. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6063, 0.5617],
  3542. [0.6216, 0.4167, 0.8588, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667],
  3543. [0.6250, 0.4054, 0.8770, 0.4723, 0.4662, 0.5367, 0.6162, 0.5433]],
  3544. device='cuda:0', dtype=torch.float64)
  3545. predictions are: tensor([[0.6692, 0.4524, 0.8693, 0.4409, 0.3938, 0.4273, 0.6492, 0.5632],
  3546. [0.5813, 0.3906, 0.8782, 0.1791, 0.4501, 0.1834, 0.7445, 0.5954],
  3547. [0.6371, 0.4335, 0.8636, 0.4553, 0.4532, 0.3713, 0.6627, 0.5859],
  3548. [0.6131, 0.4125, 0.8602, 0.4216, 0.3713, 0.4308, 0.6201, 0.5651],
  3549. [0.5784, 0.3985, 0.8561, 0.4410, 0.4433, 0.4864, 0.6356, 0.5596],
  3550. [0.6493, 0.4302, 0.8929, 0.4087, 0.3751, 0.4351, 0.6520, 0.5835],
  3551. [0.6617, 0.4436, 0.8709, 0.4943, 0.4211, 0.4950, 0.6436, 0.5883],
  3552. [0.5980, 0.4068, 0.8920, 0.3774, 0.4322, 0.4469, 0.6501, 0.5687]],
  3553. device='cuda:0', grad_fn=<AddmmBackward>)
  3554. landmarks are: tensor([[[0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  3555. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  3556. [0.6189, 0.4029, 0.8375, 0.5767, 0.4745, 0.4829, 0.5551, 0.5598],
  3557. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6037, 0.4833],
  3558. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  3559. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6062, 0.5617],
  3560. [0.6216, 0.4167, 0.8587, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667],
  3561. [0.6250, 0.4054, 0.8770, 0.4723, 0.4663, 0.5367, 0.6162, 0.5433]]],
  3562. device='cuda:0')
  3563. loss_train_step before backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  3564. loss_train_step after backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  3565. loss_train: 0.04132039553951472
  3566. step: 14
  3567. running loss: 0.0029514568242510514
  3568.  
  3569. Train Steps: 14/90 Loss: 0.0030 torch.Size([8, 600, 800])
  3570. torch.Size([8, 8])
  3571. tensor([[0.6081, 0.3950, 0.8538, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954],
  3572. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  3573. [0.6118, 0.4052, 0.8463, 0.3917, 0.3538, 0.3450, 0.5053, 0.5593],
  3574. [0.6339, 0.4102, 0.8588, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089],
  3575. [0.6198, 0.3997, 0.8582, 0.5361, 0.4117, 0.5016, 0.5942, 0.5134],
  3576. [ nan, nan, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  3577. [0.6275, 0.4050, 0.9038, 0.3767, 0.3838, 0.3533, 0.7074, 0.5575],
  3578. [0.6200, 0.3978, 0.8900, 0.4550, 0.3775, 0.5200, 0.6150, 0.5367]],
  3579. device='cuda:0', dtype=torch.float64)
  3580. predictions are: tensor([[0.5942, 0.3891, 0.8576, 0.4474, 0.4046, 0.4978, 0.5956, 0.5803],
  3581. [0.6719, 0.4472, 0.8531, 0.3704, 0.3951, 0.3693, 0.6507, 0.5522],
  3582. [0.6160, 0.4064, 0.8342, 0.3875, 0.3559, 0.3626, 0.5747, 0.6071],
  3583. [0.7079, 0.4728, 0.8355, 0.2891, 0.4455, 0.2665, 0.7089, 0.5726],
  3584. [0.6223, 0.4199, 0.8382, 0.5415, 0.4098, 0.5251, 0.6462, 0.5646],
  3585. [0.3999, 0.2640, 0.8994, 0.3008, 0.4934, 0.2666, 0.7216, 0.5950],
  3586. [0.6839, 0.4563, 0.8921, 0.3541, 0.4058, 0.3401, 0.7596, 0.5681],
  3587. [0.6237, 0.4080, 0.8588, 0.4664, 0.3939, 0.5808, 0.6382, 0.6014]],
  3588. device='cuda:0', grad_fn=<AddmmBackward>)
  3589. landmarks are: tensor([[[0.6081, 0.3950, 0.8537, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954],
  3590. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  3591. [0.6118, 0.4052, 0.8462, 0.3917, 0.3537, 0.3450, 0.5053, 0.5593],
  3592. [0.6339, 0.4102, 0.8587, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089],
  3593. [0.6198, 0.3997, 0.8582, 0.5361, 0.4117, 0.5016, 0.5942, 0.5134],
  3594. [0.0000, 0.0000, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  3595. [0.6275, 0.4050, 0.9038, 0.3767, 0.3837, 0.3533, 0.7074, 0.5575],
  3596. [0.6199, 0.3978, 0.8900, 0.4550, 0.3775, 0.5200, 0.6150, 0.5367]]],
  3597. device='cuda:0')
  3598. loss_train_step before backward: tensor(0.0049, device='cuda:0', grad_fn=<MseLossBackward>)
  3599. loss_train_step after backward: tensor(0.0049, device='cuda:0', grad_fn=<MseLossBackward>)
  3600. loss_train: 0.046259972150437534
  3601. step: 15
  3602. running loss: 0.003083998143362502
  3603. Train Steps: 15/90 Loss: 0.0031 torch.Size([8, 600, 800])
  3604. torch.Size([8, 8])
  3605. tensor([[0.6236, 0.4084, 0.7738, 0.2133, 0.3663, 0.3233, 0.5813, 0.5567],
  3606. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  3607. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  3608. [0.6200, 0.3993, 0.8519, 0.4923, 0.3962, 0.4717, 0.6013, 0.5433],
  3609. [0.6228, 0.4004, 0.8750, 0.5250, 0.3825, 0.5233, 0.6362, 0.5000],
  3610. [ nan, nan, 0.6793, 0.2110, 0.4012, 0.2167, 0.5112, 0.5583],
  3611. [0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  3612. [ nan, nan, 0.7725, 0.2611, 0.3675, 0.2733, 0.5413, 0.5167]],
  3613. device='cuda:0', dtype=torch.float64)
  3614. predictions are: tensor([[0.6457, 0.4295, 0.8061, 0.3196, 0.3867, 0.3391, 0.6472, 0.5647],
  3615. [0.6592, 0.4132, 0.8562, 0.3014, 0.4840, 0.2955, 0.7247, 0.5346],
  3616. [0.6431, 0.4059, 0.9184, 0.3719, 0.3832, 0.3503, 0.6498, 0.5410],
  3617. [0.6882, 0.4392, 0.8831, 0.5727, 0.3650, 0.5405, 0.6501, 0.5698],
  3618. [0.6855, 0.4245, 0.9158, 0.5830, 0.3851, 0.6142, 0.6524, 0.5226],
  3619. [0.3411, 0.2191, 0.7176, 0.2231, 0.4243, 0.2191, 0.5480, 0.5725],
  3620. [0.6789, 0.4679, 0.8847, 0.6112, 0.3784, 0.5188, 0.6684, 0.5326],
  3621. [0.2940, 0.1794, 0.7829, 0.2765, 0.3914, 0.2806, 0.5706, 0.5670]],
  3622. device='cuda:0', grad_fn=<AddmmBackward>)
  3623. landmarks are: tensor([[[0.6236, 0.4084, 0.7738, 0.2133, 0.3663, 0.3233, 0.5813, 0.5567],
  3624. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  3625. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  3626. [0.6200, 0.3993, 0.8519, 0.4923, 0.3963, 0.4717, 0.6012, 0.5433],
  3627. [0.6228, 0.4004, 0.8750, 0.5250, 0.3825, 0.5233, 0.6363, 0.5000],
  3628. [0.0000, 0.0000, 0.6793, 0.2110, 0.4013, 0.2167, 0.5113, 0.5583],
  3629. [0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  3630. [0.0000, 0.0000, 0.7725, 0.2611, 0.3675, 0.2733, 0.5412, 0.5167]]],
  3631. device='cuda:0')
  3632. loss_train_step before backward: tensor(0.0061, device='cuda:0', grad_fn=<MseLossBackward>)
  3633. loss_train_step after backward: tensor(0.0061, device='cuda:0', grad_fn=<MseLossBackward>)
  3634. loss_train: 0.05236703611444682
  3635. step: 16
  3636. running loss: 0.003272939757152926
  3637. Train Steps: 16/90 Loss: 0.0033 torch.Size([8, 600, 800])
  3638. torch.Size([8, 8])
  3639. tensor([[0.6163, 0.4114, 0.7650, 0.2017, 0.3763, 0.2867, 0.5631, 0.5071],
  3640. [0.6202, 0.4064, 0.7879, 0.2179, 0.4567, 0.1725, 0.5955, 0.5478],
  3641. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  3642. [0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050],
  3643. [0.6127, 0.4115, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  3644. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  3645. [0.6228, 0.4119, 0.7938, 0.2233, 0.4674, 0.1773, 0.6188, 0.5433],
  3646. [0.6055, 0.4015, 0.7425, 0.2033, 0.4113, 0.1883, 0.5217, 0.4823]],
  3647. device='cuda:0', dtype=torch.float64)
  3648. predictions are: tensor([[0.5899, 0.3427, 0.7989, 0.3637, 0.3790, 0.3644, 0.6021, 0.5073],
  3649. [0.5794, 0.3500, 0.7829, 0.3558, 0.4715, 0.2606, 0.6293, 0.5235],
  3650. [0.5524, 0.3206, 0.8937, 0.4374, 0.4395, 0.4080, 0.6941, 0.5116],
  3651. [0.5741, 0.3633, 0.9493, 0.5738, 0.3724, 0.6298, 0.6568, 0.5525],
  3652. [0.5215, 0.3038, 0.7782, 0.3703, 0.3806, 0.3965, 0.5499, 0.5504],
  3653. [0.6112, 0.3637, 0.8144, 0.3356, 0.3792, 0.4078, 0.6076, 0.5224],
  3654. [0.5639, 0.3496, 0.8526, 0.3172, 0.4734, 0.2581, 0.6480, 0.5238],
  3655. [0.4399, 0.2583, 0.7541, 0.2975, 0.3986, 0.2429, 0.5493, 0.5150]],
  3656. device='cuda:0', grad_fn=<AddmmBackward>)
  3657. landmarks are: tensor([[[0.6163, 0.4114, 0.7650, 0.2017, 0.3762, 0.2867, 0.5631, 0.5071],
  3658. [0.6202, 0.4064, 0.7879, 0.2179, 0.4567, 0.1725, 0.5955, 0.5478],
  3659. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  3660. [0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050],
  3661. [0.6127, 0.4114, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  3662. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  3663. [0.6228, 0.4119, 0.7937, 0.2233, 0.4674, 0.1773, 0.6187, 0.5433],
  3664. [0.6055, 0.4015, 0.7425, 0.2033, 0.4112, 0.1883, 0.5217, 0.4823]]],
  3665. device='cuda:0')
  3666. loss_train_step before backward: tensor(0.0050, device='cuda:0', grad_fn=<MseLossBackward>)
  3667. loss_train_step after backward: tensor(0.0050, device='cuda:0', grad_fn=<MseLossBackward>)
  3668. loss_train: 0.057400591555051506
  3669. step: 17
  3670. running loss: 0.003376505385591265
  3671. Train Steps: 17/90 Loss: 0.0034 torch.Size([8, 600, 800])
  3672. torch.Size([8, 8])
  3673. tensor([[0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950],
  3674. [0.6149, 0.4054, 0.6713, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  3675. [0.6131, 0.4037, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  3676. [ nan, nan, 0.6469, 0.1943, 0.4025, 0.2000, 0.5125, 0.5533],
  3677. [0.6109, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  3678. [0.6239, 0.4174, 0.8425, 0.5733, 0.4825, 0.4500, 0.5625, 0.5933],
  3679. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167],
  3680. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320]],
  3681. device='cuda:0', dtype=torch.float64)
  3682. predictions are: tensor([[0.5694, 0.3072, 0.9180, 0.5808, 0.4260, 0.5578, 0.5957, 0.4780],
  3683. [0.4657, 0.2800, 0.7147, 0.2770, 0.4011, 0.2651, 0.5142, 0.5039],
  3684. [0.5343, 0.2985, 0.7305, 0.2937, 0.3831, 0.3645, 0.5357, 0.5165],
  3685. [0.3305, 0.1637, 0.7213, 0.2061, 0.4155, 0.2128, 0.5293, 0.5100],
  3686. [0.5894, 0.3573, 0.9104, 0.5207, 0.3549, 0.4583, 0.5722, 0.4644],
  3687. [0.6676, 0.4207, 0.8576, 0.6183, 0.4383, 0.4724, 0.6427, 0.5154],
  3688. [0.5665, 0.3570, 0.8540, 0.3346, 0.3749, 0.3530, 0.5816, 0.4958],
  3689. [0.5989, 0.3703, 0.8312, 0.2364, 0.4190, 0.3152, 0.6319, 0.4938]],
  3690. device='cuda:0', grad_fn=<AddmmBackward>)
  3691. landmarks are: tensor([[[0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950],
  3692. [0.6149, 0.4054, 0.6712, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  3693. [0.6131, 0.4036, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  3694. [0.0000, 0.0000, 0.6469, 0.1943, 0.4025, 0.2000, 0.5125, 0.5533],
  3695. [0.6108, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  3696. [0.6239, 0.4174, 0.8425, 0.5733, 0.4825, 0.4500, 0.5625, 0.5933],
  3697. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167],
  3698. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320]]],
  3699. device='cuda:0')
  3700. loss_train_step before backward: tensor(0.0045, device='cuda:0', grad_fn=<MseLossBackward>)
  3701. loss_train_step after backward: tensor(0.0045, device='cuda:0', grad_fn=<MseLossBackward>)
  3702. loss_train: 0.06192554801236838
  3703. step: 18
  3704. running loss: 0.0034403082229093546
  3705.  
  3706. Train Steps: 18/90 Loss: 0.0034 torch.Size([8, 600, 800])
  3707. torch.Size([8, 8])
  3708. tensor([[0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284],
  3709. [0.6076, 0.3953, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954],
  3710. [0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  3711. [0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364],
  3712. [0.6185, 0.4129, 0.8900, 0.4567, 0.3937, 0.5417, 0.5734, 0.5110],
  3713. [0.6227, 0.4083, 0.8938, 0.4800, 0.3800, 0.2950, 0.5737, 0.5350],
  3714. [0.6275, 0.4024, 0.8500, 0.5383, 0.3912, 0.4883, 0.6288, 0.5100],
  3715. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391]],
  3716. device='cuda:0', dtype=torch.float64)
  3717. predictions are: tensor([[0.5125, 0.2871, 0.7511, 0.2094, 0.4505, 0.1997, 0.5630, 0.4911],
  3718. [0.4933, 0.3047, 0.8090, 0.3301, 0.3396, 0.3870, 0.5353, 0.5002],
  3719. [0.5513, 0.3362, 0.8229, 0.4283, 0.3745, 0.5247, 0.5657, 0.4955],
  3720. [0.5318, 0.3230, 0.7729, 0.2162, 0.4420, 0.2273, 0.6116, 0.5084],
  3721. [0.5149, 0.2947, 0.9065, 0.5164, 0.4007, 0.5175, 0.5539, 0.5059],
  3722. [0.6284, 0.3912, 0.8496, 0.4788, 0.4001, 0.3236, 0.5528, 0.5050],
  3723. [0.5536, 0.3453, 0.8191, 0.5324, 0.3937, 0.4933, 0.5993, 0.4796],
  3724. [0.5261, 0.3023, 0.7768, 0.2523, 0.4267, 0.2545, 0.5609, 0.4950]],
  3725. device='cuda:0', grad_fn=<AddmmBackward>)
  3726. landmarks are: tensor([[[0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284],
  3727. [0.6076, 0.3952, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954],
  3728. [0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  3729. [0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364],
  3730. [0.6186, 0.4129, 0.8900, 0.4567, 0.3938, 0.5417, 0.5734, 0.5110],
  3731. [0.6227, 0.4083, 0.8938, 0.4800, 0.3800, 0.2950, 0.5738, 0.5350],
  3732. [0.6275, 0.4024, 0.8500, 0.5383, 0.3913, 0.4883, 0.6288, 0.5100],
  3733. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391]]],
  3734. device='cuda:0')
  3735. loss_train_step before backward: tensor(0.0025, device='cuda:0', grad_fn=<MseLossBackward>)
  3736. loss_train_step after backward: tensor(0.0025, device='cuda:0', grad_fn=<MseLossBackward>)
  3737. loss_train: 0.0644436435541138
  3738. step: 19
  3739. running loss: 0.003391770713374411
  3740. Train Steps: 19/90 Loss: 0.0034 torch.Size([8, 600, 800])
  3741. torch.Size([8, 8])
  3742. tensor([[0.6197, 0.4090, 0.7825, 0.2500, 0.4200, 0.2483, 0.5988, 0.5667],
  3743. [0.6339, 0.4123, 0.8638, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  3744. [0.6200, 0.3993, 0.8519, 0.4923, 0.3962, 0.4717, 0.6013, 0.5433],
  3745. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  3746. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5363, 0.5550],
  3747. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  3748. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  3749. [0.6097, 0.3988, 0.8650, 0.5250, 0.4213, 0.5200, 0.5675, 0.5050]],
  3750. device='cuda:0', dtype=torch.float64)
  3751. predictions are: tensor([[0.5718, 0.3611, 0.7569, 0.1923, 0.4113, 0.2334, 0.5492, 0.5280],
  3752. [0.5646, 0.3674, 0.8176, 0.4881, 0.4099, 0.5417, 0.6300, 0.5248],
  3753. [0.5698, 0.3466, 0.8024, 0.4662, 0.3711, 0.4423, 0.5449, 0.5125],
  3754. [0.6244, 0.3827, 0.8229, 0.3401, 0.3483, 0.3685, 0.5368, 0.5160],
  3755. [0.5406, 0.3454, 0.6538, 0.2001, 0.3887, 0.2130, 0.4810, 0.4885],
  3756. [0.5050, 0.3236, 0.8718, 0.3897, 0.4116, 0.2701, 0.5983, 0.4890],
  3757. [0.4879, 0.3313, 0.8646, 0.3113, 0.4415, 0.2104, 0.6023, 0.4683],
  3758. [0.5498, 0.3600, 0.8173, 0.4844, 0.4335, 0.4855, 0.4931, 0.4710]],
  3759. device='cuda:0', grad_fn=<AddmmBackward>)
  3760. landmarks are: tensor([[[0.6197, 0.4090, 0.7825, 0.2500, 0.4200, 0.2483, 0.5987, 0.5667],
  3761. [0.6339, 0.4123, 0.8637, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  3762. [0.6200, 0.3993, 0.8519, 0.4923, 0.3963, 0.4717, 0.6012, 0.5433],
  3763. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  3764. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5362, 0.5550],
  3765. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  3766. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  3767. [0.6097, 0.3988, 0.8650, 0.5250, 0.4212, 0.5200, 0.5675, 0.5050]]],
  3768. device='cuda:0')
  3769. loss_train_step before backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  3770. loss_train_step after backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  3771. loss_train: 0.06718559598084539
  3772. step: 20
  3773. running loss: 0.0033592797990422696
  3774. Train Steps: 20/90 Loss: 0.0034 torch.Size([8, 600, 800])
  3775. torch.Size([8, 8])
  3776. tensor([[0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  3777. [0.6179, 0.4118, 0.7278, 0.4237, 0.3588, 0.3400, 0.5675, 0.5917],
  3778. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  3779. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  3780. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  3781. [ nan, nan, 0.7648, 0.2722, 0.3962, 0.2183, 0.5060, 0.5422],
  3782. [0.6268, 0.4029, 0.8500, 0.2683, 0.3937, 0.3500, 0.6860, 0.5297],
  3783. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869]],
  3784. device='cuda:0', dtype=torch.float64)
  3785. predictions are: tensor([[0.5591, 0.3825, 0.8259, 0.4147, 0.4141, 0.3745, 0.5544, 0.5147],
  3786. [0.5839, 0.4089, 0.7758, 0.2976, 0.3758, 0.2668, 0.5041, 0.5495],
  3787. [0.5310, 0.3788, 0.8315, 0.4217, 0.4207, 0.3984, 0.5459, 0.5369],
  3788. [0.6807, 0.4494, 0.8215, 0.5131, 0.3843, 0.3442, 0.5462, 0.5332],
  3789. [0.5852, 0.4040, 0.8322, 0.2623, 0.3493, 0.4323, 0.6110, 0.5552],
  3790. [0.3095, 0.2227, 0.7204, 0.1732, 0.4180, 0.1595, 0.4772, 0.5234],
  3791. [0.6958, 0.4777, 0.7973, 0.2119, 0.3881, 0.2798, 0.6438, 0.5153],
  3792. [0.5824, 0.4102, 0.8456, 0.4937, 0.4169, 0.3476, 0.5649, 0.5045]],
  3793. device='cuda:0', grad_fn=<AddmmBackward>)
  3794. landmarks are: tensor([[[0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  3795. [0.6179, 0.4118, 0.7278, 0.4237, 0.3587, 0.3400, 0.5675, 0.5917],
  3796. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  3797. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  3798. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  3799. [0.0000, 0.0000, 0.7648, 0.2722, 0.3963, 0.2183, 0.5060, 0.5422],
  3800. [0.6268, 0.4029, 0.8500, 0.2683, 0.3938, 0.3500, 0.6860, 0.5297],
  3801. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869]]],
  3802. device='cuda:0')
  3803. loss_train_step before backward: tensor(0.0048, device='cuda:0', grad_fn=<MseLossBackward>)
  3804. loss_train_step after backward: tensor(0.0048, device='cuda:0', grad_fn=<MseLossBackward>)
  3805. loss_train: 0.07193988130893558
  3806. step: 21
  3807. running loss: 0.0034257086337588375
  3808. Train Steps: 21/90 Loss: 0.0034 torch.Size([8, 600, 800])
  3809. torch.Size([8, 8])
  3810. tensor([[0.6148, 0.4076, 0.8666, 0.4820, 0.4138, 0.5067, 0.5250, 0.5767],
  3811. [0.6265, 0.4071, 0.8875, 0.3367, 0.3975, 0.3350, 0.6312, 0.5250],
  3812. [0.6059, 0.4002, 0.7562, 0.2767, 0.3538, 0.3033, 0.5529, 0.5455],
  3813. [0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915],
  3814. [0.6138, 0.4054, 0.8750, 0.4750, 0.4363, 0.5017, 0.5086, 0.5822],
  3815. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  3816. [ nan, nan, 0.7512, 0.2117, 0.4288, 0.2000, 0.5600, 0.5617],
  3817. [0.6098, 0.3991, 0.8638, 0.4717, 0.4263, 0.4967, 0.5212, 0.5650]],
  3818. device='cuda:0', dtype=torch.float64)
  3819. predictions are: tensor([[0.6355, 0.4614, 0.8250, 0.4840, 0.3991, 0.4634, 0.5587, 0.5688],
  3820. [0.5501, 0.3784, 0.8598, 0.2704, 0.3954, 0.2741, 0.6230, 0.5681],
  3821. [0.6268, 0.4654, 0.7134, 0.2568, 0.3502, 0.2695, 0.5437, 0.5430],
  3822. [0.6804, 0.4713, 0.8506, 0.4430, 0.3397, 0.2755, 0.5728, 0.5126],
  3823. [0.5909, 0.4225, 0.8107, 0.4818, 0.3905, 0.4480, 0.5322, 0.5546],
  3824. [0.5062, 0.3929, 0.8450, 0.2082, 0.5034, 0.1870, 0.6914, 0.5587],
  3825. [0.3170, 0.2427, 0.7268, 0.1864, 0.4102, 0.1592, 0.4986, 0.5714],
  3826. [0.5695, 0.4248, 0.7992, 0.4864, 0.3944, 0.4476, 0.5511, 0.5406]],
  3827. device='cuda:0', grad_fn=<AddmmBackward>)
  3828. landmarks are: tensor([[[0.6148, 0.4076, 0.8666, 0.4820, 0.4137, 0.5067, 0.5250, 0.5767],
  3829. [0.6265, 0.4071, 0.8875, 0.3367, 0.3975, 0.3350, 0.6313, 0.5250],
  3830. [0.6059, 0.4002, 0.7563, 0.2767, 0.3537, 0.3033, 0.5529, 0.5455],
  3831. [0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915],
  3832. [0.6138, 0.4054, 0.8750, 0.4750, 0.4363, 0.5017, 0.5086, 0.5822],
  3833. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  3834. [0.0000, 0.0000, 0.7513, 0.2117, 0.4288, 0.2000, 0.5600, 0.5617],
  3835. [0.6098, 0.3991, 0.8637, 0.4717, 0.4263, 0.4967, 0.5213, 0.5650]]],
  3836. device='cuda:0')
  3837. loss_train_step before backward: tensor(0.0041, device='cuda:0', grad_fn=<MseLossBackward>)
  3838. loss_train_step after backward: tensor(0.0041, device='cuda:0', grad_fn=<MseLossBackward>)
  3839. loss_train: 0.07605478435289115
  3840. step: 22
  3841. running loss: 0.003457035652404143
  3842.  
  3843. Train Steps: 22/90 Loss: 0.0035 torch.Size([8, 600, 800])
  3844. torch.Size([8, 8])
  3845. tensor([[0.6300, 0.4133, 0.8538, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  3846. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  3847. [0.6346, 0.4086, 0.7938, 0.5500, 0.3962, 0.4867, 0.7343, 0.5702],
  3848. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  3849. [0.6265, 0.4251, 0.7113, 0.3550, 0.4375, 0.2117, 0.5587, 0.6118],
  3850. [0.6206, 0.4001, 0.8900, 0.3933, 0.3588, 0.3567, 0.5837, 0.5083],
  3851. [0.6197, 0.4091, 0.8800, 0.4783, 0.3538, 0.4767, 0.5950, 0.5550],
  3852. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533]],
  3853. device='cuda:0', dtype=torch.float64)
  3854. predictions are: tensor([[0.4578, 0.3519, 0.8790, 0.2221, 0.5106, 0.2299, 0.6964, 0.5784],
  3855. [0.5498, 0.3720, 0.7646, 0.2784, 0.3612, 0.3090, 0.5244, 0.5695],
  3856. [0.6279, 0.4609, 0.8070, 0.5108, 0.3879, 0.4680, 0.6271, 0.5730],
  3857. [0.6880, 0.4932, 0.8739, 0.4629, 0.3677, 0.3500, 0.5938, 0.5400],
  3858. [0.5222, 0.4169, 0.7394, 0.2919, 0.3977, 0.2462, 0.5193, 0.6034],
  3859. [0.6170, 0.4211, 0.8745, 0.3801, 0.3498, 0.3347, 0.5682, 0.5492],
  3860. [0.6112, 0.4401, 0.8528, 0.4759, 0.3710, 0.4550, 0.5889, 0.5859],
  3861. [0.5333, 0.4033, 0.8534, 0.3997, 0.4135, 0.4587, 0.5547, 0.5614]],
  3862. device='cuda:0', grad_fn=<AddmmBackward>)
  3863. landmarks are: tensor([[[0.6300, 0.4133, 0.8537, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  3864. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  3865. [0.6346, 0.4086, 0.7937, 0.5500, 0.3963, 0.4867, 0.7343, 0.5702],
  3866. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  3867. [0.6265, 0.4251, 0.7113, 0.3550, 0.4375, 0.2117, 0.5587, 0.6118],
  3868. [0.6206, 0.4001, 0.8900, 0.3933, 0.3587, 0.3567, 0.5838, 0.5083],
  3869. [0.6197, 0.4091, 0.8800, 0.4783, 0.3537, 0.4767, 0.5950, 0.5550],
  3870. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533]]],
  3871. device='cuda:0')
  3872. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  3873. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  3874. loss_train: 0.07791255612391979
  3875. step: 23
  3876. running loss: 0.0033875024401704254
  3877. Train Steps: 23/90 Loss: 0.0034 torch.Size([8, 600, 800])
  3878. torch.Size([8, 8])
  3879. tensor([[0.6125, 0.4035, 0.7825, 0.3100, 0.3463, 0.4900, 0.5832, 0.5637],
  3880. [0.6137, 0.4038, 0.8563, 0.4050, 0.3813, 0.2550, 0.5106, 0.4954],
  3881. [0.6261, 0.4066, 0.8325, 0.2150, 0.4763, 0.2667, 0.7002, 0.5633],
  3882. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250],
  3883. [0.6267, 0.4080, 0.8438, 0.2633, 0.4763, 0.1800, 0.6259, 0.5240],
  3884. [ nan, nan, 0.8850, 0.2817, 0.5112, 0.2183, 0.7184, 0.5436],
  3885. [0.6307, 0.4029, 0.8988, 0.4817, 0.3937, 0.3500, 0.7311, 0.5378],
  3886. [0.6069, 0.3975, 0.8625, 0.5083, 0.4388, 0.5483, 0.5650, 0.4967]],
  3887. device='cuda:0', dtype=torch.float64)
  3888. predictions are: tensor([[0.5630, 0.4143, 0.7481, 0.3400, 0.3264, 0.4892, 0.5747, 0.5941],
  3889. [0.5112, 0.3690, 0.7888, 0.3976, 0.3908, 0.2515, 0.4968, 0.5846],
  3890. [0.5933, 0.4247, 0.7916, 0.2289, 0.4429, 0.2337, 0.6436, 0.6058],
  3891. [0.6395, 0.4488, 0.8615, 0.4161, 0.3459, 0.3268, 0.5817, 0.5449],
  3892. [0.5625, 0.4243, 0.8189, 0.2745, 0.4324, 0.2153, 0.5916, 0.5724],
  3893. [0.4407, 0.3259, 0.8781, 0.2678, 0.4856, 0.2394, 0.6811, 0.5764],
  3894. [0.6495, 0.4705, 0.8612, 0.4888, 0.3962, 0.3317, 0.6322, 0.5577],
  3895. [0.6276, 0.4688, 0.8113, 0.5343, 0.4078, 0.5237, 0.5561, 0.6000]],
  3896. device='cuda:0', grad_fn=<AddmmBackward>)
  3897. landmarks are: tensor([[[0.6125, 0.4035, 0.7825, 0.3100, 0.3462, 0.4900, 0.5832, 0.5637],
  3898. [0.6137, 0.4038, 0.8562, 0.4050, 0.3812, 0.2550, 0.5106, 0.4954],
  3899. [0.6261, 0.4066, 0.8325, 0.2150, 0.4762, 0.2667, 0.7002, 0.5633],
  3900. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250],
  3901. [0.6267, 0.4080, 0.8438, 0.2633, 0.4762, 0.1800, 0.6259, 0.5240],
  3902. [0.0000, 0.0000, 0.8850, 0.2817, 0.5113, 0.2183, 0.7184, 0.5436],
  3903. [0.6307, 0.4029, 0.8988, 0.4817, 0.3938, 0.3500, 0.7311, 0.5378],
  3904. [0.6069, 0.3975, 0.8625, 0.5083, 0.4387, 0.5483, 0.5650, 0.4967]]],
  3905. device='cuda:0')
  3906. loss_train_step before backward: tensor(0.0063, device='cuda:0', grad_fn=<MseLossBackward>)
  3907. loss_train_step after backward: tensor(0.0063, device='cuda:0', grad_fn=<MseLossBackward>)
  3908. loss_train: 0.08417737309355289
  3909. step: 24
  3910. running loss: 0.0035073905455647036
  3911. Train Steps: 24/90 Loss: 0.0035 torch.Size([8, 600, 800])
  3912. torch.Size([8, 8])
  3913. tensor([[ nan, nan, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  3914. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250],
  3915. [0.6346, 0.4092, 0.7712, 0.5917, 0.4037, 0.4767, 0.7343, 0.5725],
  3916. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5513, 0.5750],
  3917. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  3918. [0.6133, 0.4094, 0.8495, 0.4028, 0.3588, 0.3200, 0.5003, 0.5407],
  3919. [0.6075, 0.4007, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280],
  3920. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600]],
  3921. device='cuda:0', dtype=torch.float64)
  3922. predictions are: tensor([[0.3853, 0.2412, 0.7946, 0.2664, 0.3667, 0.2621, 0.5691, 0.5883],
  3923. [0.6728, 0.4472, 0.9294, 0.3902, 0.3840, 0.3261, 0.6674, 0.5337],
  3924. [0.6466, 0.4458, 0.8176, 0.4990, 0.4050, 0.4815, 0.6699, 0.6091],
  3925. [0.2990, 0.2104, 0.7380, 0.1906, 0.4629, 0.1769, 0.5771, 0.6129],
  3926. [0.6577, 0.4726, 0.9106, 0.4686, 0.3657, 0.4160, 0.6843, 0.5803],
  3927. [0.5791, 0.3975, 0.8685, 0.3931, 0.3932, 0.3091, 0.5628, 0.5776],
  3928. [0.6461, 0.4384, 0.8341, 0.4709, 0.4037, 0.5106, 0.6007, 0.5653],
  3929. [0.7838, 0.5244, 0.8995, 0.5693, 0.4290, 0.5269, 0.6741, 0.5730]],
  3930. device='cuda:0', grad_fn=<AddmmBackward>)
  3931. landmarks are: tensor([[[0.0000, 0.0000, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  3932. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250],
  3933. [0.6346, 0.4092, 0.7713, 0.5917, 0.4038, 0.4767, 0.7343, 0.5725],
  3934. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5512, 0.5750],
  3935. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  3936. [0.6133, 0.4094, 0.8495, 0.4028, 0.3587, 0.3200, 0.5003, 0.5407],
  3937. [0.6075, 0.4006, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280],
  3938. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600]]],
  3939. device='cuda:0')
  3940. loss_train_step before backward: tensor(0.0073, device='cuda:0', grad_fn=<MseLossBackward>)
  3941. loss_train_step after backward: tensor(0.0073, device='cuda:0', grad_fn=<MseLossBackward>)
  3942. loss_train: 0.09143890452105552
  3943. step: 25
  3944. running loss: 0.0036575561808422207
  3945. Train Steps: 25/90 Loss: 0.0037 torch.Size([8, 600, 800])
  3946. torch.Size([8, 8])
  3947. tensor([[0.6145, 0.4007, 0.8775, 0.4533, 0.4562, 0.5533, 0.6088, 0.5533],
  3948. [0.6339, 0.4123, 0.8638, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  3949. [0.6254, 0.3993, 0.8988, 0.4767, 0.3987, 0.5517, 0.6955, 0.5285],
  3950. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  3951. [0.6339, 0.4102, 0.8588, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089],
  3952. [0.6154, 0.4112, 0.7037, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  3953. [0.6265, 0.4088, 0.8025, 0.1850, 0.4163, 0.2500, 0.6290, 0.4947],
  3954. [0.6118, 0.4052, 0.8463, 0.3917, 0.3538, 0.3450, 0.5053, 0.5593]],
  3955. device='cuda:0', dtype=torch.float64)
  3956. predictions are: tensor([[0.6105, 0.3962, 0.8903, 0.5136, 0.4423, 0.5345, 0.6518, 0.5574],
  3957. [0.7116, 0.4549, 0.8984, 0.5937, 0.4033, 0.5888, 0.7316, 0.5672],
  3958. [0.7130, 0.4386, 0.9326, 0.5597, 0.3984, 0.5887, 0.6999, 0.5537],
  3959. [0.4522, 0.2795, 0.7579, 0.2800, 0.4087, 0.2454, 0.5992, 0.5730],
  3960. [0.6696, 0.4300, 0.9044, 0.3524, 0.4478, 0.2530, 0.6978, 0.5384],
  3961. [0.4153, 0.2761, 0.7562, 0.2718, 0.4272, 0.2038, 0.5829, 0.5567],
  3962. [0.4782, 0.3126, 0.8689, 0.2631, 0.4484, 0.2345, 0.6533, 0.5612],
  3963. [0.6062, 0.3808, 0.8682, 0.4297, 0.3464, 0.3637, 0.5453, 0.5646]],
  3964. device='cuda:0', grad_fn=<AddmmBackward>)
  3965. landmarks are: tensor([[[0.6145, 0.4007, 0.8775, 0.4533, 0.4563, 0.5533, 0.6087, 0.5533],
  3966. [0.6339, 0.4123, 0.8637, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  3967. [0.6254, 0.3993, 0.8988, 0.4767, 0.3988, 0.5517, 0.6955, 0.5285],
  3968. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  3969. [0.6339, 0.4102, 0.8587, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089],
  3970. [0.6154, 0.4112, 0.7038, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  3971. [0.6265, 0.4088, 0.8025, 0.1850, 0.4162, 0.2500, 0.6290, 0.4947],
  3972. [0.6118, 0.4052, 0.8462, 0.3917, 0.3537, 0.3450, 0.5053, 0.5593]]],
  3973. device='cuda:0')
  3974. loss_train_step before backward: tensor(0.0035, device='cuda:0', grad_fn=<MseLossBackward>)
  3975. loss_train_step after backward: tensor(0.0035, device='cuda:0', grad_fn=<MseLossBackward>)
  3976. loss_train: 0.09492857556324452
  3977. step: 26
  3978. running loss: 0.0036510990601247894
  3979.  
  3980. Train Steps: 26/90 Loss: 0.0037 torch.Size([8, 600, 800])
  3981. torch.Size([8, 8])
  3982. tensor([[0.6339, 0.4149, 0.8800, 0.5000, 0.3900, 0.5283, 0.7541, 0.5424],
  3983. [ nan, nan, 0.7335, 0.2569, 0.3788, 0.2667, 0.5066, 0.5578],
  3984. [0.6252, 0.4158, 0.8988, 0.4083, 0.3788, 0.4783, 0.6225, 0.5633],
  3985. [0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6262, 0.5167],
  3986. [0.6250, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6088, 0.5183],
  3987. [0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  3988. [0.6159, 0.4085, 0.6900, 0.2283, 0.4088, 0.1950, 0.5123, 0.5397],
  3989. [0.6141, 0.4038, 0.8650, 0.4833, 0.4839, 0.5176, 0.5787, 0.5600]],
  3990. device='cuda:0', dtype=torch.float64)
  3991. predictions are: tensor([[0.7710, 0.4596, 0.9157, 0.5361, 0.3850, 0.5504, 0.7209, 0.5340],
  3992. [0.1755, 0.0911, 0.7831, 0.2816, 0.4241, 0.2348, 0.5595, 0.5750],
  3993. [0.6858, 0.4413, 0.8724, 0.4188, 0.3791, 0.4858, 0.6770, 0.5232],
  3994. [0.6731, 0.4134, 0.9235, 0.4454, 0.3964, 0.3642, 0.6883, 0.5083],
  3995. [0.6874, 0.3868, 0.9253, 0.4927, 0.4442, 0.5354, 0.6994, 0.5263],
  3996. [0.6664, 0.3893, 0.8907, 0.4971, 0.4151, 0.4613, 0.6301, 0.5303],
  3997. [0.4252, 0.2396, 0.7291, 0.2757, 0.4188, 0.1851, 0.5661, 0.5395],
  3998. [0.6321, 0.3918, 0.8892, 0.5291, 0.4634, 0.4708, 0.6284, 0.5273]],
  3999. device='cuda:0', grad_fn=<AddmmBackward>)
  4000. landmarks are: tensor([[[0.6339, 0.4149, 0.8800, 0.5000, 0.3900, 0.5283, 0.7541, 0.5424],
  4001. [0.0000, 0.0000, 0.7335, 0.2569, 0.3787, 0.2667, 0.5066, 0.5578],
  4002. [0.6252, 0.4158, 0.8988, 0.4083, 0.3787, 0.4783, 0.6225, 0.5633],
  4003. [0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6263, 0.5167],
  4004. [0.6251, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6087, 0.5183],
  4005. [0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  4006. [0.6159, 0.4085, 0.6900, 0.2283, 0.4087, 0.1950, 0.5123, 0.5397],
  4007. [0.6141, 0.4038, 0.8650, 0.4833, 0.4839, 0.5176, 0.5788, 0.5600]]],
  4008. device='cuda:0')
  4009. loss_train_step before backward: tensor(0.0031, device='cuda:0', grad_fn=<MseLossBackward>)
  4010. loss_train_step after backward: tensor(0.0031, device='cuda:0', grad_fn=<MseLossBackward>)
  4011. loss_train: 0.09805791999679059
  4012. step: 27
  4013. running loss: 0.003631774814695948
  4014. Train Steps: 27/90 Loss: 0.0036 torch.Size([8, 600, 800])
  4015. torch.Size([8, 8])
  4016. tensor([[0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  4017. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  4018. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  4019. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683],
  4020. [0.6311, 0.3998, 0.7975, 0.5767, 0.3838, 0.4850, 0.7327, 0.5343],
  4021. [0.6229, 0.4066, 0.7612, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  4022. [0.6305, 0.3983, 0.8950, 0.4833, 0.3688, 0.4683, 0.6375, 0.5117],
  4023. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461]],
  4024. device='cuda:0', dtype=torch.float64)
  4025. predictions are: tensor([[0.5901, 0.3425, 0.8881, 0.3529, 0.3891, 0.5200, 0.6666, 0.5400],
  4026. [0.6469, 0.3587, 0.8641, 0.3298, 0.3718, 0.4317, 0.6578, 0.5125],
  4027. [0.4706, 0.2787, 0.8005, 0.2850, 0.4421, 0.2119, 0.6169, 0.5088],
  4028. [0.7143, 0.4069, 0.8990, 0.5666, 0.3859, 0.4117, 0.5918, 0.5235],
  4029. [0.5770, 0.3445, 0.8539, 0.5257, 0.4015, 0.4867, 0.6826, 0.5029],
  4030. [0.6092, 0.3633, 0.8144, 0.3106, 0.4484, 0.2981, 0.6340, 0.5114],
  4031. [0.6334, 0.3657, 0.9173, 0.5061, 0.3971, 0.5020, 0.6485, 0.4940],
  4032. [0.6372, 0.3549, 0.8855, 0.5803, 0.4888, 0.5151, 0.5938, 0.5249]],
  4033. device='cuda:0', grad_fn=<AddmmBackward>)
  4034. landmarks are: tensor([[[0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  4035. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  4036. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  4037. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683],
  4038. [0.6311, 0.3998, 0.7975, 0.5767, 0.3837, 0.4850, 0.7327, 0.5343],
  4039. [0.6229, 0.4066, 0.7613, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  4040. [0.6305, 0.3983, 0.8950, 0.4833, 0.3688, 0.4683, 0.6375, 0.5117],
  4041. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461]]],
  4042. device='cuda:0')
  4043. loss_train_step before backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  4044. loss_train_step after backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  4045. loss_train: 0.1000079937512055
  4046. step: 28
  4047. running loss: 0.0035717140625430538
  4048. Train Steps: 28/90 Loss: 0.0036 torch.Size([8, 600, 800])
  4049. torch.Size([8, 8])
  4050. tensor([[0.6239, 0.4206, 0.8750, 0.5400, 0.3688, 0.4850, 0.5737, 0.5700],
  4051. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5837, 0.5583],
  4052. [ nan, nan, 0.9088, 0.3783, 0.4562, 0.2617, 0.6741, 0.5575],
  4053. [0.6277, 0.4118, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  4054. [0.6226, 0.4098, 0.8912, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  4055. [0.6272, 0.4071, 0.8738, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  4056. [0.6102, 0.3999, 0.8750, 0.5133, 0.3825, 0.4750, 0.5637, 0.5083],
  4057. [0.6022, 0.3994, 0.8025, 0.3350, 0.3350, 0.4400, 0.5565, 0.5025]],
  4058. device='cuda:0', dtype=torch.float64)
  4059. predictions are: tensor([[0.7127, 0.4073, 0.8357, 0.5393, 0.4057, 0.5256, 0.6161, 0.5145],
  4060. [0.5640, 0.3250, 0.8582, 0.3967, 0.4815, 0.4831, 0.5866, 0.5296],
  4061. [0.4414, 0.2499, 0.8859, 0.3510, 0.4725, 0.3156, 0.6743, 0.5210],
  4062. [0.5667, 0.3196, 0.8532, 0.3744, 0.4023, 0.2934, 0.6232, 0.5004],
  4063. [0.5504, 0.3089, 0.8489, 0.3942, 0.4237, 0.3078, 0.5773, 0.4963],
  4064. [0.7316, 0.4151, 0.8370, 0.5199, 0.3864, 0.4018, 0.6152, 0.4532],
  4065. [0.6351, 0.3639, 0.8403, 0.4770, 0.4038, 0.4873, 0.5663, 0.4826],
  4066. [0.5848, 0.3555, 0.7833, 0.3287, 0.3813, 0.4267, 0.5954, 0.5164]],
  4067. device='cuda:0', grad_fn=<AddmmBackward>)
  4068. landmarks are: tensor([[[0.6239, 0.4206, 0.8750, 0.5400, 0.3688, 0.4850, 0.5738, 0.5700],
  4069. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5838, 0.5583],
  4070. [0.0000, 0.0000, 0.9087, 0.3783, 0.4563, 0.2617, 0.6741, 0.5575],
  4071. [0.6277, 0.4117, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  4072. [0.6226, 0.4098, 0.8913, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  4073. [0.6272, 0.4071, 0.8737, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  4074. [0.6102, 0.3999, 0.8750, 0.5133, 0.3825, 0.4750, 0.5638, 0.5083],
  4075. [0.6022, 0.3994, 0.8025, 0.3350, 0.3350, 0.4400, 0.5565, 0.5025]]],
  4076. device='cuda:0')
  4077. loss_train_step before backward: tensor(0.0057, device='cuda:0', grad_fn=<MseLossBackward>)
  4078. loss_train_step after backward: tensor(0.0057, device='cuda:0', grad_fn=<MseLossBackward>)
  4079. loss_train: 0.10573010763619095
  4080. step: 29
  4081. running loss: 0.003645865780558309
  4082. Train Steps: 29/90 Loss: 0.0036 torch.Size([8, 600, 800])
  4083. torch.Size([8, 8])
  4084. tensor([[0.6134, 0.4090, 0.6926, 0.2819, 0.3538, 0.3233, 0.5563, 0.5667],
  4085. [0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  4086. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  4087. [0.6129, 0.4069, 0.8750, 0.5067, 0.3875, 0.4233, 0.5235, 0.5881],
  4088. [0.6127, 0.4115, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  4089. [0.6085, 0.4008, 0.8588, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  4090. [ nan, nan, 0.9088, 0.3783, 0.4562, 0.2617, 0.6741, 0.5575],
  4091. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131]],
  4092. device='cuda:0', dtype=torch.float64)
  4093. predictions are: tensor([[0.5710, 0.3733, 0.7081, 0.3260, 0.3558, 0.3593, 0.5647, 0.5183],
  4094. [0.6567, 0.4062, 0.8642, 0.4450, 0.4036, 0.5435, 0.6240, 0.4872],
  4095. [0.5467, 0.3397, 0.8757, 0.5004, 0.4346, 0.4875, 0.5935, 0.5118],
  4096. [0.6675, 0.4063, 0.8734, 0.5478, 0.3892, 0.4632, 0.5665, 0.5052],
  4097. [0.5614, 0.3626, 0.7455, 0.3123, 0.3859, 0.3185, 0.5494, 0.5276],
  4098. [0.6181, 0.3625, 0.8679, 0.5189, 0.4946, 0.4535, 0.6002, 0.4781],
  4099. [0.4248, 0.2752, 0.9296, 0.4028, 0.4848, 0.3101, 0.6864, 0.5216],
  4100. [0.6556, 0.4074, 0.8741, 0.4087, 0.3746, 0.4201, 0.6057, 0.4803]],
  4101. device='cuda:0', grad_fn=<AddmmBackward>)
  4102. landmarks are: tensor([[[0.6134, 0.4090, 0.6926, 0.2819, 0.3537, 0.3233, 0.5562, 0.5667],
  4103. [0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  4104. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  4105. [0.6129, 0.4069, 0.8750, 0.5067, 0.3875, 0.4233, 0.5235, 0.5881],
  4106. [0.6127, 0.4114, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  4107. [0.6084, 0.4008, 0.8587, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  4108. [0.0000, 0.0000, 0.9087, 0.3783, 0.4563, 0.2617, 0.6741, 0.5575],
  4109. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131]]],
  4110. device='cuda:0')
  4111. loss_train_step before backward: tensor(0.0052, device='cuda:0', grad_fn=<MseLossBackward>)
  4112. loss_train_step after backward: tensor(0.0052, device='cuda:0', grad_fn=<MseLossBackward>)
  4113. loss_train: 0.11088850733358413
  4114. step: 30
  4115. running loss: 0.003696283577786138
  4116.  
  4117. Train Steps: 30/90 Loss: 0.0037 torch.Size([8, 600, 800])
  4118. torch.Size([8, 8])
  4119. tensor([[0.6082, 0.4024, 0.8738, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  4120. [ nan, nan, 0.7268, 0.2333, 0.4125, 0.1933, 0.5112, 0.5383],
  4121. [ nan, nan, 0.6469, 0.1943, 0.4025, 0.2000, 0.5125, 0.5533],
  4122. [0.6110, 0.3984, 0.8750, 0.4933, 0.4625, 0.4950, 0.5578, 0.5676],
  4123. [0.6102, 0.3999, 0.8750, 0.5133, 0.3825, 0.4750, 0.5637, 0.5083],
  4124. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5513, 0.5750],
  4125. [ nan, nan, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  4126. [0.6332, 0.4165, 0.9100, 0.3350, 0.4188, 0.3683, 0.7438, 0.5528]],
  4127. device='cuda:0', dtype=torch.float64)
  4128. predictions are: tensor([[0.5922, 0.4017, 0.8713, 0.4405, 0.3612, 0.4238, 0.5184, 0.5151],
  4129. [0.3582, 0.2298, 0.7556, 0.3063, 0.4014, 0.2411, 0.5185, 0.5444],
  4130. [0.2946, 0.1886, 0.7170, 0.2601, 0.4136, 0.2340, 0.5188, 0.5606],
  4131. [0.7030, 0.4408, 0.9341, 0.5412, 0.4614, 0.5772, 0.5865, 0.5240],
  4132. [0.7953, 0.5029, 0.9016, 0.5643, 0.3886, 0.5612, 0.5478, 0.4993],
  4133. [0.5802, 0.3809, 0.7125, 0.2922, 0.4219, 0.2625, 0.5282, 0.5478],
  4134. [0.3664, 0.2401, 0.7380, 0.2883, 0.4304, 0.2344, 0.5059, 0.5507],
  4135. [0.6292, 0.4025, 0.9526, 0.4345, 0.4087, 0.4079, 0.6802, 0.5200]],
  4136. device='cuda:0', grad_fn=<AddmmBackward>)
  4137. landmarks are: tensor([[[0.6082, 0.4024, 0.8737, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  4138. [0.0000, 0.0000, 0.7268, 0.2333, 0.4125, 0.1933, 0.5113, 0.5383],
  4139. [0.0000, 0.0000, 0.6469, 0.1943, 0.4025, 0.2000, 0.5125, 0.5533],
  4140. [0.6110, 0.3984, 0.8750, 0.4933, 0.4625, 0.4950, 0.5578, 0.5676],
  4141. [0.6102, 0.3999, 0.8750, 0.5133, 0.3825, 0.4750, 0.5638, 0.5083],
  4142. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5512, 0.5750],
  4143. [0.0000, 0.0000, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  4144. [0.6332, 0.4165, 0.9100, 0.3350, 0.4187, 0.3683, 0.7438, 0.5528]]],
  4145. device='cuda:0')
  4146. loss_train_step before backward: tensor(0.0100, device='cuda:0', grad_fn=<MseLossBackward>)
  4147. loss_train_step after backward: tensor(0.0100, device='cuda:0', grad_fn=<MseLossBackward>)
  4148. loss_train: 0.12087913427967578
  4149. step: 31
  4150. running loss: 0.003899326912247606
  4151. Train Steps: 31/90 Loss: 0.0039 torch.Size([8, 600, 800])
  4152. torch.Size([8, 8])
  4153. tensor([[0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  4154. [0.6126, 0.4073, 0.8750, 0.5133, 0.3800, 0.4333, 0.4986, 0.5378],
  4155. [0.6230, 0.4152, 0.7588, 0.2283, 0.4012, 0.2883, 0.6200, 0.5767],
  4156. [0.6198, 0.4076, 0.7945, 0.2453, 0.4475, 0.1871, 0.5897, 0.5350],
  4157. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864],
  4158. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  4159. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  4160. [0.6183, 0.4076, 0.8838, 0.4517, 0.3813, 0.4483, 0.5775, 0.5633]],
  4161. device='cuda:0', dtype=torch.float64)
  4162. predictions are: tensor([[0.4806, 0.2998, 0.8918, 0.4452, 0.4107, 0.5104, 0.5396, 0.5420],
  4163. [0.4820, 0.3437, 0.8743, 0.5086, 0.3730, 0.4249, 0.5120, 0.5413],
  4164. [0.4814, 0.3564, 0.7628, 0.3216, 0.4086, 0.2817, 0.5722, 0.5560],
  4165. [0.5544, 0.3869, 0.7628, 0.2905, 0.4290, 0.2236, 0.5466, 0.5370],
  4166. [0.4566, 0.3315, 0.8674, 0.4907, 0.3861, 0.4601, 0.6026, 0.5633],
  4167. [0.5705, 0.4234, 0.7099, 0.3097, 0.3921, 0.2522, 0.5098, 0.5591],
  4168. [0.5269, 0.3574, 0.8905, 0.4368, 0.4251, 0.5545, 0.5447, 0.5600],
  4169. [0.4295, 0.2856, 0.8784, 0.4406, 0.3848, 0.4491, 0.5455, 0.5579]],
  4170. device='cuda:0', grad_fn=<AddmmBackward>)
  4171. landmarks are: tensor([[[0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  4172. [0.6126, 0.4073, 0.8750, 0.5133, 0.3800, 0.4333, 0.4986, 0.5378],
  4173. [0.6230, 0.4152, 0.7588, 0.2283, 0.4013, 0.2883, 0.6200, 0.5767],
  4174. [0.6198, 0.4076, 0.7945, 0.2453, 0.4475, 0.1871, 0.5897, 0.5350],
  4175. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864],
  4176. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  4177. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  4178. [0.6183, 0.4076, 0.8838, 0.4517, 0.3812, 0.4483, 0.5775, 0.5633]]],
  4179. device='cuda:0')
  4180. loss_train_step before backward: tensor(0.0040, device='cuda:0', grad_fn=<MseLossBackward>)
  4181. loss_train_step after backward: tensor(0.0040, device='cuda:0', grad_fn=<MseLossBackward>)
  4182. loss_train: 0.12484196957666427
  4183. step: 32
  4184. running loss: 0.0039013115492707584
  4185. Train Steps: 32/90 Loss: 0.0039 torch.Size([8, 600, 800])
  4186. torch.Size([8, 8])
  4187. tensor([[0.6140, 0.4070, 0.8700, 0.5000, 0.4612, 0.4900, 0.5260, 0.5852],
  4188. [0.6147, 0.4026, 0.6600, 0.2467, 0.4088, 0.2150, 0.5489, 0.5773],
  4189. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  4190. [0.6224, 0.4097, 0.7438, 0.2267, 0.3850, 0.2850, 0.5988, 0.5250],
  4191. [0.6265, 0.4071, 0.8875, 0.3367, 0.3975, 0.3350, 0.6312, 0.5250],
  4192. [0.6353, 0.4128, 0.8488, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  4193. [0.6058, 0.3978, 0.8287, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461],
  4194. [ nan, nan, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621]],
  4195. device='cuda:0', dtype=torch.float64)
  4196. predictions are: tensor([[0.5385, 0.3726, 0.8863, 0.5260, 0.4290, 0.5047, 0.5160, 0.5818],
  4197. [0.6499, 0.4474, 0.6659, 0.2844, 0.3601, 0.2414, 0.4926, 0.5716],
  4198. [0.4714, 0.3264, 0.7618, 0.3114, 0.3952, 0.2498, 0.5081, 0.5671],
  4199. [0.6327, 0.4728, 0.7146, 0.2756, 0.3698, 0.3003, 0.5684, 0.5530],
  4200. [0.5012, 0.3294, 0.8662, 0.3438, 0.3913, 0.3386, 0.5953, 0.5621],
  4201. [0.1845, 0.1296, 0.8389, 0.2707, 0.5110, 0.2170, 0.5929, 0.5841],
  4202. [0.6692, 0.4743, 0.8151, 0.3947, 0.3312, 0.4365, 0.5337, 0.5487],
  4203. [0.0382, 0.0638, 0.8499, 0.2784, 0.4862, 0.1974, 0.5803, 0.5828]],
  4204. device='cuda:0', grad_fn=<AddmmBackward>)
  4205. landmarks are: tensor([[[0.6140, 0.4070, 0.8700, 0.5000, 0.4613, 0.4900, 0.5260, 0.5852],
  4206. [0.6147, 0.4026, 0.6600, 0.2467, 0.4087, 0.2150, 0.5489, 0.5773],
  4207. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  4208. [0.6224, 0.4097, 0.7437, 0.2267, 0.3850, 0.2850, 0.5987, 0.5250],
  4209. [0.6265, 0.4071, 0.8875, 0.3367, 0.3975, 0.3350, 0.6313, 0.5250],
  4210. [0.6353, 0.4128, 0.8487, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  4211. [0.6058, 0.3978, 0.8288, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461],
  4212. [0.0000, 0.0000, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621]]],
  4213. device='cuda:0')
  4214. loss_train_step before backward: tensor(0.0065, device='cuda:0', grad_fn=<MseLossBackward>)
  4215. loss_train_step after backward: tensor(0.0065, device='cuda:0', grad_fn=<MseLossBackward>)
  4216. loss_train: 0.1313714758725837
  4217. step: 33
  4218. running loss: 0.003980953814320717
  4219. Train Steps: 33/90 Loss: 0.0040 torch.Size([8, 600, 800])
  4220. torch.Size([8, 8])
  4221. tensor([[ nan, nan, 0.8888, 0.3100, 0.5262, 0.2817, 0.7145, 0.6003],
  4222. [0.6170, 0.4102, 0.7468, 0.3695, 0.3463, 0.3767, 0.5238, 0.5823],
  4223. [0.6104, 0.4029, 0.8738, 0.4900, 0.4088, 0.4533, 0.5070, 0.5510],
  4224. [0.6136, 0.4085, 0.6688, 0.2317, 0.3862, 0.2367, 0.5517, 0.5783],
  4225. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  4226. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  4227. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  4228. [0.6361, 0.4076, 0.8862, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297]],
  4229. device='cuda:0', dtype=torch.float64)
  4230. predictions are: tensor([[0.2869, 0.1952, 0.8825, 0.2657, 0.4910, 0.2325, 0.6424, 0.5585],
  4231. [0.5847, 0.4135, 0.7391, 0.3092, 0.3318, 0.3839, 0.5174, 0.6051],
  4232. [0.5349, 0.3649, 0.8738, 0.4602, 0.4067, 0.4506, 0.5075, 0.5688],
  4233. [0.5407, 0.3837, 0.6720, 0.2071, 0.3599, 0.2307, 0.5037, 0.5578],
  4234. [0.3352, 0.2542, 0.6870, 0.2115, 0.3747, 0.2059, 0.5195, 0.5942],
  4235. [0.4570, 0.3381, 0.8863, 0.4052, 0.4083, 0.2791, 0.6182, 0.5715],
  4236. [0.4635, 0.3456, 0.8060, 0.3912, 0.4406, 0.2866, 0.5146, 0.6138],
  4237. [0.6070, 0.4359, 0.8601, 0.5168, 0.3663, 0.4816, 0.6266, 0.5784]],
  4238. device='cuda:0', grad_fn=<AddmmBackward>)
  4239. landmarks are: tensor([[[0.0000, 0.0000, 0.8888, 0.3100, 0.5263, 0.2817, 0.7145, 0.6003],
  4240. [0.6170, 0.4102, 0.7468, 0.3695, 0.3462, 0.3767, 0.5238, 0.5823],
  4241. [0.6104, 0.4029, 0.8737, 0.4900, 0.4087, 0.4533, 0.5070, 0.5510],
  4242. [0.6136, 0.4085, 0.6687, 0.2317, 0.3862, 0.2367, 0.5517, 0.5783],
  4243. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  4244. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  4245. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  4246. [0.6361, 0.4076, 0.8863, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297]]],
  4247. device='cuda:0')
  4248. loss_train_step before backward: tensor(0.0055, device='cuda:0', grad_fn=<MseLossBackward>)
  4249. loss_train_step after backward: tensor(0.0055, device='cuda:0', grad_fn=<MseLossBackward>)
  4250. loss_train: 0.1368311574915424
  4251. step: 34
  4252. running loss: 0.004024445808574776
  4253.  
  4254. Train Steps: 34/90 Loss: 0.0040 torch.Size([8, 600, 800])
  4255. torch.Size([8, 8])
  4256. tensor([[0.6142, 0.4127, 0.7575, 0.3067, 0.3438, 0.4383, 0.5778, 0.5207],
  4257. [0.6219, 0.3934, 0.8688, 0.5267, 0.4313, 0.4967, 0.5988, 0.4983],
  4258. [0.6310, 0.4017, 0.8563, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006],
  4259. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  4260. [0.6124, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  4261. [0.6053, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  4262. [0.6197, 0.4050, 0.7527, 0.2000, 0.4042, 0.2249, 0.5895, 0.4995],
  4263. [0.6131, 0.4064, 0.8638, 0.5200, 0.4788, 0.4783, 0.5258, 0.5867]],
  4264. device='cuda:0', dtype=torch.float64)
  4265. predictions are: tensor([[0.4963, 0.3862, 0.7514, 0.2513, 0.3589, 0.3394, 0.5619, 0.5993],
  4266. [0.4562, 0.3419, 0.8819, 0.4533, 0.4124, 0.4145, 0.5854, 0.5761],
  4267. [0.6197, 0.4237, 0.8685, 0.5307, 0.3926, 0.4467, 0.5880, 0.6016],
  4268. [0.5460, 0.3521, 0.9146, 0.4189, 0.4082, 0.4825, 0.5842, 0.5735],
  4269. [0.5055, 0.3693, 0.7325, 0.2684, 0.3590, 0.2630, 0.5277, 0.6086],
  4270. [0.3609, 0.2583, 0.6952, 0.1869, 0.4173, 0.1355, 0.5613, 0.5715],
  4271. [0.5359, 0.4054, 0.7737, 0.1995, 0.4128, 0.1637, 0.6384, 0.5586],
  4272. [0.5642, 0.3956, 0.8697, 0.4497, 0.4499, 0.4201, 0.6075, 0.5898]],
  4273. device='cuda:0', grad_fn=<AddmmBackward>)
  4274. landmarks are: tensor([[[0.6142, 0.4127, 0.7575, 0.3067, 0.3438, 0.4383, 0.5778, 0.5207],
  4275. [0.6219, 0.3934, 0.8687, 0.5267, 0.4313, 0.4967, 0.5987, 0.4983],
  4276. [0.6310, 0.4017, 0.8562, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006],
  4277. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  4278. [0.6123, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  4279. [0.6054, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  4280. [0.6197, 0.4050, 0.7527, 0.2000, 0.4042, 0.2249, 0.5895, 0.4995],
  4281. [0.6132, 0.4063, 0.8637, 0.5200, 0.4787, 0.4783, 0.5258, 0.5867]]],
  4282. device='cuda:0')
  4283. loss_train_step before backward: tensor(0.0043, device='cuda:0', grad_fn=<MseLossBackward>)
  4284. loss_train_step after backward: tensor(0.0043, device='cuda:0', grad_fn=<MseLossBackward>)
  4285. loss_train: 0.14112150168512017
  4286. step: 35
  4287. running loss: 0.004032042905289148
  4288. Train Steps: 35/90 Loss: 0.0040 torch.Size([8, 600, 800])
  4289. torch.Size([8, 8])
  4290. tensor([[0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  4291. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  4292. [0.6038, 0.3946, 0.8413, 0.4883, 0.3563, 0.4550, 0.5266, 0.4693],
  4293. [0.6257, 0.4034, 0.8287, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  4294. [0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500],
  4295. [0.6142, 0.4127, 0.7575, 0.3067, 0.3438, 0.4383, 0.5778, 0.5207],
  4296. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  4297. [ nan, nan, 0.7192, 0.2346, 0.4037, 0.2050, 0.5138, 0.5650]],
  4298. device='cuda:0', dtype=torch.float64)
  4299. predictions are: tensor([[0.4819, 0.3458, 0.6869, 0.2910, 0.3688, 0.2382, 0.5282, 0.5730],
  4300. [0.5417, 0.3932, 0.8482, 0.2817, 0.5320, 0.2175, 0.7606, 0.5362],
  4301. [0.6417, 0.4459, 0.8624, 0.4964, 0.3670, 0.4702, 0.5220, 0.5442],
  4302. [0.6376, 0.4358, 0.8091, 0.2879, 0.3997, 0.2430, 0.6488, 0.5202],
  4303. [0.6563, 0.4553, 0.7634, 0.2758, 0.3470, 0.3813, 0.6249, 0.5541],
  4304. [0.5801, 0.4157, 0.7382, 0.2946, 0.3624, 0.3834, 0.5646, 0.5530],
  4305. [0.5191, 0.3747, 0.8364, 0.2815, 0.5005, 0.2101, 0.6680, 0.5510],
  4306. [0.2037, 0.1414, 0.7333, 0.2334, 0.4252, 0.2144, 0.5101, 0.5678]],
  4307. device='cuda:0', grad_fn=<AddmmBackward>)
  4308. landmarks are: tensor([[[0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  4309. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  4310. [0.6038, 0.3946, 0.8413, 0.4883, 0.3562, 0.4550, 0.5266, 0.4693],
  4311. [0.6257, 0.4034, 0.8288, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  4312. [0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500],
  4313. [0.6142, 0.4127, 0.7575, 0.3067, 0.3438, 0.4383, 0.5778, 0.5207],
  4314. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  4315. [0.0000, 0.0000, 0.7192, 0.2346, 0.4038, 0.2050, 0.5138, 0.5650]]],
  4316. device='cuda:0')
  4317. loss_train_step before backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  4318. loss_train_step after backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  4319. loss_train: 0.14336391224060208
  4320. step: 36
  4321. running loss: 0.00398233089557228
  4322. Train Steps: 36/90 Loss: 0.0040 torch.Size([8, 600, 800])
  4323. torch.Size([8, 8])
  4324. tensor([[0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  4325. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  4326. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  4327. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  4328. [0.6081, 0.3950, 0.8538, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954],
  4329. [0.6329, 0.4196, 0.9238, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748],
  4330. [0.6240, 0.4217, 0.8150, 0.3133, 0.4425, 0.2650, 0.5650, 0.5817],
  4331. [0.6177, 0.4086, 0.8738, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700]],
  4332. device='cuda:0', dtype=torch.float64)
  4333. predictions are: tensor([[0.7228, 0.4681, 0.7525, 0.3056, 0.3635, 0.3611, 0.5924, 0.4883],
  4334. [0.6475, 0.4415, 0.7789, 0.4468, 0.3754, 0.3749, 0.5925, 0.5845],
  4335. [0.5587, 0.3827, 0.8006, 0.4904, 0.4415, 0.3936, 0.5719, 0.4867],
  4336. [0.4817, 0.3434, 0.8249, 0.4164, 0.4155, 0.4739, 0.6631, 0.5691],
  4337. [0.5975, 0.3980, 0.8270, 0.3494, 0.3865, 0.3966, 0.5595, 0.4871],
  4338. [0.5687, 0.3749, 0.8677, 0.3652, 0.4322, 0.1858, 0.6998, 0.4778],
  4339. [0.4849, 0.3461, 0.7653, 0.2249, 0.4421, 0.1733, 0.6068, 0.5384],
  4340. [0.5811, 0.3800, 0.8053, 0.2997, 0.3964, 0.4467, 0.6727, 0.5288]],
  4341. device='cuda:0', grad_fn=<AddmmBackward>)
  4342. landmarks are: tensor([[[0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  4343. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  4344. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  4345. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  4346. [0.6081, 0.3950, 0.8537, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954],
  4347. [0.6329, 0.4196, 0.9237, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748],
  4348. [0.6240, 0.4217, 0.8150, 0.3133, 0.4425, 0.2650, 0.5650, 0.5817],
  4349. [0.6177, 0.4085, 0.8737, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700]]],
  4350. device='cuda:0')
  4351. loss_train_step before backward: tensor(0.0046, device='cuda:0', grad_fn=<MseLossBackward>)
  4352. loss_train_step after backward: tensor(0.0046, device='cuda:0', grad_fn=<MseLossBackward>)
  4353. loss_train: 0.14793575985822827
  4354. step: 37
  4355. running loss: 0.003998263779952115
  4356. Train Steps: 37/90 Loss: 0.0040 torch.Size([8, 600, 800])
  4357. torch.Size([8, 8])
  4358. tensor([[0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  4359. [0.6250, 0.4131, 0.8688, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  4360. [0.6263, 0.4039, 0.9000, 0.4400, 0.3625, 0.4667, 0.6424, 0.4804],
  4361. [0.6170, 0.4102, 0.7468, 0.3695, 0.3463, 0.3767, 0.5238, 0.5823],
  4362. [0.6201, 0.4017, 0.8871, 0.4621, 0.3517, 0.4675, 0.5999, 0.5106],
  4363. [0.6273, 0.4110, 0.8900, 0.3817, 0.4188, 0.2167, 0.5858, 0.4835],
  4364. [0.6143, 0.4034, 0.8800, 0.4833, 0.4512, 0.5367, 0.5289, 0.5097],
  4365. [0.6205, 0.4062, 0.8337, 0.2683, 0.3675, 0.4283, 0.6338, 0.5250]],
  4366. device='cuda:0', dtype=torch.float64)
  4367. predictions are: tensor([[0.5927, 0.3793, 0.8371, 0.4345, 0.4220, 0.4288, 0.6075, 0.5332],
  4368. [0.6252, 0.3783, 0.8342, 0.2596, 0.4505, 0.2143, 0.6791, 0.5136],
  4369. [0.6643, 0.4283, 0.8630, 0.4127, 0.3536, 0.4288, 0.6292, 0.4873],
  4370. [0.6172, 0.4089, 0.7182, 0.3220, 0.3551, 0.3615, 0.5789, 0.5728],
  4371. [0.6835, 0.4250, 0.7887, 0.4249, 0.3495, 0.4700, 0.5967, 0.5164],
  4372. [0.6390, 0.4108, 0.8453, 0.3511, 0.4372, 0.1673, 0.6323, 0.4891],
  4373. [0.5556, 0.3422, 0.8279, 0.4370, 0.4475, 0.4216, 0.5880, 0.5037],
  4374. [0.6281, 0.4205, 0.7479, 0.2551, 0.3832, 0.3778, 0.6559, 0.5277]],
  4375. device='cuda:0', grad_fn=<AddmmBackward>)
  4376. landmarks are: tensor([[[0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  4377. [0.6250, 0.4131, 0.8687, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  4378. [0.6263, 0.4038, 0.9000, 0.4400, 0.3625, 0.4667, 0.6424, 0.4804],
  4379. [0.6170, 0.4102, 0.7468, 0.3695, 0.3462, 0.3767, 0.5238, 0.5823],
  4380. [0.6201, 0.4017, 0.8871, 0.4621, 0.3517, 0.4675, 0.5999, 0.5106],
  4381. [0.6273, 0.4110, 0.8900, 0.3817, 0.4187, 0.2167, 0.5858, 0.4835],
  4382. [0.6143, 0.4034, 0.8800, 0.4833, 0.4512, 0.5367, 0.5289, 0.5097],
  4383. [0.6205, 0.4062, 0.8338, 0.2683, 0.3675, 0.4283, 0.6338, 0.5250]]],
  4384. device='cuda:0')
  4385. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  4386. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  4387. loss_train: 0.1494324274826795
  4388. step: 38
  4389. running loss: 0.003932432302175776
  4390.  
  4391. Train Steps: 38/90 Loss: 0.0039 torch.Size([8, 600, 800])
  4392. torch.Size([8, 8])
  4393. tensor([[0.6058, 0.3986, 0.8324, 0.4626, 0.3838, 0.4983, 0.5147, 0.5466],
  4394. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864],
  4395. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  4396. [0.6182, 0.4058, 0.8738, 0.4350, 0.3563, 0.3400, 0.5290, 0.5822],
  4397. [0.6034, 0.4011, 0.7350, 0.2533, 0.3438, 0.3367, 0.5516, 0.5084],
  4398. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  4399. [0.6098, 0.3991, 0.8638, 0.4717, 0.4263, 0.4967, 0.5212, 0.5650],
  4400. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235]],
  4401. device='cuda:0', dtype=torch.float64)
  4402. predictions are: tensor([[0.6082, 0.3901, 0.8104, 0.4446, 0.3821, 0.4699, 0.5674, 0.5060],
  4403. [0.6610, 0.4154, 0.8537, 0.5049, 0.3697, 0.4719, 0.6829, 0.5151],
  4404. [0.6313, 0.4098, 0.8679, 0.4890, 0.4500, 0.4796, 0.6458, 0.4985],
  4405. [0.6464, 0.4064, 0.8509, 0.4027, 0.3388, 0.3724, 0.5377, 0.5235],
  4406. [0.7227, 0.4560, 0.6997, 0.2580, 0.3332, 0.3190, 0.5898, 0.4940],
  4407. [0.6985, 0.4387, 0.7837, 0.2486, 0.4359, 0.2209, 0.6642, 0.5135],
  4408. [0.6163, 0.3792, 0.8439, 0.4967, 0.4222, 0.4776, 0.5871, 0.5058],
  4409. [0.6270, 0.3982, 0.8587, 0.4695, 0.4115, 0.5075, 0.6632, 0.5138]],
  4410. device='cuda:0', grad_fn=<AddmmBackward>)
  4411. landmarks are: tensor([[[0.6058, 0.3986, 0.8324, 0.4626, 0.3837, 0.4983, 0.5147, 0.5466],
  4412. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864],
  4413. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  4414. [0.6182, 0.4058, 0.8737, 0.4350, 0.3562, 0.3400, 0.5290, 0.5822],
  4415. [0.6033, 0.4011, 0.7350, 0.2533, 0.3438, 0.3367, 0.5516, 0.5084],
  4416. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  4417. [0.6098, 0.3991, 0.8637, 0.4717, 0.4263, 0.4967, 0.5213, 0.5650],
  4418. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235]]],
  4419. device='cuda:0')
  4420. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  4421. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  4422. loss_train: 0.1505707015749067
  4423. step: 39
  4424. running loss: 0.0038607872198694027
  4425. Train Steps: 39/90 Loss: 0.0039 torch.Size([8, 600, 800])
  4426. torch.Size([8, 8])
  4427. tensor([[0.6219, 0.4097, 0.8738, 0.3400, 0.3563, 0.4117, 0.5975, 0.5683],
  4428. [0.6166, 0.4008, 0.8563, 0.5667, 0.4388, 0.4933, 0.5575, 0.5567],
  4429. [0.6198, 0.4115, 0.7762, 0.2717, 0.3713, 0.3200, 0.5837, 0.5683],
  4430. [0.6257, 0.4034, 0.8287, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  4431. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  4432. [0.6257, 0.4167, 0.8775, 0.3433, 0.3563, 0.4133, 0.6200, 0.5667],
  4433. [0.6307, 0.4029, 0.8988, 0.4817, 0.3937, 0.3500, 0.7311, 0.5378],
  4434. [0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915]],
  4435. device='cuda:0', dtype=torch.float64)
  4436. predictions are: tensor([[0.6763, 0.4203, 0.8212, 0.3882, 0.3392, 0.4437, 0.5854, 0.5210],
  4437. [0.5837, 0.3779, 0.8425, 0.5387, 0.4452, 0.5220, 0.5871, 0.5394],
  4438. [0.6857, 0.4187, 0.7855, 0.3141, 0.3728, 0.3482, 0.5546, 0.5442],
  4439. [0.6829, 0.4323, 0.7807, 0.3014, 0.3951, 0.3038, 0.6372, 0.4918],
  4440. [0.6924, 0.4353, 0.8441, 0.4242, 0.3433, 0.4517, 0.5365, 0.5052],
  4441. [0.6510, 0.3992, 0.8452, 0.3819, 0.3662, 0.4419, 0.6248, 0.5265],
  4442. [0.6469, 0.4096, 0.8938, 0.4680, 0.4248, 0.3783, 0.6646, 0.4929],
  4443. [0.7247, 0.4555, 0.8720, 0.4785, 0.3858, 0.3604, 0.6007, 0.4724]],
  4444. device='cuda:0', grad_fn=<AddmmBackward>)
  4445. landmarks are: tensor([[[0.6219, 0.4097, 0.8737, 0.3400, 0.3562, 0.4117, 0.5975, 0.5683],
  4446. [0.6166, 0.4008, 0.8562, 0.5667, 0.4387, 0.4933, 0.5575, 0.5567],
  4447. [0.6198, 0.4115, 0.7763, 0.2717, 0.3713, 0.3200, 0.5838, 0.5683],
  4448. [0.6257, 0.4034, 0.8288, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  4449. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  4450. [0.6257, 0.4167, 0.8775, 0.3433, 0.3562, 0.4133, 0.6200, 0.5667],
  4451. [0.6307, 0.4029, 0.8988, 0.4817, 0.3938, 0.3500, 0.7311, 0.5378],
  4452. [0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915]]],
  4453. device='cuda:0')
  4454. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  4455. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  4456. loss_train: 0.15187458659056574
  4457. step: 40
  4458. running loss: 0.0037968646647641435
  4459. Train Steps: 40/90 Loss: 0.0038 torch.Size([8, 600, 800])
  4460. torch.Size([8, 8])
  4461. tensor([[0.6201, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  4462. [0.6108, 0.4011, 0.8037, 0.3400, 0.3700, 0.2933, 0.5658, 0.5617],
  4463. [0.6148, 0.3996, 0.8488, 0.3867, 0.3488, 0.4067, 0.5863, 0.5000],
  4464. [0.6179, 0.4082, 0.6688, 0.2667, 0.3588, 0.3317, 0.5750, 0.5783],
  4465. [0.6261, 0.3987, 0.8688, 0.4917, 0.4300, 0.5333, 0.7010, 0.5309],
  4466. [0.6317, 0.4038, 0.8287, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  4467. [ nan, nan, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  4468. [0.6055, 0.4015, 0.7425, 0.2033, 0.4113, 0.1883, 0.5217, 0.4823]],
  4469. device='cuda:0', dtype=torch.float64)
  4470. predictions are: tensor([[0.7187, 0.4657, 0.8306, 0.2767, 0.4167, 0.2786, 0.6047, 0.5362],
  4471. [0.6980, 0.4498, 0.8665, 0.3995, 0.3762, 0.4197, 0.5474, 0.5448],
  4472. [0.6695, 0.4380, 0.9158, 0.4737, 0.3652, 0.5028, 0.6096, 0.5110],
  4473. [0.7887, 0.4816, 0.7583, 0.3533, 0.3352, 0.4078, 0.5599, 0.5510],
  4474. [0.6274, 0.4131, 0.9500, 0.5838, 0.4386, 0.6303, 0.6465, 0.5380],
  4475. [0.7863, 0.5082, 0.9138, 0.6277, 0.3542, 0.5731, 0.5936, 0.5166],
  4476. [0.5299, 0.3224, 0.9347, 0.3950, 0.4768, 0.3456, 0.6791, 0.5457],
  4477. [0.6023, 0.3797, 0.7298, 0.2535, 0.3751, 0.2667, 0.5376, 0.5283]],
  4478. device='cuda:0', grad_fn=<AddmmBackward>)
  4479. landmarks are: tensor([[[0.6202, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  4480. [0.6108, 0.4011, 0.8037, 0.3400, 0.3700, 0.2933, 0.5658, 0.5617],
  4481. [0.6148, 0.3996, 0.8487, 0.3867, 0.3487, 0.4067, 0.5863, 0.5000],
  4482. [0.6179, 0.4082, 0.6687, 0.2667, 0.3587, 0.3317, 0.5750, 0.5783],
  4483. [0.6261, 0.3987, 0.8687, 0.4917, 0.4300, 0.5333, 0.7010, 0.5309],
  4484. [0.6317, 0.4038, 0.8288, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  4485. [0.0000, 0.0000, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  4486. [0.6055, 0.4015, 0.7425, 0.2033, 0.4112, 0.1883, 0.5217, 0.4823]]],
  4487. device='cuda:0')
  4488. loss_train_step before backward: tensor(0.0102, device='cuda:0', grad_fn=<MseLossBackward>)
  4489. loss_train_step after backward: tensor(0.0102, device='cuda:0', grad_fn=<MseLossBackward>)
  4490. loss_train: 0.16205621568951756
  4491. step: 41
  4492. running loss: 0.003952590626573599
  4493. Train Steps: 41/90 Loss: 0.0040 torch.Size([8, 600, 800])
  4494. torch.Size([8, 8])
  4495. tensor([[0.6092, 0.4001, 0.8638, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  4496. [0.6263, 0.4065, 0.9038, 0.4317, 0.3588, 0.4550, 0.6325, 0.5250],
  4497. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6188, 0.5283],
  4498. [0.6147, 0.4107, 0.8137, 0.3333, 0.3750, 0.2683, 0.5006, 0.5412],
  4499. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  4500. [0.6109, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  4501. [0.6143, 0.4040, 0.8237, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  4502. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284]],
  4503. device='cuda:0', dtype=torch.float64)
  4504. predictions are: tensor([[0.6168, 0.3942, 0.8755, 0.5418, 0.4090, 0.6057, 0.5533, 0.5243],
  4505. [0.6289, 0.3965, 0.9262, 0.4993, 0.3608, 0.5306, 0.6536, 0.5308],
  4506. [0.7041, 0.4402, 0.9332, 0.4216, 0.4070, 0.3354, 0.6179, 0.5408],
  4507. [0.5029, 0.3301, 0.8102, 0.3966, 0.3733, 0.3770, 0.5184, 0.5792],
  4508. [0.6606, 0.4101, 0.8494, 0.2925, 0.4726, 0.2737, 0.6511, 0.5564],
  4509. [0.7221, 0.4742, 0.8928, 0.5254, 0.3563, 0.5033, 0.5423, 0.5139],
  4510. [0.5314, 0.3336, 0.8073, 0.3495, 0.4271, 0.3116, 0.5455, 0.5472],
  4511. [0.6883, 0.4347, 0.7809, 0.2945, 0.4331, 0.2745, 0.5999, 0.5528]],
  4512. device='cuda:0', grad_fn=<AddmmBackward>)
  4513. landmarks are: tensor([[[0.6092, 0.4001, 0.8637, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  4514. [0.6263, 0.4065, 0.9038, 0.4317, 0.3587, 0.4550, 0.6325, 0.5250],
  4515. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6187, 0.5283],
  4516. [0.6147, 0.4107, 0.8138, 0.3333, 0.3750, 0.2683, 0.5006, 0.5412],
  4517. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  4518. [0.6108, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  4519. [0.6143, 0.4040, 0.8238, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  4520. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284]]],
  4521. device='cuda:0')
  4522. loss_train_step before backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  4523. loss_train_step after backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  4524. loss_train: 0.16464617906603962
  4525. step: 42
  4526. running loss: 0.003920147120619991
  4527.  
  4528. Train Steps: 42/90 Loss: 0.0039 torch.Size([8, 600, 800])
  4529. torch.Size([8, 8])
  4530. tensor([[0.6161, 0.4024, 0.8662, 0.4683, 0.4935, 0.5364, 0.6063, 0.5567],
  4531. [0.6260, 0.4120, 0.8013, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895],
  4532. [0.6307, 0.4029, 0.8988, 0.4817, 0.3937, 0.3500, 0.7311, 0.5378],
  4533. [ nan, nan, 0.7225, 0.2167, 0.3987, 0.2283, 0.5427, 0.5181],
  4534. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5413, 0.5717],
  4535. [0.6321, 0.4048, 0.8738, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  4536. [0.6277, 0.4057, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  4537. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116]],
  4538. device='cuda:0', dtype=torch.float64)
  4539. predictions are: tensor([[0.5914, 0.3750, 0.9267, 0.4842, 0.4291, 0.5331, 0.5822, 0.5713],
  4540. [0.6727, 0.4229, 0.8425, 0.2422, 0.4783, 0.1994, 0.6245, 0.5530],
  4541. [0.6520, 0.4122, 0.9716, 0.4921, 0.4072, 0.3805, 0.6626, 0.5454],
  4542. [0.4037, 0.2400, 0.7607, 0.2332, 0.4031, 0.2527, 0.5299, 0.5721],
  4543. [0.6678, 0.4063, 0.9150, 0.5728, 0.4048, 0.5287, 0.5675, 0.5446],
  4544. [0.7441, 0.4790, 0.9230, 0.5637, 0.3632, 0.5035, 0.5689, 0.5169],
  4545. [0.6529, 0.4008, 0.8589, 0.2868, 0.4395, 0.2580, 0.6256, 0.5510],
  4546. [0.5707, 0.3659, 0.9312, 0.4682, 0.4123, 0.6192, 0.5800, 0.5642]],
  4547. device='cuda:0', grad_fn=<AddmmBackward>)
  4548. landmarks are: tensor([[[0.6161, 0.4024, 0.8662, 0.4683, 0.4935, 0.5364, 0.6062, 0.5567],
  4549. [0.6259, 0.4120, 0.8012, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895],
  4550. [0.6307, 0.4029, 0.8988, 0.4817, 0.3938, 0.3500, 0.7311, 0.5378],
  4551. [0.0000, 0.0000, 0.7225, 0.2167, 0.3988, 0.2283, 0.5427, 0.5181],
  4552. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5412, 0.5717],
  4553. [0.6321, 0.4048, 0.8737, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  4554. [0.6277, 0.4056, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  4555. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116]]],
  4556. device='cuda:0')
  4557. loss_train_step before backward: tensor(0.0052, device='cuda:0', grad_fn=<MseLossBackward>)
  4558. loss_train_step after backward: tensor(0.0052, device='cuda:0', grad_fn=<MseLossBackward>)
  4559. loss_train: 0.16981427988503128
  4560. step: 43
  4561. running loss: 0.00394916929965189
  4562. Train Steps: 43/90 Loss: 0.0039 torch.Size([8, 600, 800])
  4563. torch.Size([8, 8])
  4564. tensor([[0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200],
  4565. [0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  4566. [ nan, nan, 0.7412, 0.2200, 0.4450, 0.1517, 0.5312, 0.4983],
  4567. [0.6198, 0.4130, 0.8762, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  4568. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  4569. [0.6274, 0.4099, 0.8625, 0.3233, 0.4400, 0.1983, 0.5876, 0.4869],
  4570. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  4571. [0.6286, 0.4078, 0.8063, 0.2267, 0.4788, 0.1533, 0.5953, 0.4913]],
  4572. device='cuda:0', dtype=torch.float64)
  4573. predictions are: tensor([[0.5744, 0.3610, 0.8981, 0.3053, 0.4541, 0.2640, 0.6367, 0.5374],
  4574. [0.6513, 0.4209, 0.9170, 0.4760, 0.3727, 0.5281, 0.6176, 0.5163],
  4575. [0.3297, 0.2064, 0.7455, 0.2015, 0.4621, 0.1717, 0.5486, 0.5469],
  4576. [0.6995, 0.4713, 0.9346, 0.4540, 0.3628, 0.4868, 0.5914, 0.5479],
  4577. [0.5862, 0.3670, 0.9249, 0.5698, 0.4216, 0.4958, 0.5751, 0.5130],
  4578. [0.5996, 0.3651, 0.8751, 0.3118, 0.4596, 0.2157, 0.5981, 0.5374],
  4579. [0.6333, 0.4009, 0.9072, 0.5587, 0.4147, 0.4863, 0.5798, 0.5360],
  4580. [0.5015, 0.3147, 0.8200, 0.2512, 0.4896, 0.1724, 0.5748, 0.5407]],
  4581. device='cuda:0', grad_fn=<AddmmBackward>)
  4582. landmarks are: tensor([[[0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200],
  4583. [0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  4584. [0.0000, 0.0000, 0.7412, 0.2200, 0.4450, 0.1517, 0.5312, 0.4983],
  4585. [0.6198, 0.4130, 0.8763, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  4586. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  4587. [0.6274, 0.4099, 0.8625, 0.3233, 0.4400, 0.1983, 0.5876, 0.4869],
  4588. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  4589. [0.6286, 0.4078, 0.8062, 0.2267, 0.4787, 0.1533, 0.5953, 0.4913]]],
  4590. device='cuda:0')
  4591. loss_train_step before backward: tensor(0.0039, device='cuda:0', grad_fn=<MseLossBackward>)
  4592. loss_train_step after backward: tensor(0.0039, device='cuda:0', grad_fn=<MseLossBackward>)
  4593. loss_train: 0.1736702184425667
  4594. step: 44
  4595. running loss: 0.003947050419149243
  4596. Train Steps: 44/90 Loss: 0.0039 torch.Size([8, 600, 800])
  4597. torch.Size([8, 8])
  4598. tensor([[0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5988, 0.5700],
  4599. [0.6038, 0.3946, 0.8413, 0.4883, 0.3563, 0.4550, 0.5266, 0.4693],
  4600. [0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767],
  4601. [0.6249, 0.4142, 0.8350, 0.3283, 0.3613, 0.3700, 0.6188, 0.5400],
  4602. [0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500],
  4603. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  4604. [0.6286, 0.4060, 0.9188, 0.4333, 0.3675, 0.4167, 0.7034, 0.5528],
  4605. [0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6262, 0.5367]],
  4606. device='cuda:0', dtype=torch.float64)
  4607. predictions are: tensor([[0.4946, 0.3198, 0.8330, 0.2777, 0.4325, 0.2888, 0.5966, 0.5509],
  4608. [0.5412, 0.3576, 0.8926, 0.4817, 0.4078, 0.4237, 0.5483, 0.4943],
  4609. [0.5436, 0.3551, 0.8983, 0.5374, 0.4359, 0.4431, 0.5623, 0.5453],
  4610. [0.6110, 0.3883, 0.8518, 0.3097, 0.4324, 0.2751, 0.5913, 0.5272],
  4611. [0.5764, 0.3854, 0.8313, 0.2675, 0.3994, 0.3353, 0.5966, 0.5245],
  4612. [0.6086, 0.3816, 0.9137, 0.5039, 0.4031, 0.3933, 0.5720, 0.5005],
  4613. [0.5202, 0.3427, 0.9587, 0.4177, 0.4165, 0.3402, 0.6581, 0.5295],
  4614. [0.5557, 0.3643, 0.9287, 0.4716, 0.4642, 0.5385, 0.6326, 0.5063]],
  4615. device='cuda:0', grad_fn=<AddmmBackward>)
  4616. landmarks are: tensor([[[0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5987, 0.5700],
  4617. [0.6038, 0.3946, 0.8413, 0.4883, 0.3562, 0.4550, 0.5266, 0.4693],
  4618. [0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767],
  4619. [0.6249, 0.4142, 0.8350, 0.3283, 0.3613, 0.3700, 0.6187, 0.5400],
  4620. [0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500],
  4621. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  4622. [0.6286, 0.4060, 0.9187, 0.4333, 0.3675, 0.4167, 0.7034, 0.5528],
  4623. [0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6263, 0.5367]]],
  4624. device='cuda:0')
  4625. loss_train_step before backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  4626. loss_train_step after backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  4627. loss_train: 0.17585839342791587
  4628. step: 45
  4629. running loss: 0.003907964298398131
  4630. Train Steps: 45/90 Loss: 0.0039 torch.Size([8, 600, 800])
  4631. torch.Size([8, 8])
  4632. tensor([[0.6224, 0.4061, 0.8988, 0.4300, 0.3838, 0.4750, 0.6112, 0.5483],
  4633. [0.6339, 0.4102, 0.8588, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089],
  4634. [0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  4635. [0.6107, 0.4013, 0.8700, 0.4650, 0.5049, 0.5176, 0.5850, 0.5550],
  4636. [0.6152, 0.4131, 0.6863, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  4637. [0.6162, 0.4134, 0.6700, 0.2467, 0.3962, 0.2533, 0.5737, 0.5467],
  4638. [0.6273, 0.4143, 0.8750, 0.5700, 0.3987, 0.4717, 0.6013, 0.5467],
  4639. [0.6159, 0.4085, 0.6900, 0.2283, 0.4088, 0.1950, 0.5123, 0.5397]],
  4640. device='cuda:0', dtype=torch.float64)
  4641. predictions are: tensor([[0.5982, 0.3846, 0.9121, 0.4259, 0.3752, 0.5248, 0.6178, 0.4911],
  4642. [0.5487, 0.3555, 0.8807, 0.2899, 0.4626, 0.1906, 0.6620, 0.4942],
  4643. [0.5193, 0.3321, 0.8960, 0.4542, 0.4383, 0.4488, 0.5616, 0.5277],
  4644. [0.4869, 0.3403, 0.9145, 0.4456, 0.4816, 0.4387, 0.6093, 0.5287],
  4645. [0.5725, 0.3812, 0.7642, 0.2518, 0.3758, 0.2768, 0.5590, 0.5411],
  4646. [0.5903, 0.3854, 0.7478, 0.2506, 0.3907, 0.2212, 0.5636, 0.5075],
  4647. [0.5732, 0.3680, 0.8950, 0.5185, 0.4251, 0.4477, 0.5844, 0.5426],
  4648. [0.3919, 0.2581, 0.7479, 0.2186, 0.4290, 0.1590, 0.5343, 0.5263]],
  4649. device='cuda:0', grad_fn=<AddmmBackward>)
  4650. landmarks are: tensor([[[0.6224, 0.4061, 0.8988, 0.4300, 0.3837, 0.4750, 0.6112, 0.5483],
  4651. [0.6339, 0.4102, 0.8587, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089],
  4652. [0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  4653. [0.6107, 0.4013, 0.8700, 0.4650, 0.5049, 0.5176, 0.5850, 0.5550],
  4654. [0.6152, 0.4131, 0.6862, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  4655. [0.6162, 0.4134, 0.6700, 0.2467, 0.3963, 0.2533, 0.5738, 0.5467],
  4656. [0.6273, 0.4143, 0.8750, 0.5700, 0.3988, 0.4717, 0.6012, 0.5467],
  4657. [0.6159, 0.4085, 0.6900, 0.2283, 0.4087, 0.1950, 0.5123, 0.5397]]],
  4658. device='cuda:0')
  4659. loss_train_step before backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  4660. loss_train_step after backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  4661. loss_train: 0.1786146817030385
  4662. step: 46
  4663. running loss: 0.003882927863109533
  4664.  
  4665. Train Steps: 46/90 Loss: 0.0039 torch.Size([8, 600, 800])
  4666. torch.Size([8, 8])
  4667. tensor([[0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  4668. [0.6364, 0.4165, 0.9088, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  4669. [0.6275, 0.4024, 0.8600, 0.2283, 0.5350, 0.1800, 0.7074, 0.5413],
  4670. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373],
  4671. [ nan, nan, 0.8938, 0.2850, 0.4662, 0.3117, 0.7406, 0.5528],
  4672. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  4673. [0.6223, 0.4028, 0.8988, 0.4200, 0.3763, 0.5733, 0.6375, 0.5167],
  4674. [0.6183, 0.4076, 0.8838, 0.4517, 0.3813, 0.4483, 0.5775, 0.5633]],
  4675. device='cuda:0', dtype=torch.float64)
  4676. predictions are: tensor([[0.5902, 0.4247, 0.8342, 0.4375, 0.4049, 0.2877, 0.5761, 0.4922],
  4677. [0.5506, 0.3890, 0.8556, 0.3736, 0.4278, 0.2765, 0.5867, 0.5182],
  4678. [0.4517, 0.3155, 0.8016, 0.1960, 0.5234, 0.2164, 0.6464, 0.5249],
  4679. [0.5782, 0.3937, 0.7643, 0.4926, 0.3738, 0.4246, 0.5473, 0.5271],
  4680. [0.3472, 0.2472, 0.8567, 0.2544, 0.4739, 0.2547, 0.6857, 0.5190],
  4681. [0.5657, 0.3600, 0.8437, 0.4191, 0.3642, 0.3413, 0.5603, 0.4914],
  4682. [0.5481, 0.3568, 0.8340, 0.3782, 0.3921, 0.5499, 0.6102, 0.4830],
  4683. [0.5766, 0.3872, 0.8331, 0.4085, 0.3848, 0.4229, 0.5482, 0.5033]],
  4684. device='cuda:0', grad_fn=<AddmmBackward>)
  4685. landmarks are: tensor([[[0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  4686. [0.6364, 0.4165, 0.9087, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  4687. [0.6275, 0.4024, 0.8600, 0.2283, 0.5350, 0.1800, 0.7074, 0.5413],
  4688. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373],
  4689. [0.0000, 0.0000, 0.8938, 0.2850, 0.4663, 0.3117, 0.7406, 0.5528],
  4690. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  4691. [0.6223, 0.4028, 0.8988, 0.4200, 0.3762, 0.5733, 0.6375, 0.5167],
  4692. [0.6183, 0.4076, 0.8838, 0.4517, 0.3812, 0.4483, 0.5775, 0.5633]]],
  4693. device='cuda:0')
  4694. loss_train_step before backward: tensor(0.0054, device='cuda:0', grad_fn=<MseLossBackward>)
  4695. loss_train_step after backward: tensor(0.0054, device='cuda:0', grad_fn=<MseLossBackward>)
  4696. loss_train: 0.18397039466071874
  4697. step: 47
  4698. running loss: 0.003914263716185505
  4699. Train Steps: 47/90 Loss: 0.0039 torch.Size([8, 600, 800])
  4700. torch.Size([8, 8])
  4701. tensor([[0.6128, 0.4022, 0.8738, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  4702. [0.6357, 0.4097, 0.9038, 0.3883, 0.4213, 0.2950, 0.6686, 0.5390],
  4703. [0.6248, 0.4032, 0.7738, 0.1900, 0.4813, 0.1400, 0.5941, 0.4904],
  4704. [0.6132, 0.4118, 0.8200, 0.3633, 0.3563, 0.5400, 0.5787, 0.5136],
  4705. [0.6231, 0.3973, 0.8650, 0.3950, 0.3625, 0.3183, 0.5837, 0.5167],
  4706. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  4707. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  4708. [0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274]],
  4709. device='cuda:0', dtype=torch.float64)
  4710. predictions are: tensor([[0.5356, 0.3720, 0.8169, 0.4604, 0.4491, 0.4540, 0.5478, 0.5239],
  4711. [0.5339, 0.3525, 0.8122, 0.3496, 0.3886, 0.2890, 0.6373, 0.5348],
  4712. [0.4764, 0.3318, 0.7208, 0.1869, 0.4643, 0.1142, 0.5922, 0.5236],
  4713. [0.5076, 0.3286, 0.7992, 0.3146, 0.3590, 0.5195, 0.6077, 0.5179],
  4714. [0.5510, 0.3645, 0.8016, 0.3514, 0.3689, 0.3138, 0.5630, 0.5140],
  4715. [0.5081, 0.3545, 0.8147, 0.3891, 0.3914, 0.3503, 0.6489, 0.5292],
  4716. [0.6319, 0.4548, 0.8447, 0.4583, 0.3428, 0.4245, 0.6191, 0.5207],
  4717. [0.5233, 0.3743, 0.8006, 0.2742, 0.4410, 0.2303, 0.6163, 0.5454]],
  4718. device='cuda:0', grad_fn=<AddmmBackward>)
  4719. landmarks are: tensor([[[0.6128, 0.4022, 0.8737, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  4720. [0.6357, 0.4097, 0.9038, 0.3883, 0.4212, 0.2950, 0.6686, 0.5390],
  4721. [0.6248, 0.4032, 0.7738, 0.1900, 0.4812, 0.1400, 0.5941, 0.4904],
  4722. [0.6132, 0.4118, 0.8200, 0.3633, 0.3562, 0.5400, 0.5787, 0.5136],
  4723. [0.6231, 0.3973, 0.8650, 0.3950, 0.3625, 0.3183, 0.5838, 0.5167],
  4724. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  4725. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  4726. [0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274]]],
  4727. device='cuda:0')
  4728. loss_train_step before backward: tensor(0.0029, device='cuda:0', grad_fn=<MseLossBackward>)
  4729. loss_train_step after backward: tensor(0.0029, device='cuda:0', grad_fn=<MseLossBackward>)
  4730. loss_train: 0.18689397221896797
  4731. step: 48
  4732. running loss: 0.0038936244212284996
  4733. Train Steps: 48/90 Loss: 0.0039 torch.Size([8, 600, 800])
  4734. torch.Size([8, 8])
  4735. tensor([[0.6175, 0.4093, 0.8800, 0.4433, 0.4075, 0.4367, 0.5128, 0.4901],
  4736. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290],
  4737. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  4738. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  4739. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  4740. [0.6250, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6088, 0.5183],
  4741. [0.6229, 0.4198, 0.7662, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783],
  4742. [0.6131, 0.4037, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680]],
  4743. device='cuda:0', dtype=torch.float64)
  4744. predictions are: tensor([[0.5783, 0.3915, 0.8595, 0.4291, 0.3632, 0.4066, 0.5616, 0.5293],
  4745. [0.5297, 0.3724, 0.7324, 0.2156, 0.4202, 0.1592, 0.6418, 0.5288],
  4746. [0.6896, 0.4471, 0.8466, 0.3719, 0.3241, 0.4163, 0.6480, 0.5045],
  4747. [0.5473, 0.3957, 0.8311, 0.4912, 0.4214, 0.5023, 0.5809, 0.5467],
  4748. [0.6278, 0.4192, 0.8578, 0.5170, 0.3710, 0.4178, 0.6087, 0.5449],
  4749. [0.5861, 0.3984, 0.8378, 0.4504, 0.4305, 0.5174, 0.6835, 0.5589],
  4750. [0.4583, 0.3411, 0.7667, 0.2237, 0.4115, 0.2177, 0.6164, 0.5766],
  4751. [0.5240, 0.3749, 0.6977, 0.2363, 0.3559, 0.2672, 0.5716, 0.5624]],
  4752. device='cuda:0', grad_fn=<AddmmBackward>)
  4753. landmarks are: tensor([[[0.6175, 0.4093, 0.8800, 0.4433, 0.4075, 0.4367, 0.5128, 0.4901],
  4754. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290],
  4755. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  4756. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  4757. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  4758. [0.6251, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6087, 0.5183],
  4759. [0.6229, 0.4198, 0.7663, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783],
  4760. [0.6131, 0.4036, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680]]],
  4761. device='cuda:0')
  4762. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  4763. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  4764. loss_train: 0.1887947202194482
  4765. step: 49
  4766. running loss: 0.00385295347386629
  4767. Train Steps: 49/90 Loss: 0.0039 torch.Size([8, 600, 800])
  4768. torch.Size([8, 8])
  4769. tensor([[0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  4770. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901],
  4771. [0.6304, 0.4024, 0.8925, 0.4800, 0.3937, 0.4817, 0.7485, 0.5297],
  4772. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5787, 0.5117],
  4773. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  4774. [0.6227, 0.4083, 0.8938, 0.4800, 0.3800, 0.2950, 0.5737, 0.5350],
  4775. [0.6200, 0.3978, 0.8900, 0.4550, 0.3775, 0.5200, 0.6150, 0.5367],
  4776. [0.6226, 0.4103, 0.8575, 0.3450, 0.4388, 0.2067, 0.5787, 0.5383]],
  4777. device='cuda:0', dtype=torch.float64)
  4778. predictions are: tensor([[0.5907, 0.4206, 0.8311, 0.4863, 0.3859, 0.4446, 0.6255, 0.5465],
  4779. [0.4468, 0.3266, 0.7506, 0.2761, 0.3689, 0.2431, 0.5742, 0.5514],
  4780. [0.5799, 0.3804, 0.8372, 0.4546, 0.3878, 0.4680, 0.6465, 0.5339],
  4781. [0.6617, 0.4215, 0.7334, 0.2241, 0.4029, 0.1739, 0.6141, 0.5315],
  4782. [0.6885, 0.4408, 0.8412, 0.3858, 0.3437, 0.4316, 0.6301, 0.5134],
  4783. [0.6789, 0.4871, 0.8510, 0.4682, 0.3937, 0.3102, 0.6101, 0.5566],
  4784. [0.5780, 0.3749, 0.8523, 0.4565, 0.3673, 0.5337, 0.6059, 0.5508],
  4785. [0.5551, 0.4059, 0.8128, 0.3384, 0.4452, 0.2528, 0.6119, 0.5846]],
  4786. device='cuda:0', grad_fn=<AddmmBackward>)
  4787. landmarks are: tensor([[[0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  4788. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901],
  4789. [0.6304, 0.4024, 0.8925, 0.4800, 0.3938, 0.4817, 0.7485, 0.5297],
  4790. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5788, 0.5117],
  4791. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  4792. [0.6227, 0.4083, 0.8938, 0.4800, 0.3800, 0.2950, 0.5738, 0.5350],
  4793. [0.6199, 0.3978, 0.8900, 0.4550, 0.3775, 0.5200, 0.6150, 0.5367],
  4794. [0.6226, 0.4103, 0.8575, 0.3450, 0.4387, 0.2067, 0.5788, 0.5383]]],
  4795. device='cuda:0')
  4796. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  4797. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  4798. loss_train: 0.19071150361560285
  4799. step: 50
  4800. running loss: 0.003814230072312057
  4801.  
  4802. Train Steps: 50/90 Loss: 0.0038 torch.Size([8, 600, 800])
  4803. torch.Size([8, 8])
  4804. tensor([[0.6268, 0.4052, 0.8175, 0.2250, 0.4688, 0.1917, 0.6375, 0.5267],
  4805. [0.6188, 0.4099, 0.7400, 0.2433, 0.3962, 0.2750, 0.6162, 0.5467],
  4806. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  4807. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783],
  4808. [0.6053, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  4809. [0.6257, 0.4034, 0.8287, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  4810. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  4811. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167]],
  4812. device='cuda:0', dtype=torch.float64)
  4813. predictions are: tensor([[0.6429, 0.4160, 0.8269, 0.2762, 0.4544, 0.2121, 0.6649, 0.5159],
  4814. [0.6870, 0.4637, 0.7694, 0.2872, 0.3812, 0.3079, 0.6157, 0.5234],
  4815. [0.3769, 0.2598, 0.7436, 0.2793, 0.4277, 0.2608, 0.5515, 0.5496],
  4816. [0.7011, 0.4604, 0.8506, 0.3975, 0.3754, 0.3619, 0.5322, 0.5614],
  4817. [0.5484, 0.3436, 0.7126, 0.2348, 0.4148, 0.2125, 0.5711, 0.4997],
  4818. [0.6928, 0.4471, 0.8586, 0.2866, 0.3946, 0.2889, 0.6364, 0.4866],
  4819. [0.7239, 0.4933, 0.8775, 0.5449, 0.4109, 0.5543, 0.5762, 0.5451],
  4820. [0.6751, 0.4653, 0.7992, 0.3384, 0.3962, 0.3202, 0.6100, 0.5912]],
  4821. device='cuda:0', grad_fn=<AddmmBackward>)
  4822. landmarks are: tensor([[[0.6268, 0.4052, 0.8175, 0.2250, 0.4688, 0.1917, 0.6375, 0.5267],
  4823. [0.6188, 0.4099, 0.7400, 0.2433, 0.3963, 0.2750, 0.6162, 0.5467],
  4824. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  4825. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783],
  4826. [0.6054, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  4827. [0.6257, 0.4034, 0.8288, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  4828. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  4829. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167]]],
  4830. device='cuda:0')
  4831. loss_train_step before backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  4832. loss_train_step after backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  4833. loss_train: 0.1935047539882362
  4834. step: 51
  4835. running loss: 0.003794210862514435
  4836. Train Steps: 51/90 Loss: 0.0038 torch.Size([8, 600, 800])
  4837. torch.Size([8, 8])
  4838. tensor([[0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  4839. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  4840. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650],
  4841. [ nan, nan, 0.7097, 0.2346, 0.4250, 0.1850, 0.5175, 0.5583],
  4842. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  4843. [0.6153, 0.4117, 0.8688, 0.5167, 0.4895, 0.5647, 0.5524, 0.5136],
  4844. [0.6212, 0.4033, 0.8938, 0.4167, 0.3813, 0.4267, 0.5613, 0.5583],
  4845. [0.6182, 0.4099, 0.7812, 0.3000, 0.3937, 0.2367, 0.5325, 0.5750]],
  4846. device='cuda:0', dtype=torch.float64)
  4847. predictions are: tensor([[0.7185, 0.4708, 0.8618, 0.5070, 0.4093, 0.3858, 0.6009, 0.5121],
  4848. [0.8135, 0.4977, 0.8833, 0.4586, 0.3953, 0.5580, 0.6702, 0.5256],
  4849. [0.7068, 0.4859, 0.8293, 0.3565, 0.3601, 0.2996, 0.6099, 0.5512],
  4850. [0.3108, 0.2245, 0.7162, 0.2091, 0.4363, 0.1829, 0.5488, 0.5426],
  4851. [0.7652, 0.4867, 0.8227, 0.5169, 0.3867, 0.4359, 0.5665, 0.5834],
  4852. [0.6891, 0.4566, 0.8543, 0.4532, 0.4490, 0.4629, 0.5988, 0.5428],
  4853. [0.6487, 0.4061, 0.9048, 0.4152, 0.3674, 0.4081, 0.5812, 0.5541],
  4854. [0.6153, 0.4052, 0.7894, 0.2775, 0.4123, 0.2320, 0.5595, 0.5471]],
  4855. device='cuda:0', grad_fn=<AddmmBackward>)
  4856. landmarks are: tensor([[[0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  4857. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  4858. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650],
  4859. [0.0000, 0.0000, 0.7097, 0.2346, 0.4250, 0.1850, 0.5175, 0.5583],
  4860. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  4861. [0.6154, 0.4117, 0.8687, 0.5167, 0.4895, 0.5647, 0.5524, 0.5136],
  4862. [0.6212, 0.4033, 0.8938, 0.4167, 0.3812, 0.4267, 0.5612, 0.5583],
  4863. [0.6182, 0.4099, 0.7812, 0.3000, 0.3938, 0.2367, 0.5325, 0.5750]]],
  4864. device='cuda:0')
  4865. loss_train_step before backward: tensor(0.0048, device='cuda:0', grad_fn=<MseLossBackward>)
  4866. loss_train_step after backward: tensor(0.0048, device='cuda:0', grad_fn=<MseLossBackward>)
  4867. loss_train: 0.19832616718485951
  4868. step: 52
  4869. running loss: 0.0038139647535549905
  4870. Train Steps: 52/90 Loss: 0.0038 torch.Size([8, 600, 800])
  4871. torch.Size([8, 8])
  4872. tensor([[0.6053, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  4873. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  4874. [0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5413, 0.5433],
  4875. [ nan, nan, 0.7612, 0.3250, 0.4037, 0.2533, 0.5438, 0.5767],
  4876. [0.6250, 0.4236, 0.8638, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  4877. [0.6160, 0.4093, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  4878. [0.6131, 0.4037, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  4879. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896]],
  4880. device='cuda:0', dtype=torch.float64)
  4881. predictions are: tensor([[0.6246, 0.3846, 0.7061, 0.2208, 0.4330, 0.1798, 0.5548, 0.5012],
  4882. [0.7532, 0.4474, 0.9285, 0.5140, 0.4293, 0.5499, 0.6269, 0.5106],
  4883. [0.6529, 0.4466, 0.8533, 0.3775, 0.3637, 0.3183, 0.5681, 0.5332],
  4884. [0.4785, 0.2866, 0.8067, 0.3179, 0.4354, 0.2605, 0.5388, 0.5638],
  4885. [0.7072, 0.4532, 0.8808, 0.4216, 0.4210, 0.3280, 0.5693, 0.5640],
  4886. [0.6810, 0.4414, 0.8535, 0.4714, 0.3980, 0.4581, 0.5754, 0.5703],
  4887. [0.6754, 0.4343, 0.7393, 0.2802, 0.3996, 0.2901, 0.5359, 0.5503],
  4888. [0.6758, 0.4444, 0.8548, 0.4092, 0.3865, 0.3137, 0.5375, 0.5456]],
  4889. device='cuda:0', grad_fn=<AddmmBackward>)
  4890. landmarks are: tensor([[[0.6054, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  4891. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  4892. [0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5412, 0.5433],
  4893. [0.0000, 0.0000, 0.7613, 0.3250, 0.4038, 0.2533, 0.5437, 0.5767],
  4894. [0.6250, 0.4236, 0.8637, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  4895. [0.6160, 0.4092, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  4896. [0.6131, 0.4036, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  4897. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896]]],
  4898. device='cuda:0')
  4899. loss_train_step before backward: tensor(0.0060, device='cuda:0', grad_fn=<MseLossBackward>)
  4900. loss_train_step after backward: tensor(0.0060, device='cuda:0', grad_fn=<MseLossBackward>)
  4901. loss_train: 0.2043646709062159
  4902. step: 53
  4903. running loss: 0.003855937186909734
  4904. Train Steps: 53/90 Loss: 0.0039 torch.Size([8, 600, 800])
  4905. torch.Size([8, 8])
  4906. tensor([[0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  4907. [0.6127, 0.4118, 0.8650, 0.5083, 0.4088, 0.5367, 0.5300, 0.5456],
  4908. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  4909. [0.6125, 0.3974, 0.7725, 0.2517, 0.3538, 0.3317, 0.5887, 0.5500],
  4910. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  4911. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  4912. [0.6222, 0.3957, 0.8838, 0.5017, 0.3937, 0.4600, 0.5900, 0.5017],
  4913. [0.6164, 0.4119, 0.7913, 0.2650, 0.3538, 0.3500, 0.5614, 0.5038]],
  4914. device='cuda:0', dtype=torch.float64)
  4915. predictions are: tensor([[0.6823, 0.4182, 0.8925, 0.5127, 0.3848, 0.3467, 0.5078, 0.5746],
  4916. [0.6324, 0.3610, 0.8778, 0.5452, 0.4411, 0.5536, 0.5033, 0.5969],
  4917. [0.6364, 0.3702, 0.8149, 0.3233, 0.4203, 0.2426, 0.5644, 0.5335],
  4918. [0.6659, 0.3918, 0.7969, 0.2866, 0.3929, 0.3467, 0.5385, 0.5664],
  4919. [0.6106, 0.3540, 0.8478, 0.4741, 0.3870, 0.4879, 0.5421, 0.5649],
  4920. [0.6444, 0.3753, 0.8584, 0.2748, 0.4906, 0.1783, 0.6106, 0.5320],
  4921. [0.6796, 0.4049, 0.9135, 0.5595, 0.4094, 0.4721, 0.5515, 0.5398],
  4922. [0.7185, 0.4458, 0.8179, 0.2925, 0.3749, 0.3738, 0.5725, 0.5191]],
  4923. device='cuda:0', grad_fn=<AddmmBackward>)
  4924. landmarks are: tensor([[[0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  4925. [0.6127, 0.4118, 0.8650, 0.5083, 0.4087, 0.5367, 0.5300, 0.5456],
  4926. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  4927. [0.6125, 0.3974, 0.7725, 0.2517, 0.3537, 0.3317, 0.5888, 0.5500],
  4928. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  4929. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  4930. [0.6222, 0.3957, 0.8838, 0.5017, 0.3938, 0.4600, 0.5900, 0.5017],
  4931. [0.6164, 0.4119, 0.7912, 0.2650, 0.3537, 0.3500, 0.5614, 0.5038]]],
  4932. device='cuda:0')
  4933. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  4934. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  4935. loss_train: 0.20535756752360612
  4936. step: 54
  4937. running loss: 0.003802917917103817
  4938.  
  4939. Train Steps: 54/90 Loss: 0.0038 torch.Size([8, 600, 800])
  4940. torch.Size([8, 8])
  4941. tensor([[0.6241, 0.4143, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  4942. [0.6226, 0.4098, 0.8912, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  4943. [0.6038, 0.3946, 0.8413, 0.4883, 0.3563, 0.4550, 0.5266, 0.4693],
  4944. [ nan, nan, 0.6469, 0.1943, 0.4025, 0.2000, 0.5125, 0.5533],
  4945. [0.6097, 0.3988, 0.8650, 0.5250, 0.4213, 0.5200, 0.5675, 0.5050],
  4946. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  4947. [0.6149, 0.4054, 0.6713, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  4948. [0.6274, 0.4270, 0.8938, 0.4967, 0.3550, 0.4283, 0.5700, 0.5733]],
  4949. device='cuda:0', dtype=torch.float64)
  4950. predictions are: tensor([[0.5917, 0.3723, 0.8891, 0.4751, 0.4165, 0.5489, 0.6182, 0.5760],
  4951. [0.7117, 0.4290, 0.8914, 0.4276, 0.4067, 0.2642, 0.5423, 0.5531],
  4952. [0.5924, 0.3717, 0.8861, 0.4923, 0.3642, 0.4655, 0.5262, 0.5317],
  4953. [0.2581, 0.1246, 0.7177, 0.2037, 0.4130, 0.1838, 0.5022, 0.5447],
  4954. [0.6910, 0.4236, 0.8848, 0.5404, 0.4288, 0.5179, 0.5153, 0.5370],
  4955. [0.7144, 0.4450, 0.8453, 0.4453, 0.4477, 0.3143, 0.5251, 0.5701],
  4956. [0.5903, 0.3559, 0.7124, 0.2420, 0.3795, 0.2064, 0.4764, 0.5346],
  4957. [0.6276, 0.3709, 0.8845, 0.5027, 0.3586, 0.4799, 0.5622, 0.5465]],
  4958. device='cuda:0', grad_fn=<AddmmBackward>)
  4959. landmarks are: tensor([[[0.6241, 0.4142, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  4960. [0.6226, 0.4098, 0.8913, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  4961. [0.6038, 0.3946, 0.8413, 0.4883, 0.3562, 0.4550, 0.5266, 0.4693],
  4962. [0.0000, 0.0000, 0.6469, 0.1943, 0.4025, 0.2000, 0.5125, 0.5533],
  4963. [0.6097, 0.3988, 0.8650, 0.5250, 0.4212, 0.5200, 0.5675, 0.5050],
  4964. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  4965. [0.6149, 0.4054, 0.6712, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  4966. [0.6274, 0.4270, 0.8938, 0.4967, 0.3550, 0.4283, 0.5700, 0.5733]]],
  4967. device='cuda:0')
  4968. loss_train_step before backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  4969. loss_train_step after backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  4970. loss_train: 0.20769805239979178
  4971. step: 55
  4972. running loss: 0.0037763282254507594
  4973. Train Steps: 55/90 Loss: 0.0038 torch.Size([8, 600, 800])
  4974. torch.Size([8, 8])
  4975. tensor([[0.6200, 0.3993, 0.8519, 0.4923, 0.3962, 0.4717, 0.6013, 0.5433],
  4976. [0.6200, 0.4098, 0.8237, 0.2917, 0.4012, 0.2967, 0.6000, 0.5683],
  4977. [0.6166, 0.4008, 0.8563, 0.5667, 0.4388, 0.4933, 0.5575, 0.5567],
  4978. [0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  4979. [0.6136, 0.4117, 0.8700, 0.5167, 0.4188, 0.5083, 0.5147, 0.5495],
  4980. [0.6198, 0.3997, 0.8582, 0.5361, 0.4117, 0.5016, 0.5942, 0.5134],
  4981. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  4982. [0.6147, 0.4112, 0.7988, 0.3200, 0.3775, 0.2767, 0.5150, 0.5550]],
  4983. device='cuda:0', dtype=torch.float64)
  4984. predictions are: tensor([[0.5744, 0.3695, 0.8922, 0.5000, 0.3746, 0.4588, 0.5700, 0.5711],
  4985. [0.6601, 0.4320, 0.8350, 0.3069, 0.3937, 0.2802, 0.5901, 0.5480],
  4986. [0.5936, 0.3623, 0.8516, 0.5368, 0.4231, 0.4702, 0.5124, 0.5805],
  4987. [0.5631, 0.3485, 0.7975, 0.2137, 0.4321, 0.2081, 0.6074, 0.5147],
  4988. [0.6031, 0.3717, 0.8932, 0.5464, 0.4023, 0.4963, 0.5162, 0.5645],
  4989. [0.5450, 0.3490, 0.8726, 0.5144, 0.4051, 0.4960, 0.5226, 0.5375],
  4990. [0.6051, 0.3576, 0.8927, 0.4888, 0.4008, 0.5298, 0.5622, 0.5188],
  4991. [0.4384, 0.3046, 0.7921, 0.3440, 0.3716, 0.2838, 0.4716, 0.5635]],
  4992. device='cuda:0', grad_fn=<AddmmBackward>)
  4993. landmarks are: tensor([[[0.6200, 0.3993, 0.8519, 0.4923, 0.3963, 0.4717, 0.6012, 0.5433],
  4994. [0.6200, 0.4098, 0.8238, 0.2917, 0.4013, 0.2967, 0.6000, 0.5683],
  4995. [0.6166, 0.4008, 0.8562, 0.5667, 0.4387, 0.4933, 0.5575, 0.5567],
  4996. [0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  4997. [0.6136, 0.4117, 0.8700, 0.5167, 0.4187, 0.5083, 0.5147, 0.5495],
  4998. [0.6198, 0.3997, 0.8582, 0.5361, 0.4117, 0.5016, 0.5942, 0.5134],
  4999. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  5000. [0.6147, 0.4112, 0.7987, 0.3200, 0.3775, 0.2767, 0.5150, 0.5550]]],
  5001. device='cuda:0')
  5002. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  5003. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  5004. loss_train: 0.20929075684398413
  5005. step: 56
  5006. running loss: 0.0037373349436425735
  5007. Train Steps: 56/90 Loss: 0.0037 torch.Size([8, 600, 800])
  5008. torch.Size([8, 8])
  5009. tensor([[0.6273, 0.4143, 0.8750, 0.5700, 0.3987, 0.4717, 0.6013, 0.5467],
  5010. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5413, 0.5717],
  5011. [0.6311, 0.4008, 0.7935, 0.5746, 0.3900, 0.5033, 0.6955, 0.5366],
  5012. [0.6179, 0.3998, 0.8396, 0.3505, 0.3552, 0.3768, 0.6064, 0.5329],
  5013. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  5014. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  5015. [0.6143, 0.4040, 0.8237, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  5016. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383]],
  5017. device='cuda:0', dtype=torch.float64)
  5018. predictions are: tensor([[0.5484, 0.3537, 0.8765, 0.5648, 0.4005, 0.4816, 0.5476, 0.5668],
  5019. [0.5897, 0.3917, 0.8650, 0.5099, 0.4313, 0.4779, 0.5345, 0.5749],
  5020. [0.5643, 0.3977, 0.8293, 0.5187, 0.4009, 0.4767, 0.5955, 0.5563],
  5021. [0.6404, 0.4238, 0.8493, 0.3751, 0.3361, 0.3527, 0.5680, 0.5031],
  5022. [0.5447, 0.3584, 0.8739, 0.5105, 0.4286, 0.4851, 0.5295, 0.5788],
  5023. [0.5144, 0.3560, 0.8684, 0.3694, 0.3730, 0.3466, 0.5790, 0.5290],
  5024. [0.4598, 0.3012, 0.8132, 0.3359, 0.4055, 0.2180, 0.5332, 0.5154],
  5025. [0.4954, 0.3271, 0.8524, 0.3360, 0.3561, 0.3671, 0.6126, 0.5512]],
  5026. device='cuda:0', grad_fn=<AddmmBackward>)
  5027. landmarks are: tensor([[[0.6273, 0.4143, 0.8750, 0.5700, 0.3988, 0.4717, 0.6012, 0.5467],
  5028. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5412, 0.5717],
  5029. [0.6311, 0.4008, 0.7935, 0.5746, 0.3900, 0.5033, 0.6955, 0.5366],
  5030. [0.6179, 0.3998, 0.8396, 0.3505, 0.3552, 0.3768, 0.6064, 0.5329],
  5031. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  5032. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  5033. [0.6143, 0.4040, 0.8238, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  5034. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383]]],
  5035. device='cuda:0')
  5036. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  5037. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  5038. loss_train: 0.2111712924670428
  5039. step: 57
  5040. running loss: 0.003704759516965663
  5041. Train Steps: 57/90 Loss: 0.0037 torch.Size([8, 600, 800])
  5042. torch.Size([8, 8])
  5043. tensor([[0.6229, 0.4107, 0.8137, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  5044. [0.6151, 0.4125, 0.8738, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  5045. [0.6090, 0.4010, 0.7838, 0.3483, 0.3538, 0.3783, 0.5462, 0.5077],
  5046. [0.6260, 0.4120, 0.8013, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895],
  5047. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  5048. [ nan, nan, 0.6412, 0.1900, 0.4238, 0.1883, 0.5487, 0.5700],
  5049. [0.6357, 0.4097, 0.9038, 0.3883, 0.4213, 0.2950, 0.6686, 0.5390],
  5050. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795]],
  5051. device='cuda:0', dtype=torch.float64)
  5052. predictions are: tensor([[0.5698, 0.3906, 0.7905, 0.3307, 0.4393, 0.2544, 0.5648, 0.5202],
  5053. [0.5052, 0.3361, 0.8728, 0.5273, 0.3377, 0.4399, 0.5142, 0.5462],
  5054. [0.5575, 0.3894, 0.8096, 0.3553, 0.3398, 0.4197, 0.5326, 0.4992],
  5055. [0.5576, 0.3766, 0.8028, 0.2573, 0.4620, 0.2038, 0.6238, 0.4937],
  5056. [0.5404, 0.3751, 0.8065, 0.2962, 0.4041, 0.3262, 0.5811, 0.5038],
  5057. [0.4746, 0.3273, 0.7100, 0.2725, 0.4144, 0.2445, 0.4917, 0.5355],
  5058. [0.5059, 0.3507, 0.8818, 0.4093, 0.3965, 0.3580, 0.6559, 0.5396],
  5059. [0.5314, 0.3847, 0.8582, 0.5777, 0.3785, 0.6152, 0.6551, 0.5945]],
  5060. device='cuda:0', grad_fn=<AddmmBackward>)
  5061. landmarks are: tensor([[[0.6229, 0.4107, 0.8138, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  5062. [0.6151, 0.4125, 0.8737, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  5063. [0.6090, 0.4010, 0.7837, 0.3483, 0.3537, 0.3783, 0.5462, 0.5077],
  5064. [0.6259, 0.4120, 0.8012, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895],
  5065. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  5066. [0.0000, 0.0000, 0.6413, 0.1900, 0.4238, 0.1883, 0.5487, 0.5700],
  5067. [0.6357, 0.4097, 0.9038, 0.3883, 0.4212, 0.2950, 0.6686, 0.5390],
  5068. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795]]],
  5069. device='cuda:0')
  5070. loss_train_step before backward: tensor(0.0075, device='cuda:0', grad_fn=<MseLossBackward>)
  5071. loss_train_step after backward: tensor(0.0075, device='cuda:0', grad_fn=<MseLossBackward>)
  5072. loss_train: 0.21865464956499636
  5073. step: 58
  5074. running loss: 0.003769907751120627
  5075.  
  5076. Train Steps: 58/90 Loss: 0.0038 torch.Size([8, 600, 800])
  5077. torch.Size([8, 8])
  5078. tensor([[0.6218, 0.4098, 0.7238, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  5079. [0.6079, 0.3964, 0.7420, 0.2958, 0.3563, 0.2917, 0.5351, 0.4980],
  5080. [0.6264, 0.4049, 0.8988, 0.4633, 0.3813, 0.4983, 0.6326, 0.4843],
  5081. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  5082. [0.6222, 0.3937, 0.8350, 0.5617, 0.4138, 0.4600, 0.5800, 0.5233],
  5083. [0.6124, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  5084. [0.6131, 0.4064, 0.8638, 0.5200, 0.4788, 0.4783, 0.5258, 0.5867],
  5085. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578]],
  5086. device='cuda:0', dtype=torch.float64)
  5087. predictions are: tensor([[0.5859, 0.4131, 0.7635, 0.2248, 0.4391, 0.2565, 0.6437, 0.5214],
  5088. [0.5467, 0.3847, 0.7767, 0.2812, 0.3527, 0.3002, 0.5714, 0.4764],
  5089. [0.5477, 0.3962, 0.9153, 0.4810, 0.3778, 0.5369, 0.6626, 0.4993],
  5090. [0.5260, 0.3667, 0.8673, 0.4650, 0.3777, 0.4376, 0.5970, 0.5427],
  5091. [0.4763, 0.3250, 0.8709, 0.5843, 0.4274, 0.4781, 0.5866, 0.5502],
  5092. [0.5177, 0.3731, 0.7484, 0.2848, 0.3712, 0.3071, 0.5603, 0.5259],
  5093. [0.6167, 0.4277, 0.8827, 0.5338, 0.4489, 0.4727, 0.6133, 0.5438],
  5094. [0.4654, 0.3342, 0.7150, 0.2396, 0.3987, 0.2010, 0.5648, 0.5176]],
  5095. device='cuda:0', grad_fn=<AddmmBackward>)
  5096. landmarks are: tensor([[[0.6218, 0.4098, 0.7237, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  5097. [0.6079, 0.3964, 0.7420, 0.2958, 0.3562, 0.2917, 0.5351, 0.4980],
  5098. [0.6264, 0.4049, 0.8988, 0.4633, 0.3812, 0.4983, 0.6326, 0.4843],
  5099. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  5100. [0.6222, 0.3937, 0.8350, 0.5617, 0.4137, 0.4600, 0.5800, 0.5233],
  5101. [0.6123, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  5102. [0.6132, 0.4063, 0.8637, 0.5200, 0.4787, 0.4783, 0.5258, 0.5867],
  5103. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578]]],
  5104. device='cuda:0')
  5105. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  5106. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  5107. loss_train: 0.2205590239027515
  5108. step: 59
  5109. running loss: 0.0037382885407246016
  5110. Train Steps: 59/90 Loss: 0.0037 torch.Size([8, 600, 800])
  5111. torch.Size([8, 8])
  5112. tensor([[0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  5113. [0.6273, 0.4143, 0.8750, 0.5700, 0.3987, 0.4717, 0.6013, 0.5467],
  5114. [0.6193, 0.4079, 0.7288, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  5115. [0.6175, 0.4091, 0.7863, 0.2800, 0.3638, 0.3583, 0.6188, 0.5433],
  5116. [0.6125, 0.3974, 0.7725, 0.2517, 0.3538, 0.3317, 0.5887, 0.5500],
  5117. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  5118. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5837, 0.5583],
  5119. [0.6131, 0.4064, 0.8638, 0.5200, 0.4788, 0.4783, 0.5258, 0.5867]],
  5120. device='cuda:0', dtype=torch.float64)
  5121. predictions are: tensor([[0.5923, 0.3933, 0.8322, 0.5039, 0.4399, 0.5371, 0.6048, 0.4884],
  5122. [0.5682, 0.3870, 0.8590, 0.5505, 0.3958, 0.4336, 0.6032, 0.5265],
  5123. [0.6281, 0.4465, 0.7253, 0.2506, 0.4124, 0.2161, 0.6342, 0.5426],
  5124. [0.5458, 0.3981, 0.7589, 0.2239, 0.3483, 0.3101, 0.6219, 0.4943],
  5125. [0.5405, 0.3728, 0.7586, 0.2376, 0.3597, 0.2994, 0.5904, 0.5146],
  5126. [0.5155, 0.3839, 0.8459, 0.3394, 0.3507, 0.4217, 0.7081, 0.4893],
  5127. [0.5667, 0.4076, 0.8640, 0.4323, 0.4619, 0.4661, 0.6003, 0.5212],
  5128. [0.6148, 0.4273, 0.8491, 0.5168, 0.4372, 0.4141, 0.5909, 0.5302]],
  5129. device='cuda:0', grad_fn=<AddmmBackward>)
  5130. landmarks are: tensor([[[0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  5131. [0.6273, 0.4143, 0.8750, 0.5700, 0.3988, 0.4717, 0.6012, 0.5467],
  5132. [0.6193, 0.4078, 0.7287, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  5133. [0.6175, 0.4091, 0.7862, 0.2800, 0.3638, 0.3583, 0.6187, 0.5433],
  5134. [0.6125, 0.3974, 0.7725, 0.2517, 0.3537, 0.3317, 0.5888, 0.5500],
  5135. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  5136. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5838, 0.5583],
  5137. [0.6132, 0.4063, 0.8637, 0.5200, 0.4787, 0.4783, 0.5258, 0.5867]]],
  5138. device='cuda:0')
  5139. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  5140. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  5141. loss_train: 0.22198802302591503
  5142. step: 60
  5143. running loss: 0.0036998003837652505
  5144. Train Steps: 60/90 Loss: 0.0037 torch.Size([8, 600, 800])
  5145. torch.Size([8, 8])
  5146. tensor([[ nan, nan, 0.7725, 0.2611, 0.3675, 0.2733, 0.5413, 0.5167],
  5147. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  5148. [0.6171, 0.4127, 0.8900, 0.4800, 0.4325, 0.5783, 0.5769, 0.5090],
  5149. [0.6222, 0.3937, 0.8350, 0.5617, 0.4138, 0.4600, 0.5800, 0.5233],
  5150. [0.6250, 0.3993, 0.9138, 0.4333, 0.3763, 0.5217, 0.6995, 0.5320],
  5151. [0.6229, 0.4066, 0.8513, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  5152. [0.6200, 0.4059, 0.8700, 0.4900, 0.4163, 0.5000, 0.6162, 0.5467],
  5153. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5787, 0.5117]],
  5154. device='cuda:0', dtype=torch.float64)
  5155. predictions are: tensor([[0.1789, 0.1370, 0.7120, 0.2195, 0.3751, 0.2633, 0.5566, 0.5312],
  5156. [0.5631, 0.3876, 0.8010, 0.1853, 0.4524, 0.1427, 0.6686, 0.5063],
  5157. [0.6861, 0.4646, 0.8735, 0.4319, 0.4315, 0.5165, 0.5684, 0.5173],
  5158. [0.5888, 0.4002, 0.8069, 0.5105, 0.4057, 0.4213, 0.5696, 0.5447],
  5159. [0.7186, 0.4990, 0.8426, 0.4124, 0.3694, 0.4953, 0.7160, 0.5340],
  5160. [0.6200, 0.4549, 0.8096, 0.5021, 0.4174, 0.4357, 0.6127, 0.5518],
  5161. [0.6523, 0.4633, 0.8407, 0.4444, 0.4231, 0.4848, 0.6290, 0.5298],
  5162. [0.6444, 0.4142, 0.7149, 0.1780, 0.3949, 0.1900, 0.5875, 0.5107]],
  5163. device='cuda:0', grad_fn=<AddmmBackward>)
  5164. landmarks are: tensor([[[0.0000, 0.0000, 0.7725, 0.2611, 0.3675, 0.2733, 0.5412, 0.5167],
  5165. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  5166. [0.6171, 0.4127, 0.8900, 0.4800, 0.4325, 0.5783, 0.5769, 0.5090],
  5167. [0.6222, 0.3937, 0.8350, 0.5617, 0.4137, 0.4600, 0.5800, 0.5233],
  5168. [0.6250, 0.3993, 0.9137, 0.4333, 0.3762, 0.5217, 0.6995, 0.5320],
  5169. [0.6229, 0.4066, 0.8512, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  5170. [0.6199, 0.4059, 0.8700, 0.4900, 0.4162, 0.5000, 0.6162, 0.5467],
  5171. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5788, 0.5117]]],
  5172. device='cuda:0')
  5173. loss_train_step before backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  5174. loss_train_step after backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  5175. loss_train: 0.22428614646196365
  5176. step: 61
  5177. running loss: 0.0036768220731469453
  5178. Train Steps: 61/90 Loss: 0.0037 torch.Size([8, 600, 800])
  5179. torch.Size([8, 8])
  5180. tensor([[0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  5181. [0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767],
  5182. [0.6293, 0.4097, 0.8800, 0.2517, 0.5262, 0.2600, 0.7430, 0.5378],
  5183. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  5184. [0.6284, 0.4127, 0.8538, 0.5867, 0.4363, 0.5083, 0.6038, 0.5433],
  5185. [ nan, nan, 0.7425, 0.2117, 0.3937, 0.2433, 0.5438, 0.5567],
  5186. [0.6179, 0.4118, 0.7278, 0.4237, 0.3588, 0.3400, 0.5675, 0.5917],
  5187. [0.6222, 0.3957, 0.8838, 0.5017, 0.3937, 0.4600, 0.5900, 0.5017]],
  5188. device='cuda:0', dtype=torch.float64)
  5189. predictions are: tensor([[0.5691, 0.3808, 0.7564, 0.1777, 0.4299, 0.1889, 0.6466, 0.5016],
  5190. [0.6328, 0.4316, 0.8196, 0.4589, 0.4082, 0.4745, 0.5711, 0.5704],
  5191. [0.5339, 0.3510, 0.8157, 0.1961, 0.4921, 0.2346, 0.7271, 0.5324],
  5192. [0.7159, 0.4650, 0.8122, 0.4496, 0.4231, 0.5123, 0.6721, 0.5393],
  5193. [0.6640, 0.4476, 0.8020, 0.5129, 0.4045, 0.4800, 0.5805, 0.5495],
  5194. [0.2912, 0.1803, 0.7068, 0.2265, 0.4071, 0.2261, 0.5405, 0.5527],
  5195. [0.6205, 0.4289, 0.7558, 0.3432, 0.3518, 0.3197, 0.5453, 0.5625],
  5196. [0.6474, 0.4268, 0.8590, 0.4616, 0.3832, 0.4418, 0.6134, 0.5029]],
  5197. device='cuda:0', grad_fn=<AddmmBackward>)
  5198. landmarks are: tensor([[[0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  5199. [0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767],
  5200. [0.6293, 0.4097, 0.8800, 0.2517, 0.5263, 0.2600, 0.7430, 0.5378],
  5201. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  5202. [0.6284, 0.4127, 0.8537, 0.5867, 0.4363, 0.5083, 0.6037, 0.5433],
  5203. [0.0000, 0.0000, 0.7425, 0.2117, 0.3938, 0.2433, 0.5437, 0.5567],
  5204. [0.6179, 0.4118, 0.7278, 0.4237, 0.3587, 0.3400, 0.5675, 0.5917],
  5205. [0.6222, 0.3957, 0.8838, 0.5017, 0.3938, 0.4600, 0.5900, 0.5017]]],
  5206. device='cuda:0')
  5207. loss_train_step before backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  5208. loss_train_step after backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  5209. loss_train: 0.22752876160666347
  5210. step: 62
  5211. running loss: 0.003669818735591346
  5212.  
  5213. Train Steps: 62/90 Loss: 0.0037 torch.Size([8, 600, 800])
  5214. torch.Size([8, 8])
  5215. tensor([[0.6182, 0.4058, 0.8738, 0.4350, 0.3563, 0.3400, 0.5290, 0.5822],
  5216. [0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5837, 0.5317],
  5217. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200],
  5218. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5363, 0.5550],
  5219. [0.6143, 0.4040, 0.8237, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  5220. [0.6197, 0.4090, 0.7825, 0.2500, 0.4200, 0.2483, 0.5988, 0.5667],
  5221. [0.6339, 0.4123, 0.8638, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  5222. [0.6109, 0.4036, 0.7188, 0.1750, 0.3850, 0.2550, 0.5863, 0.5567]],
  5223. device='cuda:0', dtype=torch.float64)
  5224. predictions are: tensor([[0.6034, 0.3830, 0.8470, 0.4437, 0.3738, 0.3747, 0.5338, 0.5797],
  5225. [0.5444, 0.3546, 0.8087, 0.3473, 0.3960, 0.3135, 0.6162, 0.5542],
  5226. [0.5521, 0.3261, 0.8409, 0.3032, 0.4649, 0.2440, 0.6543, 0.5259],
  5227. [0.5926, 0.3772, 0.6626, 0.2557, 0.4032, 0.2445, 0.5561, 0.5455],
  5228. [0.5108, 0.3200, 0.7701, 0.3098, 0.4265, 0.2208, 0.5590, 0.5159],
  5229. [0.6456, 0.4156, 0.7671, 0.2335, 0.4370, 0.2696, 0.6316, 0.5587],
  5230. [0.6134, 0.4167, 0.8674, 0.5409, 0.4430, 0.5818, 0.7387, 0.5563],
  5231. [0.6006, 0.3781, 0.6835, 0.2165, 0.4343, 0.2681, 0.5758, 0.5471]],
  5232. device='cuda:0', grad_fn=<AddmmBackward>)
  5233. landmarks are: tensor([[[0.6182, 0.4058, 0.8737, 0.4350, 0.3562, 0.3400, 0.5290, 0.5822],
  5234. [0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5838, 0.5317],
  5235. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200],
  5236. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5362, 0.5550],
  5237. [0.6143, 0.4040, 0.8238, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  5238. [0.6197, 0.4090, 0.7825, 0.2500, 0.4200, 0.2483, 0.5987, 0.5667],
  5239. [0.6339, 0.4123, 0.8637, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  5240. [0.6108, 0.4036, 0.7188, 0.1750, 0.3850, 0.2550, 0.5863, 0.5567]]],
  5241. device='cuda:0')
  5242. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  5243. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  5244. loss_train: 0.22869582765270025
  5245. step: 63
  5246. running loss: 0.0036300925024238136
  5247. Train Steps: 63/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5248. torch.Size([8, 8])
  5249. tensor([[0.6275, 0.4111, 0.8463, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  5250. [0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  5251. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  5252. [ nan, nan, 0.7981, 0.3194, 0.3625, 0.3167, 0.5040, 0.5563],
  5253. [ nan, nan, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  5254. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467],
  5255. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  5256. [0.6329, 0.4055, 0.9050, 0.4783, 0.3613, 0.3917, 0.6464, 0.5019]],
  5257. device='cuda:0', dtype=torch.float64)
  5258. predictions are: tensor([[0.6140, 0.3829, 0.8532, 0.2711, 0.4873, 0.2148, 0.6442, 0.5450],
  5259. [0.6847, 0.4150, 0.8289, 0.3931, 0.3653, 0.4806, 0.6040, 0.5475],
  5260. [0.7392, 0.4679, 0.8294, 0.5288, 0.4050, 0.5031, 0.6651, 0.5992],
  5261. [0.4096, 0.2648, 0.7645, 0.3139, 0.3828, 0.2889, 0.5430, 0.5730],
  5262. [0.1937, 0.0940, 0.8403, 0.2669, 0.5400, 0.2014, 0.6737, 0.5628],
  5263. [0.7099, 0.4543, 0.8390, 0.3263, 0.3754, 0.4002, 0.5924, 0.5534],
  5264. [0.7283, 0.4502, 0.8550, 0.4972, 0.4617, 0.4719, 0.5788, 0.5715],
  5265. [0.7993, 0.4937, 0.9025, 0.4761, 0.4007, 0.3892, 0.6428, 0.5437]],
  5266. device='cuda:0', grad_fn=<AddmmBackward>)
  5267. landmarks are: tensor([[[0.6275, 0.4111, 0.8462, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  5268. [0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  5269. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  5270. [0.0000, 0.0000, 0.7981, 0.3194, 0.3625, 0.3167, 0.5040, 0.5563],
  5271. [0.0000, 0.0000, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  5272. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467],
  5273. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  5274. [0.6329, 0.4055, 0.9050, 0.4783, 0.3613, 0.3917, 0.6464, 0.5019]]],
  5275. device='cuda:0')
  5276. loss_train_step before backward: tensor(0.0062, device='cuda:0', grad_fn=<MseLossBackward>)
  5277. loss_train_step after backward: tensor(0.0062, device='cuda:0', grad_fn=<MseLossBackward>)
  5278. loss_train: 0.23489011928904802
  5279. step: 64
  5280. running loss: 0.0036701581138913753
  5281. Train Steps: 64/90 Loss: 0.0037 torch.Size([8, 600, 800])
  5282. torch.Size([8, 8])
  5283. tensor([[ nan, nan, 0.9088, 0.3783, 0.4562, 0.2617, 0.6741, 0.5575],
  5284. [0.6263, 0.4030, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  5285. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896],
  5286. [0.6117, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103],
  5287. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461],
  5288. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5787, 0.5117],
  5289. [0.6102, 0.4005, 0.8688, 0.5100, 0.4813, 0.5400, 0.5404, 0.5064],
  5290. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533]],
  5291. device='cuda:0', dtype=torch.float64)
  5292. predictions are: tensor([[0.2212, 0.1345, 0.9286, 0.3959, 0.4843, 0.2741, 0.6916, 0.6015],
  5293. [0.6566, 0.4074, 0.9196, 0.4697, 0.3683, 0.5070, 0.6453, 0.4970],
  5294. [0.5537, 0.3543, 0.8277, 0.3884, 0.3790, 0.2970, 0.5461, 0.5881],
  5295. [0.5645, 0.3503, 0.6897, 0.2247, 0.3856, 0.2232, 0.5269, 0.5686],
  5296. [0.6891, 0.4353, 0.8483, 0.5375, 0.4398, 0.4830, 0.5719, 0.5832],
  5297. [0.6096, 0.3528, 0.7507, 0.2245, 0.4239, 0.2051, 0.5894, 0.5566],
  5298. [0.6337, 0.3920, 0.8727, 0.5009, 0.4832, 0.4534, 0.5873, 0.5605],
  5299. [0.6439, 0.4225, 0.8797, 0.4280, 0.4242, 0.4819, 0.6083, 0.5413]],
  5300. device='cuda:0', grad_fn=<AddmmBackward>)
  5301. landmarks are: tensor([[[0.0000, 0.0000, 0.9087, 0.3783, 0.4563, 0.2617, 0.6741, 0.5575],
  5302. [0.6263, 0.4029, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  5303. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896],
  5304. [0.6116, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103],
  5305. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461],
  5306. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5788, 0.5117],
  5307. [0.6102, 0.4005, 0.8687, 0.5100, 0.4812, 0.5400, 0.5404, 0.5064],
  5308. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533]]],
  5309. device='cuda:0')
  5310. loss_train_step before backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  5311. loss_train_step after backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  5312. loss_train: 0.23684400937054306
  5313. step: 65
  5314. running loss: 0.003643753990316047
  5315. Train Steps: 65/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5316. torch.Size([8, 8])
  5317. tensor([[0.6332, 0.4118, 0.9238, 0.4267, 0.4012, 0.4733, 0.7525, 0.5436],
  5318. [0.6147, 0.4026, 0.6600, 0.2467, 0.4088, 0.2150, 0.5489, 0.5773],
  5319. [0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  5320. [ nan, nan, 0.7225, 0.2167, 0.3987, 0.2283, 0.5427, 0.5181],
  5321. [0.6226, 0.4103, 0.8575, 0.3450, 0.4388, 0.2067, 0.5787, 0.5383],
  5322. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  5323. [0.6202, 0.4053, 0.8638, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  5324. [0.6254, 0.4076, 0.8700, 0.3267, 0.4150, 0.3083, 0.7050, 0.5609]],
  5325. device='cuda:0', dtype=torch.float64)
  5326. predictions are: tensor([[0.7355, 0.4587, 0.9337, 0.4735, 0.4093, 0.4814, 0.7177, 0.5522],
  5327. [0.5557, 0.3274, 0.7331, 0.2659, 0.4116, 0.2238, 0.4978, 0.5754],
  5328. [0.7013, 0.4191, 0.9234, 0.5755, 0.3711, 0.4749, 0.5756, 0.5564],
  5329. [0.0530, 0.0123, 0.7388, 0.2372, 0.4230, 0.2440, 0.5240, 0.5477],
  5330. [0.5336, 0.3270, 0.8849, 0.3829, 0.4673, 0.2350, 0.5589, 0.5687],
  5331. [0.6730, 0.3989, 0.9308, 0.5037, 0.4046, 0.6039, 0.6622, 0.5154],
  5332. [0.6858, 0.4156, 0.8866, 0.5624, 0.4506, 0.5007, 0.5818, 0.5358],
  5333. [0.5166, 0.3204, 0.9286, 0.3357, 0.4373, 0.2783, 0.6617, 0.5548]],
  5334. device='cuda:0', grad_fn=<AddmmBackward>)
  5335. landmarks are: tensor([[[0.6332, 0.4118, 0.9237, 0.4267, 0.4013, 0.4733, 0.7525, 0.5436],
  5336. [0.6147, 0.4026, 0.6600, 0.2467, 0.4087, 0.2150, 0.5489, 0.5773],
  5337. [0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  5338. [0.0000, 0.0000, 0.7225, 0.2167, 0.3988, 0.2283, 0.5427, 0.5181],
  5339. [0.6226, 0.4103, 0.8575, 0.3450, 0.4387, 0.2067, 0.5788, 0.5383],
  5340. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  5341. [0.6202, 0.4053, 0.8637, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  5342. [0.6254, 0.4076, 0.8700, 0.3267, 0.4150, 0.3083, 0.7050, 0.5609]]],
  5343. device='cuda:0')
  5344. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  5345. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  5346. loss_train: 0.2385742465266958
  5347. step: 66
  5348. running loss: 0.003614761311010542
  5349.  
  5350. Train Steps: 66/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5351. torch.Size([8, 8])
  5352. tensor([[0.6364, 0.4165, 0.9088, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  5353. [ nan, nan, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  5354. [0.6034, 0.4011, 0.7350, 0.2533, 0.3438, 0.3367, 0.5516, 0.5084],
  5355. [ nan, nan, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733],
  5356. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  5357. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901],
  5358. [0.6186, 0.4060, 0.8750, 0.5050, 0.3538, 0.4367, 0.5813, 0.6083],
  5359. [0.6097, 0.4024, 0.8488, 0.3717, 0.3875, 0.5517, 0.5836, 0.5591]],
  5360. device='cuda:0', dtype=torch.float64)
  5361. predictions are: tensor([[0.6671, 0.4208, 0.9785, 0.5062, 0.4243, 0.3026, 0.6318, 0.5373],
  5362. [0.1790, 0.0846, 0.8194, 0.3341, 0.4027, 0.3083, 0.5173, 0.5482],
  5363. [0.6645, 0.4287, 0.7940, 0.3145, 0.3603, 0.3541, 0.5978, 0.5310],
  5364. [0.2190, 0.1470, 0.7626, 0.2580, 0.4547, 0.2251, 0.5737, 0.5546],
  5365. [0.5184, 0.3228, 0.7564, 0.3006, 0.4239, 0.2514, 0.5888, 0.5536],
  5366. [0.5250, 0.3274, 0.8387, 0.3473, 0.3684, 0.2694, 0.5364, 0.5339],
  5367. [0.6058, 0.3866, 0.9575, 0.5627, 0.4092, 0.5112, 0.6149, 0.5469],
  5368. [0.6870, 0.4104, 0.9199, 0.4403, 0.4319, 0.5612, 0.6627, 0.5029]],
  5369. device='cuda:0', grad_fn=<AddmmBackward>)
  5370. landmarks are: tensor([[[0.6364, 0.4165, 0.9087, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  5371. [0.0000, 0.0000, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  5372. [0.6033, 0.4011, 0.7350, 0.2533, 0.3438, 0.3367, 0.5516, 0.5084],
  5373. [0.0000, 0.0000, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733],
  5374. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  5375. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901],
  5376. [0.6186, 0.4060, 0.8750, 0.5050, 0.3537, 0.4367, 0.5813, 0.6083],
  5377. [0.6097, 0.4024, 0.8487, 0.3717, 0.3875, 0.5517, 0.5836, 0.5591]]],
  5378. device='cuda:0')
  5379. loss_train_step before backward: tensor(0.0042, device='cuda:0', grad_fn=<MseLossBackward>)
  5380. loss_train_step after backward: tensor(0.0042, device='cuda:0', grad_fn=<MseLossBackward>)
  5381. loss_train: 0.24276340811047703
  5382. step: 67
  5383. running loss: 0.003623334449410105
  5384. Train Steps: 67/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5385. torch.Size([8, 8])
  5386. tensor([[0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  5387. [0.6132, 0.4066, 0.7259, 0.2402, 0.3588, 0.3300, 0.6000, 0.5600],
  5388. [0.6250, 0.4146, 0.8838, 0.3933, 0.3588, 0.4283, 0.6162, 0.5367],
  5389. [0.6193, 0.4108, 0.7438, 0.2700, 0.3650, 0.3683, 0.6238, 0.5717],
  5390. [ nan, nan, 0.6512, 0.1717, 0.4100, 0.1983, 0.5253, 0.5240],
  5391. [0.6286, 0.4055, 0.9000, 0.4717, 0.3763, 0.4683, 0.7018, 0.5494],
  5392. [0.6182, 0.3930, 0.8841, 0.3892, 0.3556, 0.4967, 0.6222, 0.5279],
  5393. [0.6197, 0.4090, 0.7825, 0.2500, 0.4200, 0.2483, 0.5988, 0.5667]],
  5394. device='cuda:0', dtype=torch.float64)
  5395. predictions are: tensor([[0.5577, 0.3404, 0.9322, 0.5424, 0.4028, 0.5637, 0.5783, 0.5157],
  5396. [0.5639, 0.3888, 0.7683, 0.2950, 0.3763, 0.3433, 0.5770, 0.5662],
  5397. [0.5509, 0.3533, 0.9084, 0.4569, 0.3602, 0.3963, 0.5602, 0.4995],
  5398. [0.6154, 0.3938, 0.8092, 0.3178, 0.3638, 0.3636, 0.5955, 0.5416],
  5399. [0.0305, 0.0186, 0.7465, 0.2533, 0.4329, 0.2216, 0.5314, 0.5542],
  5400. [0.5792, 0.3591, 0.9353, 0.5333, 0.3668, 0.4853, 0.6384, 0.5251],
  5401. [0.5785, 0.3509, 0.9181, 0.4518, 0.3553, 0.4828, 0.6251, 0.5212],
  5402. [0.5584, 0.3610, 0.8354, 0.2901, 0.4220, 0.2634, 0.5679, 0.5586]],
  5403. device='cuda:0', grad_fn=<AddmmBackward>)
  5404. landmarks are: tensor([[[0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  5405. [0.6132, 0.4066, 0.7259, 0.2402, 0.3587, 0.3300, 0.6000, 0.5600],
  5406. [0.6250, 0.4146, 0.8838, 0.3933, 0.3587, 0.4283, 0.6162, 0.5367],
  5407. [0.6193, 0.4108, 0.7437, 0.2700, 0.3650, 0.3683, 0.6237, 0.5717],
  5408. [0.0000, 0.0000, 0.6513, 0.1717, 0.4100, 0.1983, 0.5253, 0.5240],
  5409. [0.6286, 0.4055, 0.9000, 0.4717, 0.3762, 0.4683, 0.7018, 0.5494],
  5410. [0.6182, 0.3930, 0.8841, 0.3892, 0.3556, 0.4967, 0.6222, 0.5279],
  5411. [0.6197, 0.4090, 0.7825, 0.2500, 0.4200, 0.2483, 0.5987, 0.5667]]],
  5412. device='cuda:0')
  5413. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  5414. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  5415. loss_train: 0.24445695418398827
  5416. step: 68
  5417. running loss: 0.003594955208588063
  5418. Train Steps: 68/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5419. torch.Size([8, 8])
  5420. tensor([[0.6189, 0.4029, 0.8375, 0.5767, 0.4745, 0.4829, 0.5551, 0.5598],
  5421. [ nan, nan, 0.7268, 0.2333, 0.4125, 0.1933, 0.5112, 0.5383],
  5422. [0.6259, 0.4156, 0.8812, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960],
  5423. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  5424. [0.6250, 0.4013, 0.8525, 0.5417, 0.4037, 0.5117, 0.6325, 0.5017],
  5425. [0.6236, 0.3967, 0.8675, 0.5400, 0.3862, 0.4517, 0.5825, 0.5200],
  5426. [0.6208, 0.4082, 0.8538, 0.3067, 0.3588, 0.3717, 0.6112, 0.5517],
  5427. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461]],
  5428. device='cuda:0', dtype=torch.float64)
  5429. predictions are: tensor([[ 0.5516, 0.3771, 0.8252, 0.4969, 0.4173, 0.4475, 0.5955, 0.5628],
  5430. [-0.0498, -0.0271, 0.7035, 0.1818, 0.3952, 0.2227, 0.5223, 0.5519],
  5431. [ 0.4824, 0.3117, 0.8709, 0.2582, 0.4267, 0.2320, 0.6016, 0.5198],
  5432. [ 0.5551, 0.3592, 0.8599, 0.4213, 0.3784, 0.5226, 0.6311, 0.5210],
  5433. [ 0.5545, 0.3603, 0.8491, 0.4894, 0.3806, 0.5112, 0.6097, 0.5152],
  5434. [ 0.5212, 0.3685, 0.8626, 0.5007, 0.3574, 0.4597, 0.5946, 0.5317],
  5435. [ 0.5741, 0.3873, 0.8403, 0.2797, 0.3233, 0.3746, 0.6037, 0.5376],
  5436. [ 0.5667, 0.3745, 0.8423, 0.4916, 0.4029, 0.5023, 0.5392, 0.5470]],
  5437. device='cuda:0', grad_fn=<AddmmBackward>)
  5438. landmarks are: tensor([[[0.6189, 0.4029, 0.8375, 0.5767, 0.4745, 0.4829, 0.5551, 0.5598],
  5439. [0.0000, 0.0000, 0.7268, 0.2333, 0.4125, 0.1933, 0.5113, 0.5383],
  5440. [0.6259, 0.4156, 0.8813, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960],
  5441. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  5442. [0.6250, 0.4013, 0.8525, 0.5417, 0.4038, 0.5117, 0.6325, 0.5017],
  5443. [0.6236, 0.3967, 0.8675, 0.5400, 0.3862, 0.4517, 0.5825, 0.5200],
  5444. [0.6208, 0.4082, 0.8537, 0.3067, 0.3587, 0.3717, 0.6112, 0.5517],
  5445. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461]]],
  5446. device='cuda:0')
  5447. loss_train_step before backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  5448. loss_train_step after backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  5449. loss_train: 0.24625502328854054
  5450. step: 69
  5451. running loss: 0.003568913380993341
  5452. Train Steps: 69/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5453. torch.Size([8, 8])
  5454. tensor([[0.6193, 0.4050, 0.7313, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  5455. [0.6260, 0.4120, 0.8013, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895],
  5456. [0.6357, 0.4118, 0.8400, 0.2500, 0.5413, 0.1633, 0.6725, 0.5586],
  5457. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323],
  5458. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  5459. [0.6188, 0.4099, 0.7400, 0.2433, 0.3962, 0.2750, 0.6162, 0.5467],
  5460. [0.6228, 0.4004, 0.8750, 0.5250, 0.3825, 0.5233, 0.6362, 0.5000],
  5461. [0.6293, 0.3982, 0.8700, 0.5300, 0.3763, 0.4717, 0.7050, 0.5297]],
  5462. device='cuda:0', dtype=torch.float64)
  5463. predictions are: tensor([[0.4102, 0.2683, 0.7072, 0.2315, 0.3801, 0.3003, 0.5461, 0.5529],
  5464. [0.4654, 0.3223, 0.7671, 0.2061, 0.4613, 0.2163, 0.6026, 0.5000],
  5465. [0.3718, 0.2384, 0.8290, 0.2430, 0.4869, 0.2548, 0.6139, 0.5142],
  5466. [0.4628, 0.3322, 0.7671, 0.2106, 0.4033, 0.2363, 0.5672, 0.5524],
  5467. [0.4736, 0.3394, 0.8669, 0.4077, 0.3232, 0.4130, 0.5212, 0.5193],
  5468. [0.5163, 0.3695, 0.7215, 0.2309, 0.3828, 0.3317, 0.5854, 0.5381],
  5469. [0.5423, 0.3440, 0.8990, 0.5151, 0.3579, 0.5819, 0.5909, 0.5004],
  5470. [0.5993, 0.3707, 0.8675, 0.5264, 0.3271, 0.5180, 0.6483, 0.5365]],
  5471. device='cuda:0', grad_fn=<AddmmBackward>)
  5472. landmarks are: tensor([[[0.6193, 0.4050, 0.7312, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  5473. [0.6259, 0.4120, 0.8012, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895],
  5474. [0.6357, 0.4118, 0.8400, 0.2500, 0.5412, 0.1633, 0.6725, 0.5586],
  5475. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323],
  5476. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  5477. [0.6188, 0.4099, 0.7400, 0.2433, 0.3963, 0.2750, 0.6162, 0.5467],
  5478. [0.6228, 0.4004, 0.8750, 0.5250, 0.3825, 0.5233, 0.6363, 0.5000],
  5479. [0.6293, 0.3982, 0.8700, 0.5300, 0.3762, 0.4717, 0.7050, 0.5297]]],
  5480. device='cuda:0')
  5481. loss_train_step before backward: tensor(0.0054, device='cuda:0', grad_fn=<MseLossBackward>)
  5482. loss_train_step after backward: tensor(0.0054, device='cuda:0', grad_fn=<MseLossBackward>)
  5483. loss_train: 0.25164759380277246
  5484. step: 70
  5485. running loss: 0.003594965625753892
  5486.  
  5487. Train Steps: 70/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5488. torch.Size([8, 8])
  5489. tensor([[0.6118, 0.4052, 0.8463, 0.3917, 0.3538, 0.3450, 0.5053, 0.5593],
  5490. [0.6197, 0.4050, 0.7527, 0.2000, 0.4042, 0.2249, 0.5895, 0.4995],
  5491. [0.6353, 0.4128, 0.8488, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  5492. [0.6193, 0.4050, 0.7313, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  5493. [0.6286, 0.3977, 0.9038, 0.4733, 0.3900, 0.4150, 0.7074, 0.5320],
  5494. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  5495. [0.6189, 0.4049, 0.8888, 0.4417, 0.4213, 0.5200, 0.5988, 0.5633],
  5496. [0.6129, 0.4069, 0.8750, 0.5067, 0.3875, 0.4233, 0.5235, 0.5881]],
  5497. device='cuda:0', dtype=torch.float64)
  5498. predictions are: tensor([[0.4371, 0.3352, 0.8117, 0.3662, 0.3465, 0.3646, 0.4915, 0.5341],
  5499. [0.4829, 0.3549, 0.7093, 0.1705, 0.4160, 0.2348, 0.6135, 0.5019],
  5500. [0.3030, 0.1993, 0.7985, 0.2163, 0.5106, 0.2201, 0.6479, 0.5032],
  5501. [0.4934, 0.3381, 0.6827, 0.2094, 0.3904, 0.2574, 0.5596, 0.5345],
  5502. [0.7015, 0.4689, 0.8686, 0.4039, 0.3264, 0.4478, 0.6475, 0.4829],
  5503. [0.4930, 0.3494, 0.7581, 0.2368, 0.4055, 0.2341, 0.5452, 0.5235],
  5504. [0.5343, 0.3676, 0.8534, 0.4128, 0.4094, 0.5396, 0.5989, 0.5252],
  5505. [0.5998, 0.4077, 0.8622, 0.5105, 0.3764, 0.4379, 0.5306, 0.5128]],
  5506. device='cuda:0', grad_fn=<AddmmBackward>)
  5507. landmarks are: tensor([[[0.6118, 0.4052, 0.8462, 0.3917, 0.3537, 0.3450, 0.5053, 0.5593],
  5508. [0.6197, 0.4050, 0.7527, 0.2000, 0.4042, 0.2249, 0.5895, 0.4995],
  5509. [0.6353, 0.4128, 0.8487, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  5510. [0.6193, 0.4050, 0.7312, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  5511. [0.6286, 0.3977, 0.9038, 0.4733, 0.3900, 0.4150, 0.7074, 0.5320],
  5512. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  5513. [0.6189, 0.4049, 0.8888, 0.4417, 0.4212, 0.5200, 0.5987, 0.5633],
  5514. [0.6129, 0.4069, 0.8750, 0.5067, 0.3875, 0.4233, 0.5235, 0.5881]]],
  5515. device='cuda:0')
  5516. loss_train_step before backward: tensor(0.0053, device='cuda:0', grad_fn=<MseLossBackward>)
  5517. loss_train_step after backward: tensor(0.0053, device='cuda:0', grad_fn=<MseLossBackward>)
  5518. loss_train: 0.2569181363796815
  5519. step: 71
  5520. running loss: 0.003618565301122275
  5521. Train Steps: 71/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5522. torch.Size([8, 8])
  5523. tensor([[0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  5524. [0.6311, 0.4008, 0.7935, 0.5746, 0.3900, 0.5033, 0.6955, 0.5366],
  5525. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683],
  5526. [0.6147, 0.4112, 0.7988, 0.3200, 0.3775, 0.2767, 0.5150, 0.5550],
  5527. [0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  5528. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6063, 0.5617],
  5529. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  5530. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355]],
  5531. device='cuda:0', dtype=torch.float64)
  5532. predictions are: tensor([[0.7440, 0.5025, 0.8211, 0.2773, 0.4304, 0.2292, 0.6325, 0.4862],
  5533. [0.5843, 0.4262, 0.7705, 0.4431, 0.4198, 0.4516, 0.6558, 0.5235],
  5534. [0.5360, 0.3710, 0.8328, 0.4776, 0.3894, 0.3353, 0.5496, 0.5669],
  5535. [0.3983, 0.3095, 0.7359, 0.2769, 0.4089, 0.2535, 0.5024, 0.5584],
  5536. [0.6206, 0.4284, 0.8121, 0.3538, 0.3933, 0.5180, 0.5663, 0.5119],
  5537. [0.6116, 0.4225, 0.8465, 0.4352, 0.3794, 0.4521, 0.5980, 0.5525],
  5538. [0.6379, 0.4404, 0.8386, 0.3539, 0.3735, 0.4274, 0.6420, 0.5186],
  5539. [0.5376, 0.3719, 0.8287, 0.2723, 0.4434, 0.2788, 0.6565, 0.5224]],
  5540. device='cuda:0', grad_fn=<AddmmBackward>)
  5541. landmarks are: tensor([[[0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  5542. [0.6311, 0.4008, 0.7935, 0.5746, 0.3900, 0.5033, 0.6955, 0.5366],
  5543. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683],
  5544. [0.6147, 0.4112, 0.7987, 0.3200, 0.3775, 0.2767, 0.5150, 0.5550],
  5545. [0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  5546. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6062, 0.5617],
  5547. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  5548. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355]]],
  5549. device='cuda:0')
  5550. loss_train_step before backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  5551. loss_train_step after backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  5552. loss_train: 0.26006837061140686
  5553. step: 72
  5554. running loss: 0.0036120607029362065
  5555. Train Steps: 72/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5556. torch.Size([8, 8])
  5557. tensor([[0.6229, 0.4066, 0.8513, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  5558. [0.6250, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6088, 0.5183],
  5559. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  5560. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  5561. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  5562. [0.6168, 0.4029, 0.8523, 0.3417, 0.3588, 0.5000, 0.6125, 0.5400],
  5563. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  5564. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6063, 0.5617]],
  5565. device='cuda:0', dtype=torch.float64)
  5566. predictions are: tensor([[0.5909, 0.4132, 0.8230, 0.5111, 0.4550, 0.3941, 0.5952, 0.5519],
  5567. [0.6030, 0.4113, 0.8529, 0.4071, 0.4819, 0.4526, 0.6298, 0.5334],
  5568. [0.6145, 0.4201, 0.8440, 0.3966, 0.4589, 0.4681, 0.6126, 0.5336],
  5569. [0.6355, 0.4346, 0.7490, 0.2479, 0.3614, 0.2378, 0.5285, 0.5285],
  5570. [0.6995, 0.5085, 0.8196, 0.2540, 0.4860, 0.1383, 0.6536, 0.5236],
  5571. [0.6006, 0.4164, 0.8114, 0.3042, 0.3536, 0.4077, 0.6028, 0.5390],
  5572. [0.6246, 0.4432, 0.8034, 0.4662, 0.4182, 0.4467, 0.6713, 0.5607],
  5573. [0.6711, 0.4687, 0.8610, 0.4468, 0.3593, 0.3986, 0.5906, 0.5496]],
  5574. device='cuda:0', grad_fn=<AddmmBackward>)
  5575. landmarks are: tensor([[[0.6229, 0.4066, 0.8512, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  5576. [0.6251, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6087, 0.5183],
  5577. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  5578. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  5579. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  5580. [0.6168, 0.4029, 0.8523, 0.3417, 0.3587, 0.5000, 0.6125, 0.5400],
  5581. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  5582. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6062, 0.5617]]],
  5583. device='cuda:0')
  5584. loss_train_step before backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  5585. loss_train_step after backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  5586. loss_train: 0.2622584825148806
  5587. step: 73
  5588. running loss: 0.0035925819522586383
  5589. Train Steps: 73/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5590. torch.Size([8, 8])
  5591. tensor([[0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6262, 0.5367],
  5592. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  5593. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  5594. [0.6095, 0.3970, 0.8688, 0.4767, 0.4860, 0.4879, 0.5191, 0.4940],
  5595. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  5596. [0.6257, 0.4167, 0.8775, 0.3433, 0.3563, 0.4133, 0.6200, 0.5667],
  5597. [0.6289, 0.4024, 0.9088, 0.4567, 0.3937, 0.5633, 0.7058, 0.5609],
  5598. [0.6264, 0.4248, 0.8938, 0.4183, 0.3875, 0.4100, 0.5400, 0.5350]],
  5599. device='cuda:0', dtype=torch.float64)
  5600. predictions are: tensor([[0.6593, 0.4456, 0.8539, 0.4580, 0.4526, 0.5123, 0.6356, 0.5641],
  5601. [0.8243, 0.5577, 0.7724, 0.3118, 0.3512, 0.2677, 0.6180, 0.5296],
  5602. [0.6100, 0.3944, 0.8633, 0.4584, 0.3923, 0.4094, 0.6911, 0.5411],
  5603. [0.7037, 0.4569, 0.8566, 0.4849, 0.4779, 0.3771, 0.5572, 0.5403],
  5604. [0.6089, 0.4352, 0.8558, 0.4276, 0.4084, 0.3731, 0.5672, 0.5224],
  5605. [0.7343, 0.5111, 0.8361, 0.3372, 0.3965, 0.3233, 0.6683, 0.5543],
  5606. [0.6134, 0.4223, 0.8518, 0.4732, 0.4324, 0.4890, 0.7007, 0.5811],
  5607. [0.5786, 0.4307, 0.8439, 0.4056, 0.3870, 0.3162, 0.5765, 0.5565]],
  5608. device='cuda:0', grad_fn=<AddmmBackward>)
  5609. landmarks are: tensor([[[0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6263, 0.5367],
  5610. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  5611. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  5612. [0.6095, 0.3970, 0.8687, 0.4767, 0.4860, 0.4879, 0.5191, 0.4940],
  5613. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  5614. [0.6257, 0.4167, 0.8775, 0.3433, 0.3562, 0.4133, 0.6200, 0.5667],
  5615. [0.6289, 0.4024, 0.9087, 0.4567, 0.3938, 0.5633, 0.7058, 0.5609],
  5616. [0.6264, 0.4248, 0.8938, 0.4183, 0.3875, 0.4100, 0.5400, 0.5350]]],
  5617. device='cuda:0')
  5618. loss_train_step before backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  5619. loss_train_step after backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  5620. loss_train: 0.26550832379143685
  5621. step: 74
  5622. running loss: 0.0035879503215059034
  5623.  
  5624. Train Steps: 74/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5625. torch.Size([8, 8])
  5626. tensor([[0.6124, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  5627. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  5628. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  5629. [ nan, nan, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  5630. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  5631. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  5632. [0.6226, 0.4098, 0.8912, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  5633. [0.6250, 0.4106, 0.8700, 0.3717, 0.3588, 0.4967, 0.6038, 0.5167]],
  5634. device='cuda:0', dtype=torch.float64)
  5635. predictions are: tensor([[0.7221, 0.4680, 0.8072, 0.5104, 0.3845, 0.4499, 0.5519, 0.5637],
  5636. [0.6449, 0.4268, 0.8583, 0.5027, 0.4076, 0.5221, 0.6426, 0.5507],
  5637. [0.7298, 0.5031, 0.7766, 0.2269, 0.4749, 0.1387, 0.6136, 0.5198],
  5638. [0.4523, 0.3037, 0.8340, 0.2722, 0.5095, 0.1814, 0.6730, 0.5421],
  5639. [0.7061, 0.4926, 0.8780, 0.4993, 0.3746, 0.4107, 0.6573, 0.5228],
  5640. [0.7205, 0.4687, 0.8657, 0.5145, 0.4142, 0.5632, 0.6663, 0.5195],
  5641. [0.7471, 0.4946, 0.8512, 0.4310, 0.4183, 0.2610, 0.5567, 0.5546],
  5642. [0.6749, 0.4517, 0.8457, 0.3970, 0.3774, 0.4914, 0.6339, 0.5649]],
  5643. device='cuda:0', grad_fn=<AddmmBackward>)
  5644. landmarks are: tensor([[[0.6123, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  5645. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  5646. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  5647. [0.0000, 0.0000, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  5648. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  5649. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  5650. [0.6226, 0.4098, 0.8913, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  5651. [0.6250, 0.4105, 0.8700, 0.3717, 0.3587, 0.4967, 0.6037, 0.5167]]],
  5652. device='cuda:0')
  5653. loss_train_step before backward: tensor(0.0066, device='cuda:0', grad_fn=<MseLossBackward>)
  5654. loss_train_step after backward: tensor(0.0066, device='cuda:0', grad_fn=<MseLossBackward>)
  5655. loss_train: 0.27209125307854265
  5656. step: 75
  5657. running loss: 0.003627883374380569
  5658. Train Steps: 75/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5659. torch.Size([8, 8])
  5660. tensor([[0.6097, 0.3988, 0.8650, 0.5250, 0.4213, 0.5200, 0.5675, 0.5050],
  5661. [0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332],
  5662. [0.6212, 0.4159, 0.8675, 0.5783, 0.4088, 0.4317, 0.5613, 0.5917],
  5663. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  5664. [0.6085, 0.4008, 0.8588, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  5665. [0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  5666. [0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133],
  5667. [0.6346, 0.4144, 0.9088, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899]],
  5668. device='cuda:0', dtype=torch.float64)
  5669. predictions are: tensor([[0.6798, 0.4367, 0.8905, 0.5339, 0.4184, 0.5330, 0.5787, 0.5165],
  5670. [0.6814, 0.4195, 0.9055, 0.4748, 0.3659, 0.4066, 0.6685, 0.5359],
  5671. [0.6670, 0.4253, 0.8463, 0.5316, 0.3967, 0.4579, 0.5615, 0.5652],
  5672. [0.6028, 0.3785, 0.8851, 0.3418, 0.4280, 0.3274, 0.6745, 0.5367],
  5673. [0.6783, 0.4034, 0.8593, 0.4999, 0.4756, 0.4750, 0.5447, 0.5214],
  5674. [0.6905, 0.4342, 0.9075, 0.4974, 0.3788, 0.4901, 0.6273, 0.5221],
  5675. [0.7945, 0.5029, 0.8942, 0.5036, 0.3687, 0.4840, 0.6507, 0.5171],
  5676. [0.6484, 0.4060, 0.8931, 0.4606, 0.3984, 0.4084, 0.6942, 0.5413]],
  5677. device='cuda:0', grad_fn=<AddmmBackward>)
  5678. landmarks are: tensor([[[0.6097, 0.3988, 0.8650, 0.5250, 0.4212, 0.5200, 0.5675, 0.5050],
  5679. [0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332],
  5680. [0.6212, 0.4159, 0.8675, 0.5783, 0.4087, 0.4317, 0.5612, 0.5917],
  5681. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  5682. [0.6084, 0.4008, 0.8587, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  5683. [0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  5684. [0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133],
  5685. [0.6346, 0.4144, 0.9087, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899]]],
  5686. device='cuda:0')
  5687. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  5688. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  5689. loss_train: 0.27344133728183806
  5690. step: 76
  5691. running loss: 0.003597912332655764
  5692. Train Steps: 76/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5693. torch.Size([8, 8])
  5694. tensor([[0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  5695. [0.6198, 0.4101, 0.8838, 0.5283, 0.3763, 0.5267, 0.5913, 0.5567],
  5696. [0.6229, 0.4107, 0.8137, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  5697. [0.6072, 0.4029, 0.7037, 0.2150, 0.3912, 0.2267, 0.5516, 0.5507],
  5698. [0.6246, 0.4008, 0.8757, 0.5088, 0.4101, 0.5392, 0.6644, 0.5133],
  5699. [0.6250, 0.4103, 0.8950, 0.4400, 0.3912, 0.5650, 0.6050, 0.5133],
  5700. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  5701. [0.6168, 0.4029, 0.8523, 0.3417, 0.3588, 0.5000, 0.6125, 0.5400]],
  5702. device='cuda:0', dtype=torch.float64)
  5703. predictions are: tensor([[0.4302, 0.2588, 0.7816, 0.3117, 0.4392, 0.2675, 0.5540, 0.5430],
  5704. [0.6821, 0.4257, 0.9284, 0.6394, 0.3763, 0.5872, 0.6291, 0.5547],
  5705. [0.7801, 0.5038, 0.8543, 0.3555, 0.4846, 0.2312, 0.6044, 0.5342],
  5706. [0.7169, 0.4645, 0.7590, 0.2761, 0.4219, 0.2503, 0.6018, 0.5394],
  5707. [0.6197, 0.3730, 0.9462, 0.5948, 0.3995, 0.5777, 0.6588, 0.5362],
  5708. [0.7014, 0.4260, 0.9745, 0.5130, 0.4179, 0.6349, 0.6801, 0.5268],
  5709. [0.6904, 0.4121, 0.9245, 0.5656, 0.4052, 0.5342, 0.6547, 0.5416],
  5710. [0.6215, 0.3756, 0.9185, 0.4246, 0.3443, 0.5273, 0.6487, 0.5402]],
  5711. device='cuda:0', grad_fn=<AddmmBackward>)
  5712. landmarks are: tensor([[[0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  5713. [0.6198, 0.4101, 0.8838, 0.5283, 0.3762, 0.5267, 0.5913, 0.5567],
  5714. [0.6229, 0.4107, 0.8138, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  5715. [0.6072, 0.4029, 0.7038, 0.2150, 0.3913, 0.2267, 0.5516, 0.5507],
  5716. [0.6246, 0.4008, 0.8757, 0.5088, 0.4101, 0.5392, 0.6644, 0.5133],
  5717. [0.6250, 0.4103, 0.8950, 0.4400, 0.3913, 0.5650, 0.6050, 0.5133],
  5718. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  5719. [0.6168, 0.4029, 0.8523, 0.3417, 0.3587, 0.5000, 0.6125, 0.5400]]],
  5720. device='cuda:0')
  5721. loss_train_step before backward: tensor(0.0037, device='cuda:0', grad_fn=<MseLossBackward>)
  5722. loss_train_step after backward: tensor(0.0037, device='cuda:0', grad_fn=<MseLossBackward>)
  5723. loss_train: 0.27711278037168086
  5724. step: 77
  5725. running loss: 0.0035988672775542968
  5726. Train Steps: 77/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5727. torch.Size([8, 8])
  5728. tensor([[0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6262, 0.5167],
  5729. [ nan, nan, 0.7097, 0.2346, 0.4250, 0.1850, 0.5175, 0.5583],
  5730. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  5731. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5988, 0.5667],
  5732. [0.6261, 0.3987, 0.8688, 0.4917, 0.4300, 0.5333, 0.7010, 0.5309],
  5733. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317],
  5734. [0.6219, 0.3934, 0.8688, 0.5267, 0.4313, 0.4967, 0.5988, 0.4983],
  5735. [0.6136, 0.4085, 0.6688, 0.2317, 0.3862, 0.2367, 0.5517, 0.5783]],
  5736. device='cuda:0', dtype=torch.float64)
  5737. predictions are: tensor([[0.6488, 0.4085, 0.9239, 0.4528, 0.3710, 0.4379, 0.6246, 0.5328],
  5738. [0.1281, 0.0702, 0.7535, 0.2990, 0.4289, 0.2780, 0.5352, 0.5354],
  5739. [0.7497, 0.4416, 0.8740, 0.3162, 0.4783, 0.2598, 0.6582, 0.5157],
  5740. [0.7471, 0.4585, 0.8960, 0.4175, 0.3671, 0.4120, 0.5890, 0.5436],
  5741. [0.6345, 0.3725, 0.9412, 0.5425, 0.4347, 0.6288, 0.6831, 0.5394],
  5742. [0.7109, 0.4313, 0.9336, 0.5875, 0.3706, 0.4795, 0.5975, 0.5337],
  5743. [0.6542, 0.3831, 0.9247, 0.5839, 0.4145, 0.5602, 0.6040, 0.5012],
  5744. [0.6355, 0.3950, 0.7365, 0.2905, 0.3895, 0.3092, 0.5418, 0.5230]],
  5745. device='cuda:0', grad_fn=<AddmmBackward>)
  5746. landmarks are: tensor([[[0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6263, 0.5167],
  5747. [0.0000, 0.0000, 0.7097, 0.2346, 0.4250, 0.1850, 0.5175, 0.5583],
  5748. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  5749. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5987, 0.5667],
  5750. [0.6261, 0.3987, 0.8687, 0.4917, 0.4300, 0.5333, 0.7010, 0.5309],
  5751. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317],
  5752. [0.6219, 0.3934, 0.8687, 0.5267, 0.4313, 0.4967, 0.5987, 0.4983],
  5753. [0.6136, 0.4085, 0.6687, 0.2317, 0.3862, 0.2367, 0.5517, 0.5783]]],
  5754. device='cuda:0')
  5755. loss_train_step before backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  5756. loss_train_step after backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  5757. loss_train: 0.27983771497383714
  5758. step: 78
  5759. running loss: 0.0035876630124850916
  5760.  
  5761. Train Steps: 78/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5762. torch.Size([8, 8])
  5763. tensor([[0.6178, 0.4012, 0.8900, 0.4495, 0.3891, 0.5617, 0.5972, 0.4879],
  5764. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  5765. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  5766. [0.6317, 0.4038, 0.8287, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  5767. [0.6122, 0.4006, 0.8850, 0.4217, 0.4088, 0.5517, 0.6063, 0.5517],
  5768. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  5769. [0.6289, 0.4081, 0.8720, 0.3487, 0.3900, 0.3183, 0.6703, 0.5376],
  5770. [0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950]],
  5771. device='cuda:0', dtype=torch.float64)
  5772. predictions are: tensor([[0.6002, 0.3573, 0.9323, 0.4695, 0.4067, 0.5875, 0.6304, 0.5184],
  5773. [0.6629, 0.4128, 0.8068, 0.3113, 0.3855, 0.2987, 0.5554, 0.5189],
  5774. [0.5464, 0.3306, 0.9350, 0.4519, 0.4138, 0.4141, 0.6858, 0.5519],
  5775. [0.6337, 0.3700, 0.9142, 0.5837, 0.3704, 0.5209, 0.6383, 0.5275],
  5776. [0.5967, 0.3291, 0.9264, 0.4726, 0.4439, 0.5952, 0.5902, 0.5238],
  5777. [0.5697, 0.3110, 0.8690, 0.5939, 0.4222, 0.5291, 0.5780, 0.5865],
  5778. [0.6790, 0.4061, 0.9367, 0.3756, 0.4064, 0.3228, 0.6563, 0.5449],
  5779. [0.5955, 0.3229, 0.9189, 0.5381, 0.4464, 0.5383, 0.5872, 0.5075]],
  5780. device='cuda:0', grad_fn=<AddmmBackward>)
  5781. landmarks are: tensor([[[0.6178, 0.4012, 0.8900, 0.4495, 0.3891, 0.5617, 0.5972, 0.4879],
  5782. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  5783. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  5784. [0.6317, 0.4038, 0.8288, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  5785. [0.6122, 0.4006, 0.8850, 0.4217, 0.4087, 0.5517, 0.6062, 0.5517],
  5786. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  5787. [0.6289, 0.4081, 0.8720, 0.3487, 0.3900, 0.3183, 0.6703, 0.5376],
  5788. [0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950]]],
  5789. device='cuda:0')
  5790. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  5791. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  5792. loss_train: 0.28146227076649666
  5793. step: 79
  5794. running loss: 0.0035628135540062868
  5795. Train Steps: 79/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5796. torch.Size([8, 8])
  5797. tensor([[0.6346, 0.4165, 0.9138, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471],
  5798. [0.6264, 0.4049, 0.8988, 0.4633, 0.3813, 0.4983, 0.6326, 0.4843],
  5799. [ nan, nan, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  5800. [0.6240, 0.4217, 0.8150, 0.3133, 0.4425, 0.2650, 0.5650, 0.5817],
  5801. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  5802. [0.6261, 0.4066, 0.8325, 0.2150, 0.4763, 0.2667, 0.7002, 0.5633],
  5803. [0.6109, 0.4015, 0.7668, 0.3639, 0.3513, 0.3667, 0.5200, 0.5641],
  5804. [0.6271, 0.4040, 0.9138, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413]],
  5805. device='cuda:0', dtype=torch.float64)
  5806. predictions are: tensor([[0.5119, 0.3425, 0.9265, 0.4695, 0.3875, 0.4551, 0.6786, 0.5260],
  5807. [0.6343, 0.3938, 0.9262, 0.5178, 0.3744, 0.5502, 0.6083, 0.4991],
  5808. [0.1974, 0.1064, 0.7876, 0.3150, 0.3990, 0.3172, 0.5073, 0.5274],
  5809. [0.6346, 0.4311, 0.8191, 0.3264, 0.4551, 0.3047, 0.5770, 0.5565],
  5810. [0.6159, 0.3697, 0.8444, 0.6172, 0.3703, 0.5573, 0.5787, 0.4977],
  5811. [0.7068, 0.4193, 0.8361, 0.2817, 0.4721, 0.2977, 0.6492, 0.5381],
  5812. [0.6086, 0.3895, 0.8116, 0.4116, 0.3440, 0.3997, 0.5280, 0.5469],
  5813. [0.5484, 0.3277, 0.9446, 0.4219, 0.4743, 0.3307, 0.6781, 0.5497]],
  5814. device='cuda:0', grad_fn=<AddmmBackward>)
  5815. landmarks are: tensor([[[0.6346, 0.4165, 0.9137, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471],
  5816. [0.6264, 0.4049, 0.8988, 0.4633, 0.3812, 0.4983, 0.6326, 0.4843],
  5817. [0.0000, 0.0000, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  5818. [0.6240, 0.4217, 0.8150, 0.3133, 0.4425, 0.2650, 0.5650, 0.5817],
  5819. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  5820. [0.6261, 0.4066, 0.8325, 0.2150, 0.4762, 0.2667, 0.7002, 0.5633],
  5821. [0.6109, 0.4015, 0.7668, 0.3639, 0.3512, 0.3667, 0.5200, 0.5641],
  5822. [0.6271, 0.4040, 0.9137, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413]]],
  5823. device='cuda:0')
  5824. loss_train_step before backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  5825. loss_train_step after backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  5826. loss_train: 0.28381356899626553
  5827. step: 80
  5828. running loss: 0.0035476696124533192
  5829. Train Steps: 80/90 Loss: 0.0035 torch.Size([8, 600, 800])
  5830. torch.Size([8, 8])
  5831. tensor([[0.6357, 0.4097, 0.9038, 0.3883, 0.4213, 0.2950, 0.6686, 0.5390],
  5832. [0.6346, 0.4144, 0.9088, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  5833. [0.6159, 0.4085, 0.6900, 0.2283, 0.4088, 0.1950, 0.5123, 0.5397],
  5834. [0.6112, 0.4029, 0.8638, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  5835. [0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717],
  5836. [0.6178, 0.4059, 0.7525, 0.2250, 0.4313, 0.1783, 0.5404, 0.5240],
  5837. [ nan, nan, 0.6412, 0.1900, 0.4238, 0.1883, 0.5487, 0.5700],
  5838. [ nan, nan, 0.7240, 0.2722, 0.3900, 0.2567, 0.5168, 0.5933]],
  5839. device='cuda:0', dtype=torch.float64)
  5840. predictions are: tensor([[0.7318, 0.4761, 0.9288, 0.4338, 0.4220, 0.3790, 0.6864, 0.5237],
  5841. [0.7237, 0.4574, 0.9326, 0.5243, 0.3879, 0.5031, 0.6863, 0.5338],
  5842. [0.5232, 0.3285, 0.7327, 0.2817, 0.4121, 0.2591, 0.5487, 0.5335],
  5843. [0.6395, 0.4100, 0.9086, 0.5295, 0.4753, 0.5576, 0.5786, 0.5316],
  5844. [0.6670, 0.4423, 0.7967, 0.3256, 0.3888, 0.3411, 0.5705, 0.5651],
  5845. [0.3493, 0.2167, 0.7714, 0.2595, 0.4272, 0.2022, 0.5383, 0.5256],
  5846. [0.2071, 0.1384, 0.7526, 0.2718, 0.4370, 0.2432, 0.5629, 0.5550],
  5847. [0.3592, 0.2399, 0.7592, 0.3142, 0.3981, 0.2982, 0.5345, 0.5534]],
  5848. device='cuda:0', grad_fn=<AddmmBackward>)
  5849. landmarks are: tensor([[[0.6357, 0.4097, 0.9038, 0.3883, 0.4212, 0.2950, 0.6686, 0.5390],
  5850. [0.6346, 0.4144, 0.9087, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  5851. [0.6159, 0.4085, 0.6900, 0.2283, 0.4087, 0.1950, 0.5123, 0.5397],
  5852. [0.6112, 0.4029, 0.8637, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  5853. [0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717],
  5854. [0.6178, 0.4059, 0.7525, 0.2250, 0.4313, 0.1783, 0.5404, 0.5240],
  5855. [0.0000, 0.0000, 0.6413, 0.1900, 0.4238, 0.1883, 0.5487, 0.5700],
  5856. [0.0000, 0.0000, 0.7240, 0.2722, 0.3900, 0.2567, 0.5168, 0.5933]]],
  5857. device='cuda:0')
  5858. loss_train_step before backward: tensor(0.0076, device='cuda:0', grad_fn=<MseLossBackward>)
  5859. loss_train_step after backward: tensor(0.0076, device='cuda:0', grad_fn=<MseLossBackward>)
  5860. loss_train: 0.29142689728178084
  5861. step: 81
  5862. running loss: 0.003597862929404702
  5863. Train Steps: 81/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5864. torch.Size([8, 8])
  5865. tensor([[0.6248, 0.4185, 0.8500, 0.5767, 0.4463, 0.4550, 0.5613, 0.5917],
  5866. [0.6259, 0.4156, 0.8812, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960],
  5867. [0.6156, 0.4125, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  5868. [0.6299, 0.4008, 0.8450, 0.5350, 0.4213, 0.5000, 0.6350, 0.5100],
  5869. [0.6127, 0.4118, 0.8650, 0.5083, 0.4088, 0.5367, 0.5300, 0.5456],
  5870. [0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533],
  5871. [0.6264, 0.4071, 0.9038, 0.3867, 0.3663, 0.3917, 0.6338, 0.5283],
  5872. [0.6267, 0.4065, 0.8313, 0.2467, 0.4788, 0.1733, 0.6312, 0.5133]],
  5873. device='cuda:0', dtype=torch.float64)
  5874. predictions are: tensor([[0.5037, 0.3305, 0.8141, 0.5500, 0.4043, 0.4282, 0.5797, 0.5804],
  5875. [0.6664, 0.4200, 0.8563, 0.2748, 0.4429, 0.2194, 0.6277, 0.5476],
  5876. [0.5060, 0.3348, 0.8612, 0.4524, 0.4320, 0.5289, 0.5627, 0.5324],
  5877. [0.4562, 0.3147, 0.8126, 0.4894, 0.3906, 0.4595, 0.6114, 0.5158],
  5878. [0.4270, 0.2897, 0.8398, 0.4804, 0.4076, 0.5038, 0.5437, 0.5549],
  5879. [0.4705, 0.3189, 0.7083, 0.2346, 0.3568, 0.2952, 0.5956, 0.5650],
  5880. [0.5339, 0.3557, 0.8859, 0.3443, 0.3659, 0.3642, 0.6454, 0.5390],
  5881. [0.5492, 0.3757, 0.8003, 0.2359, 0.4695, 0.1793, 0.6644, 0.5451]],
  5882. device='cuda:0', grad_fn=<AddmmBackward>)
  5883. landmarks are: tensor([[[0.6248, 0.4185, 0.8500, 0.5767, 0.4462, 0.4550, 0.5612, 0.5917],
  5884. [0.6259, 0.4156, 0.8813, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960],
  5885. [0.6155, 0.4124, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  5886. [0.6299, 0.4008, 0.8450, 0.5350, 0.4212, 0.5000, 0.6350, 0.5100],
  5887. [0.6127, 0.4118, 0.8650, 0.5083, 0.4087, 0.5367, 0.5300, 0.5456],
  5888. [0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533],
  5889. [0.6264, 0.4071, 0.9038, 0.3867, 0.3663, 0.3917, 0.6338, 0.5283],
  5890. [0.6266, 0.4065, 0.8313, 0.2467, 0.4787, 0.1733, 0.6313, 0.5133]]],
  5891. device='cuda:0')
  5892. loss_train_step before backward: tensor(0.0033, device='cuda:0', grad_fn=<MseLossBackward>)
  5893. loss_train_step after backward: tensor(0.0033, device='cuda:0', grad_fn=<MseLossBackward>)
  5894. loss_train: 0.29475518641993403
  5895. step: 82
  5896. running loss: 0.003594575444145537
  5897.  
  5898. Train Steps: 82/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5899. torch.Size([8, 8])
  5900. tensor([[0.6198, 0.4115, 0.7762, 0.2717, 0.3713, 0.3200, 0.5837, 0.5683],
  5901. [0.6161, 0.4024, 0.8662, 0.4683, 0.4935, 0.5364, 0.6063, 0.5567],
  5902. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  5903. [ nan, nan, 0.7515, 0.2708, 0.3987, 0.2267, 0.5162, 0.5567],
  5904. [0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  5905. [0.6336, 0.4191, 0.8938, 0.5167, 0.3937, 0.3517, 0.7343, 0.5748],
  5906. [0.6060, 0.3924, 0.8450, 0.5717, 0.4200, 0.5217, 0.5253, 0.4752],
  5907. [0.6299, 0.4303, 0.7963, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222]],
  5908. device='cuda:0', dtype=torch.float64)
  5909. predictions are: tensor([[0.5219, 0.3660, 0.7790, 0.2321, 0.3643, 0.2673, 0.5809, 0.5409],
  5910. [0.5195, 0.3591, 0.8312, 0.3807, 0.4510, 0.4457, 0.5938, 0.5612],
  5911. [0.5682, 0.3913, 0.6999, 0.2407, 0.3874, 0.2158, 0.6085, 0.5697],
  5912. [0.1475, 0.1097, 0.6956, 0.2113, 0.4007, 0.1901, 0.5284, 0.5563],
  5913. [0.5345, 0.3710, 0.8218, 0.4980, 0.3978, 0.4215, 0.6464, 0.4959],
  5914. [0.5906, 0.4026, 0.8150, 0.4026, 0.3914, 0.3041, 0.6396, 0.5236],
  5915. [0.4948, 0.3323, 0.8203, 0.5143, 0.4422, 0.4626, 0.5775, 0.5119],
  5916. [0.6208, 0.4077, 0.7593, 0.2987, 0.4598, 0.2385, 0.5518, 0.5721]],
  5917. device='cuda:0', grad_fn=<AddmmBackward>)
  5918. landmarks are: tensor([[[0.6198, 0.4115, 0.7763, 0.2717, 0.3713, 0.3200, 0.5838, 0.5683],
  5919. [0.6161, 0.4024, 0.8662, 0.4683, 0.4935, 0.5364, 0.6062, 0.5567],
  5920. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  5921. [0.0000, 0.0000, 0.7515, 0.2708, 0.3988, 0.2267, 0.5163, 0.5567],
  5922. [0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  5923. [0.6336, 0.4191, 0.8938, 0.5167, 0.3938, 0.3517, 0.7343, 0.5748],
  5924. [0.6060, 0.3924, 0.8450, 0.5717, 0.4200, 0.5217, 0.5253, 0.4752],
  5925. [0.6299, 0.4303, 0.7962, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222]]],
  5926. device='cuda:0')
  5927. loss_train_step before backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  5928. loss_train_step after backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  5929. loss_train: 0.29796311957761645
  5930. step: 83
  5931. running loss: 0.0035899171033447765
  5932. Train Steps: 83/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5933. torch.Size([8, 8])
  5934. tensor([[0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166],
  5935. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  5936. [0.6110, 0.3984, 0.8750, 0.4933, 0.4625, 0.4950, 0.5578, 0.5676],
  5937. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  5938. [0.6152, 0.4131, 0.6863, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  5939. [0.6179, 0.4040, 0.7412, 0.1850, 0.3825, 0.2783, 0.5837, 0.5600],
  5940. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  5941. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150]],
  5942. device='cuda:0', dtype=torch.float64)
  5943. predictions are: tensor([[0.5181, 0.3539, 0.8390, 0.4595, 0.4247, 0.4804, 0.6175, 0.5306],
  5944. [0.4836, 0.3487, 0.7682, 0.2725, 0.3916, 0.2523, 0.6011, 0.5581],
  5945. [0.5561, 0.3936, 0.8207, 0.4423, 0.4917, 0.4533, 0.5457, 0.5479],
  5946. [0.5201, 0.3584, 0.8710, 0.3262, 0.3958, 0.2486, 0.6272, 0.5657],
  5947. [0.4927, 0.3735, 0.6764, 0.2544, 0.3768, 0.2560, 0.5766, 0.5698],
  5948. [0.4877, 0.3685, 0.7300, 0.2531, 0.3999, 0.2156, 0.5919, 0.5682],
  5949. [0.4900, 0.3573, 0.7261, 0.2373, 0.4414, 0.1238, 0.5944, 0.5538],
  5950. [0.5658, 0.4001, 0.8245, 0.4372, 0.4641, 0.5247, 0.6171, 0.5283]],
  5951. device='cuda:0', grad_fn=<AddmmBackward>)
  5952. landmarks are: tensor([[[0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166],
  5953. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  5954. [0.6110, 0.3984, 0.8750, 0.4933, 0.4625, 0.4950, 0.5578, 0.5676],
  5955. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  5956. [0.6152, 0.4131, 0.6862, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  5957. [0.6179, 0.4040, 0.7412, 0.1850, 0.3825, 0.2783, 0.5838, 0.5600],
  5958. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  5959. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150]]],
  5960. device='cuda:0')
  5961. loss_train_step before backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  5962. loss_train_step after backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  5963. loss_train: 0.3006193444598466
  5964. step: 84
  5965. running loss: 0.0035788017197600787
  5966. Train Steps: 84/90 Loss: 0.0036 torch.Size([8, 600, 800])
  5967. torch.Size([8, 8])
  5968. tensor([[0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  5969. [0.6190, 0.4135, 0.8000, 0.4883, 0.3566, 0.3647, 0.5613, 0.5900],
  5970. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  5971. [0.6275, 0.4050, 0.9038, 0.3767, 0.3838, 0.3533, 0.7074, 0.5575],
  5972. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167],
  5973. [0.6179, 0.4118, 0.7278, 0.4237, 0.3588, 0.3400, 0.5675, 0.5917],
  5974. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  5975. [0.6128, 0.4084, 0.8738, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397]],
  5976. device='cuda:0', dtype=torch.float64)
  5977. predictions are: tensor([[0.5401, 0.3604, 0.7687, 0.5018, 0.4050, 0.4576, 0.5831, 0.5146],
  5978. [0.6179, 0.4281, 0.7783, 0.3962, 0.3832, 0.3099, 0.5524, 0.5581],
  5979. [0.5014, 0.3646, 0.8290, 0.4376, 0.4013, 0.4111, 0.6210, 0.5631],
  5980. [0.5474, 0.3766, 0.8633, 0.2975, 0.4234, 0.2974, 0.6899, 0.5606],
  5981. [0.5735, 0.4356, 0.7765, 0.2658, 0.3967, 0.2545, 0.5893, 0.5601],
  5982. [0.5691, 0.4091, 0.7424, 0.3332, 0.3733, 0.2847, 0.5574, 0.5884],
  5983. [0.4941, 0.3615, 0.7728, 0.1914, 0.4640, 0.2244, 0.6177, 0.5739],
  5984. [0.5280, 0.3710, 0.8275, 0.4217, 0.3933, 0.3405, 0.5366, 0.5362]],
  5985. device='cuda:0', grad_fn=<AddmmBackward>)
  5986. landmarks are: tensor([[[0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  5987. [0.6190, 0.4135, 0.8000, 0.4883, 0.3566, 0.3647, 0.5612, 0.5900],
  5988. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  5989. [0.6275, 0.4050, 0.9038, 0.3767, 0.3837, 0.3533, 0.7074, 0.5575],
  5990. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167],
  5991. [0.6179, 0.4118, 0.7278, 0.4237, 0.3587, 0.3400, 0.5675, 0.5917],
  5992. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  5993. [0.6127, 0.4084, 0.8737, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397]]],
  5994. device='cuda:0')
  5995. loss_train_step before backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  5996. loss_train_step after backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  5997. loss_train: 0.3032755733001977
  5998. step: 85
  5999. running loss: 0.0035679479211787966
  6000. Train Steps: 85/90 Loss: 0.0036 torch.Size([8, 600, 800])
  6001. torch.Size([8, 8])
  6002. tensor([[0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  6003. [0.6307, 0.4029, 0.8988, 0.4817, 0.3937, 0.3500, 0.7311, 0.5378],
  6004. [0.6138, 0.4101, 0.8800, 0.5083, 0.4637, 0.5950, 0.5587, 0.5077],
  6005. [0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915],
  6006. [0.6133, 0.4094, 0.8495, 0.4028, 0.3588, 0.3200, 0.5003, 0.5407],
  6007. [0.6115, 0.3998, 0.7063, 0.2383, 0.4037, 0.1950, 0.5320, 0.4993],
  6008. [0.6236, 0.3967, 0.8675, 0.5400, 0.3862, 0.4517, 0.5825, 0.5200],
  6009. [0.6135, 0.3994, 0.7913, 0.3050, 0.3625, 0.3050, 0.5837, 0.5050]],
  6010. device='cuda:0', dtype=torch.float64)
  6011. predictions are: tensor([[0.5689, 0.3898, 0.6745, 0.2629, 0.3627, 0.2696, 0.5545, 0.5796],
  6012. [0.5396, 0.3793, 0.8791, 0.4103, 0.4065, 0.3312, 0.6782, 0.5538],
  6013. [0.6021, 0.4020, 0.8272, 0.4906, 0.4783, 0.5328, 0.5690, 0.5541],
  6014. [0.6105, 0.4080, 0.8580, 0.4126, 0.3853, 0.2962, 0.6221, 0.5206],
  6015. [0.6513, 0.4556, 0.8229, 0.3616, 0.3921, 0.2947, 0.5291, 0.5429],
  6016. [0.4928, 0.3361, 0.6731, 0.1926, 0.4042, 0.1822, 0.5665, 0.5280],
  6017. [0.5830, 0.4160, 0.8283, 0.5111, 0.3814, 0.4505, 0.5676, 0.5232],
  6018. [0.5854, 0.4016, 0.7740, 0.2581, 0.3595, 0.2813, 0.6416, 0.5413]],
  6019. device='cuda:0', grad_fn=<AddmmBackward>)
  6020. landmarks are: tensor([[[0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  6021. [0.6307, 0.4029, 0.8988, 0.4817, 0.3938, 0.3500, 0.7311, 0.5378],
  6022. [0.6138, 0.4101, 0.8800, 0.5083, 0.4638, 0.5950, 0.5587, 0.5077],
  6023. [0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915],
  6024. [0.6133, 0.4094, 0.8495, 0.4028, 0.3587, 0.3200, 0.5003, 0.5407],
  6025. [0.6115, 0.3998, 0.7063, 0.2383, 0.4038, 0.1950, 0.5320, 0.4993],
  6026. [0.6236, 0.3967, 0.8675, 0.5400, 0.3862, 0.4517, 0.5825, 0.5200],
  6027. [0.6135, 0.3994, 0.7912, 0.3050, 0.3625, 0.3050, 0.5838, 0.5050]]],
  6028. device='cuda:0')
  6029. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  6030. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  6031. loss_train: 0.3045844287844375
  6032. step: 86
  6033. running loss: 0.0035416794044702033
  6034.  
  6035. Train Steps: 86/90 Loss: 0.0035 torch.Size([8, 600, 800])
  6036. torch.Size([8, 8])
  6037. tensor([[0.6179, 0.4008, 0.8600, 0.4015, 0.3932, 0.2515, 0.5711, 0.5438],
  6038. [0.6267, 0.4080, 0.8438, 0.2633, 0.4763, 0.1800, 0.6259, 0.5240],
  6039. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  6040. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  6041. [0.6173, 0.4114, 0.7325, 0.2500, 0.4213, 0.1917, 0.5338, 0.5700],
  6042. [0.6175, 0.3957, 0.8700, 0.4817, 0.4662, 0.5133, 0.5800, 0.5517],
  6043. [ nan, nan, 0.6688, 0.2513, 0.4113, 0.2117, 0.5193, 0.5933],
  6044. [ nan, nan, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552]],
  6045. device='cuda:0', dtype=torch.float64)
  6046. predictions are: tensor([[0.6938, 0.4627, 0.7745, 0.3595, 0.3673, 0.2659, 0.5422, 0.5378],
  6047. [0.7675, 0.5108, 0.8090, 0.2835, 0.4165, 0.2215, 0.5889, 0.5316],
  6048. [0.7989, 0.5596, 0.8286, 0.5296, 0.3607, 0.5234, 0.5537, 0.5094],
  6049. [0.6746, 0.4583, 0.8689, 0.3938, 0.4130, 0.2602, 0.6126, 0.5131],
  6050. [0.4435, 0.2818, 0.6643, 0.2294, 0.3795, 0.2038, 0.5428, 0.5347],
  6051. [0.7686, 0.5124, 0.8225, 0.4737, 0.4038, 0.5151, 0.5450, 0.4887],
  6052. [0.3795, 0.2579, 0.6317, 0.2452, 0.3483, 0.2038, 0.5323, 0.5397],
  6053. [0.2499, 0.1584, 0.8249, 0.2532, 0.4639, 0.2239, 0.6906, 0.5346]],
  6054. device='cuda:0', grad_fn=<AddmmBackward>)
  6055. landmarks are: tensor([[[0.6179, 0.4008, 0.8600, 0.4015, 0.3932, 0.2515, 0.5711, 0.5438],
  6056. [0.6267, 0.4080, 0.8438, 0.2633, 0.4762, 0.1800, 0.6259, 0.5240],
  6057. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  6058. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  6059. [0.6173, 0.4114, 0.7325, 0.2500, 0.4212, 0.1917, 0.5337, 0.5700],
  6060. [0.6175, 0.3957, 0.8700, 0.4817, 0.4663, 0.5133, 0.5800, 0.5517],
  6061. [0.0000, 0.0000, 0.6688, 0.2513, 0.4112, 0.2117, 0.5193, 0.5933],
  6062. [0.0000, 0.0000, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552]]],
  6063. device='cuda:0')
  6064. loss_train_step before backward: tensor(0.0087, device='cuda:0', grad_fn=<MseLossBackward>)
  6065. loss_train_step after backward: tensor(0.0087, device='cuda:0', grad_fn=<MseLossBackward>)
  6066. loss_train: 0.31329219427425414
  6067. step: 87
  6068. running loss: 0.0036010597043017716
  6069. Train Steps: 87/90 Loss: 0.0036 torch.Size([8, 600, 800])
  6070. torch.Size([8, 8])
  6071. tensor([[0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  6072. [0.6136, 0.4085, 0.6688, 0.2317, 0.3862, 0.2367, 0.5517, 0.5783],
  6073. [0.6131, 0.4064, 0.8638, 0.5200, 0.4788, 0.4783, 0.5258, 0.5867],
  6074. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  6075. [0.6267, 0.4080, 0.8438, 0.2633, 0.4763, 0.1800, 0.6259, 0.5240],
  6076. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  6077. [0.6132, 0.4118, 0.8200, 0.3633, 0.3563, 0.5400, 0.5787, 0.5136],
  6078. [0.6126, 0.4073, 0.8750, 0.5133, 0.3800, 0.4333, 0.4986, 0.5378]],
  6079. device='cuda:0', dtype=torch.float64)
  6080. predictions are: tensor([[0.7099, 0.4488, 0.8918, 0.5150, 0.3862, 0.4957, 0.6400, 0.4723],
  6081. [0.5831, 0.3690, 0.6823, 0.2410, 0.3665, 0.2161, 0.5593, 0.5425],
  6082. [0.6480, 0.4269, 0.8681, 0.5214, 0.4285, 0.4433, 0.5788, 0.5561],
  6083. [0.7184, 0.4610, 0.8753, 0.4972, 0.4210, 0.5490, 0.6182, 0.5110],
  6084. [0.6838, 0.4470, 0.8567, 0.2835, 0.4509, 0.1811, 0.6441, 0.5405],
  6085. [0.5792, 0.3969, 0.7905, 0.3778, 0.3279, 0.3846, 0.5698, 0.5413],
  6086. [0.6387, 0.4136, 0.8354, 0.3753, 0.3431, 0.4887, 0.6001, 0.5222],
  6087. [0.6564, 0.4403, 0.8673, 0.5548, 0.3776, 0.4223, 0.5660, 0.5180]],
  6088. device='cuda:0', grad_fn=<AddmmBackward>)
  6089. landmarks are: tensor([[[0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  6090. [0.6136, 0.4085, 0.6687, 0.2317, 0.3862, 0.2367, 0.5517, 0.5783],
  6091. [0.6132, 0.4063, 0.8637, 0.5200, 0.4787, 0.4783, 0.5258, 0.5867],
  6092. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  6093. [0.6267, 0.4080, 0.8438, 0.2633, 0.4762, 0.1800, 0.6259, 0.5240],
  6094. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  6095. [0.6132, 0.4118, 0.8200, 0.3633, 0.3562, 0.5400, 0.5787, 0.5136],
  6096. [0.6126, 0.4073, 0.8750, 0.5133, 0.3800, 0.4333, 0.4986, 0.5378]]],
  6097. device='cuda:0')
  6098. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  6099. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  6100. loss_train: 0.3143081188900396
  6101. step: 88
  6102. running loss: 0.003571683169204996
  6103. Train Steps: 88/90 Loss: 0.0036 torch.Size([8, 600, 800])
  6104. torch.Size([8, 8])
  6105. tensor([[0.6229, 0.4198, 0.7662, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783],
  6106. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  6107. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  6108. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  6109. [0.6154, 0.4112, 0.7037, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  6110. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  6111. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  6112. [0.6140, 0.4070, 0.8700, 0.5000, 0.4612, 0.4900, 0.5260, 0.5852]],
  6113. device='cuda:0', dtype=torch.float64)
  6114. predictions are: tensor([[0.5503, 0.3433, 0.8042, 0.2656, 0.4390, 0.2441, 0.5905, 0.5752],
  6115. [0.4924, 0.2846, 0.7159, 0.2282, 0.3960, 0.2076, 0.5492, 0.5356],
  6116. [0.7602, 0.4789, 0.9267, 0.5469, 0.4050, 0.5384, 0.5807, 0.4911],
  6117. [0.7297, 0.4698, 0.8829, 0.5198, 0.3922, 0.5400, 0.6037, 0.5108],
  6118. [0.5250, 0.3183, 0.7260, 0.2407, 0.4251, 0.2064, 0.5545, 0.5222],
  6119. [0.6908, 0.4439, 0.9060, 0.5334, 0.3466, 0.3828, 0.5711, 0.5140],
  6120. [0.6375, 0.4331, 0.9175, 0.4850, 0.3439, 0.4380, 0.6430, 0.5255],
  6121. [0.7003, 0.4664, 0.9140, 0.5508, 0.4439, 0.5048, 0.5472, 0.5362]],
  6122. device='cuda:0', grad_fn=<AddmmBackward>)
  6123. landmarks are: tensor([[[0.6229, 0.4198, 0.7663, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783],
  6124. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  6125. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  6126. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  6127. [0.6154, 0.4112, 0.7038, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  6128. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  6129. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  6130. [0.6140, 0.4070, 0.8700, 0.5000, 0.4613, 0.4900, 0.5260, 0.5852]]],
  6131. device='cuda:0')
  6132. loss_train_step before backward: tensor(0.0025, device='cuda:0', grad_fn=<MseLossBackward>)
  6133. loss_train_step after backward: tensor(0.0025, device='cuda:0', grad_fn=<MseLossBackward>)
  6134. loss_train: 0.31682477600406855
  6135. step: 89
  6136. running loss: 0.0035598289438659387
  6137. Train Steps: 89/90 Loss: 0.0036 torch.Size([8, 600, 800])
  6138. torch.Size([8, 8])
  6139. tensor([[0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5837, 0.5317],
  6140. [0.6059, 0.4002, 0.7562, 0.2767, 0.3538, 0.3033, 0.5529, 0.5455],
  6141. [0.6272, 0.4071, 0.8738, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  6142. [0.6276, 0.4120, 0.8738, 0.3133, 0.4225, 0.2217, 0.6203, 0.4892],
  6143. [0.6289, 0.4019, 0.8113, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332],
  6144. [0.6196, 0.4088, 0.8888, 0.4583, 0.4500, 0.5683, 0.6138, 0.5883],
  6145. [ nan, nan, 0.7268, 0.2333, 0.4125, 0.1933, 0.5112, 0.5383],
  6146. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683]],
  6147. device='cuda:0', dtype=torch.float64)
  6148. predictions are: tensor([[0.6008, 0.3638, 0.8715, 0.3505, 0.3895, 0.3336, 0.5966, 0.5717],
  6149. [0.6301, 0.4162, 0.7736, 0.2896, 0.3732, 0.3400, 0.5808, 0.5527],
  6150. [0.7209, 0.4506, 0.9106, 0.5676, 0.3906, 0.4272, 0.5947, 0.5073],
  6151. [0.6660, 0.4047, 0.9162, 0.3639, 0.4761, 0.2391, 0.6249, 0.5336],
  6152. [0.7238, 0.4595, 0.8319, 0.5199, 0.3889, 0.5274, 0.6024, 0.5363],
  6153. [0.7352, 0.4680, 0.9079, 0.4925, 0.4598, 0.5879, 0.5617, 0.5313],
  6154. [0.2352, 0.1203, 0.7305, 0.2232, 0.4102, 0.2230, 0.5104, 0.5519],
  6155. [0.7288, 0.4646, 0.8875, 0.5587, 0.3863, 0.4118, 0.5313, 0.5509]],
  6156. device='cuda:0', grad_fn=<AddmmBackward>)
  6157. landmarks are: tensor([[[0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5838, 0.5317],
  6158. [0.6059, 0.4002, 0.7563, 0.2767, 0.3537, 0.3033, 0.5529, 0.5455],
  6159. [0.6272, 0.4071, 0.8737, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  6160. [0.6276, 0.4120, 0.8737, 0.3133, 0.4225, 0.2217, 0.6203, 0.4892],
  6161. [0.6289, 0.4019, 0.8112, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332],
  6162. [0.6196, 0.4088, 0.8888, 0.4583, 0.4500, 0.5683, 0.6137, 0.5883],
  6163. [0.0000, 0.0000, 0.7268, 0.2333, 0.4125, 0.1933, 0.5113, 0.5383],
  6164. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683]]],
  6165. device='cuda:0')
  6166. loss_train_step before backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  6167. loss_train_step after backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  6168. loss_train: 0.31965578126255423
  6169. step: 90
  6170. running loss: 0.003551730902917269
  6171.  
  6172. Valid Steps: 10/10 Loss: nan 36
  6173. --------------------------------------------------
  6174. Epoch: 2 Train Loss: 0.0036 Valid Loss: nan
  6175. --------------------------------------------------
  6176. size of train loader is: 90
  6177. torch.Size([8, 600, 800])
  6178. torch.Size([8, 8])
  6179. tensor([[0.6186, 0.4060, 0.8750, 0.5050, 0.3538, 0.4367, 0.5813, 0.6083],
  6180. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  6181. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  6182. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  6183. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  6184. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5513, 0.5750],
  6185. [0.6219, 0.3934, 0.8688, 0.5267, 0.4313, 0.4967, 0.5988, 0.4983],
  6186. [0.6200, 0.4049, 0.8638, 0.5617, 0.4125, 0.5100, 0.6013, 0.5317]],
  6187. device='cuda:0', dtype=torch.float64)
  6188. predictions are: tensor([[0.6002, 0.3815, 0.8933, 0.5012, 0.4067, 0.4831, 0.5530, 0.5773],
  6189. [0.6410, 0.4013, 0.9355, 0.4723, 0.4100, 0.4706, 0.5617, 0.5201],
  6190. [0.5869, 0.3487, 0.8109, 0.3059, 0.3688, 0.3311, 0.5628, 0.5760],
  6191. [0.6995, 0.4231, 0.9138, 0.5172, 0.4408, 0.4911, 0.5576, 0.5447],
  6192. [0.6746, 0.4112, 0.9438, 0.5046, 0.4605, 0.5980, 0.6225, 0.5350],
  6193. [0.3981, 0.2286, 0.7361, 0.2425, 0.4561, 0.2127, 0.5571, 0.5786],
  6194. [0.7015, 0.4287, 0.9322, 0.5515, 0.4549, 0.4995, 0.5993, 0.5136],
  6195. [0.6751, 0.4332, 0.8995, 0.5880, 0.4387, 0.5118, 0.5768, 0.5340]],
  6196. device='cuda:0', grad_fn=<AddmmBackward>)
  6197. landmarks are: tensor([[[0.6186, 0.4060, 0.8750, 0.5050, 0.3537, 0.4367, 0.5813, 0.6083],
  6198. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  6199. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  6200. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  6201. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  6202. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5512, 0.5750],
  6203. [0.6219, 0.3934, 0.8687, 0.5267, 0.4313, 0.4967, 0.5987, 0.4983],
  6204. [0.6199, 0.4049, 0.8637, 0.5617, 0.4125, 0.5100, 0.6012, 0.5317]]],
  6205. device='cuda:0')
  6206. loss_train_step before backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  6207. loss_train_step after backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  6208. loss_train: 0.002288524992763996
  6209. step: 1
  6210. running loss: 0.002288524992763996
  6211. Train Steps: 1/90 Loss: 0.0023 torch.Size([8, 600, 800])
  6212. torch.Size([8, 8])
  6213. tensor([[0.6145, 0.4008, 0.8750, 0.5383, 0.3975, 0.4650, 0.5563, 0.5533],
  6214. [0.6250, 0.4236, 0.8638, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  6215. [0.6030, 0.3969, 0.7988, 0.3917, 0.3450, 0.3667, 0.5266, 0.4700],
  6216. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  6217. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  6218. [0.6084, 0.3981, 0.8588, 0.5233, 0.4600, 0.5367, 0.5680, 0.5006],
  6219. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  6220. [0.6224, 0.4179, 0.8700, 0.5683, 0.4037, 0.4683, 0.5650, 0.5600]],
  6221. device='cuda:0', dtype=torch.float64)
  6222. predictions are: tensor([[0.5669, 0.3610, 0.8866, 0.5139, 0.4175, 0.4738, 0.5647, 0.5553],
  6223. [0.5978, 0.3696, 0.8827, 0.4233, 0.4464, 0.3263, 0.5813, 0.5786],
  6224. [0.5948, 0.3882, 0.8695, 0.4031, 0.3815, 0.3790, 0.5696, 0.5258],
  6225. [0.5551, 0.3560, 0.9126, 0.4513, 0.4443, 0.5119, 0.5792, 0.5305],
  6226. [0.4847, 0.2875, 0.7358, 0.2585, 0.4195, 0.2411, 0.5638, 0.5675],
  6227. [0.5462, 0.3488, 0.8821, 0.5407, 0.4778, 0.5131, 0.5600, 0.5296],
  6228. [0.6726, 0.4098, 0.9118, 0.5105, 0.4304, 0.5799, 0.6213, 0.5177],
  6229. [0.6760, 0.4343, 0.8956, 0.5829, 0.4141, 0.4797, 0.5930, 0.5445]],
  6230. device='cuda:0', grad_fn=<AddmmBackward>)
  6231. landmarks are: tensor([[[0.6145, 0.4008, 0.8750, 0.5383, 0.3975, 0.4650, 0.5562, 0.5533],
  6232. [0.6250, 0.4236, 0.8637, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  6233. [0.6030, 0.3969, 0.7987, 0.3917, 0.3450, 0.3667, 0.5266, 0.4700],
  6234. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  6235. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  6236. [0.6084, 0.3981, 0.8587, 0.5233, 0.4600, 0.5367, 0.5680, 0.5006],
  6237. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  6238. [0.6224, 0.4179, 0.8700, 0.5683, 0.4038, 0.4683, 0.5650, 0.5600]]],
  6239. device='cuda:0')
  6240. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  6241. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  6242. loss_train: 0.0036328728310763836
  6243. step: 2
  6244. running loss: 0.0018164364155381918
  6245. Train Steps: 2/90 Loss: 0.0018 torch.Size([8, 600, 800])
  6246. torch.Size([8, 8])
  6247. tensor([[0.6229, 0.4066, 0.8513, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  6248. [0.6267, 0.4080, 0.8438, 0.2633, 0.4763, 0.1800, 0.6259, 0.5240],
  6249. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  6250. [ nan, nan, 0.7850, 0.2700, 0.4288, 0.1717, 0.5199, 0.4999],
  6251. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167],
  6252. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  6253. [0.6264, 0.4248, 0.8938, 0.4183, 0.3875, 0.4100, 0.5400, 0.5350],
  6254. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290]],
  6255. device='cuda:0', dtype=torch.float64)
  6256. predictions are: tensor([[0.6359, 0.4082, 0.8793, 0.6344, 0.4539, 0.5514, 0.5787, 0.5307],
  6257. [0.6908, 0.4302, 0.8753, 0.3369, 0.4884, 0.2461, 0.6141, 0.5311],
  6258. [0.5758, 0.3436, 0.8435, 0.3540, 0.4212, 0.2781, 0.5877, 0.5331],
  6259. [0.3117, 0.1714, 0.7735, 0.2756, 0.4538, 0.2254, 0.5412, 0.5513],
  6260. [0.5630, 0.3703, 0.7986, 0.3373, 0.4174, 0.3504, 0.5790, 0.6113],
  6261. [0.6433, 0.3916, 0.7148, 0.3738, 0.3802, 0.3490, 0.5193, 0.5787],
  6262. [0.6158, 0.4063, 0.9110, 0.5084, 0.3871, 0.4496, 0.5415, 0.5369],
  6263. [0.5936, 0.3585, 0.8877, 0.3856, 0.3611, 0.4360, 0.6028, 0.5406]],
  6264. device='cuda:0', grad_fn=<AddmmBackward>)
  6265. landmarks are: tensor([[[0.6229, 0.4066, 0.8512, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  6266. [0.6267, 0.4080, 0.8438, 0.2633, 0.4762, 0.1800, 0.6259, 0.5240],
  6267. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  6268. [0.0000, 0.0000, 0.7850, 0.2700, 0.4288, 0.1717, 0.5199, 0.4999],
  6269. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167],
  6270. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  6271. [0.6264, 0.4248, 0.8938, 0.4183, 0.3875, 0.4100, 0.5400, 0.5350],
  6272. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290]]],
  6273. device='cuda:0')
  6274. loss_train_step before backward: tensor(0.0035, device='cuda:0', grad_fn=<MseLossBackward>)
  6275. loss_train_step after backward: tensor(0.0035, device='cuda:0', grad_fn=<MseLossBackward>)
  6276. loss_train: 0.00717794056981802
  6277. step: 3
  6278. running loss: 0.0023926468566060066
  6279. Train Steps: 3/90 Loss: 0.0024 torch.Size([8, 600, 800])
  6280. torch.Size([8, 8])
  6281. tensor([[0.6072, 0.4029, 0.7037, 0.2150, 0.3912, 0.2267, 0.5516, 0.5507],
  6282. [0.6206, 0.4001, 0.8900, 0.3933, 0.3588, 0.3567, 0.5837, 0.5083],
  6283. [0.6299, 0.4008, 0.8450, 0.5350, 0.4213, 0.5000, 0.6350, 0.5100],
  6284. [0.6203, 0.4073, 0.8189, 0.2398, 0.4400, 0.2054, 0.5929, 0.5501],
  6285. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290],
  6286. [0.6264, 0.4049, 0.8988, 0.4633, 0.3813, 0.4983, 0.6326, 0.4843],
  6287. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  6288. [0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5413, 0.5433]],
  6289. device='cuda:0', dtype=torch.float64)
  6290. predictions are: tensor([[0.6296, 0.4197, 0.6843, 0.2552, 0.4132, 0.2502, 0.5549, 0.5667],
  6291. [0.4891, 0.3154, 0.9120, 0.4422, 0.3716, 0.3887, 0.5510, 0.5275],
  6292. [0.5897, 0.3899, 0.8455, 0.5802, 0.4190, 0.5284, 0.5896, 0.5246],
  6293. [0.6052, 0.3880, 0.7896, 0.2977, 0.4430, 0.2461, 0.5919, 0.5701],
  6294. [0.5638, 0.3631, 0.7524, 0.2583, 0.4390, 0.2381, 0.6076, 0.5492],
  6295. [0.5158, 0.3366, 0.9171, 0.5129, 0.3831, 0.5358, 0.5797, 0.4969],
  6296. [0.5390, 0.3384, 0.7859, 0.3426, 0.3885, 0.3116, 0.5779, 0.5692],
  6297. [0.5654, 0.3864, 0.8075, 0.4019, 0.3687, 0.3412, 0.5067, 0.5633]],
  6298. device='cuda:0', grad_fn=<AddmmBackward>)
  6299. landmarks are: tensor([[[0.6072, 0.4029, 0.7038, 0.2150, 0.3913, 0.2267, 0.5516, 0.5507],
  6300. [0.6206, 0.4001, 0.8900, 0.3933, 0.3587, 0.3567, 0.5838, 0.5083],
  6301. [0.6299, 0.4008, 0.8450, 0.5350, 0.4212, 0.5000, 0.6350, 0.5100],
  6302. [0.6203, 0.4073, 0.8189, 0.2398, 0.4400, 0.2054, 0.5929, 0.5501],
  6303. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290],
  6304. [0.6264, 0.4049, 0.8988, 0.4633, 0.3812, 0.4983, 0.6326, 0.4843],
  6305. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  6306. [0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5412, 0.5433]]],
  6307. device='cuda:0')
  6308. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  6309. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  6310. loss_train: 0.008925004513002932
  6311. step: 4
  6312. running loss: 0.002231251128250733
  6313.  
  6314. Train Steps: 4/90 Loss: 0.0022 torch.Size([8, 600, 800])
  6315. torch.Size([8, 8])
  6316. tensor([[0.6275, 0.4050, 0.9038, 0.3767, 0.3838, 0.3533, 0.7074, 0.5575],
  6317. [0.6339, 0.4149, 0.8800, 0.5000, 0.3900, 0.5283, 0.7541, 0.5424],
  6318. [0.6257, 0.4034, 0.8287, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  6319. [0.6178, 0.4012, 0.8900, 0.4495, 0.3891, 0.5617, 0.5972, 0.4879],
  6320. [0.6277, 0.4057, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  6321. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667],
  6322. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6362, 0.5200],
  6323. [0.6129, 0.4063, 0.8738, 0.5250, 0.4313, 0.4733, 0.5230, 0.5874]],
  6324. device='cuda:0', dtype=torch.float64)
  6325. predictions are: tensor([[0.6101, 0.4008, 0.8775, 0.3697, 0.3979, 0.3186, 0.6877, 0.5551],
  6326. [0.5939, 0.4007, 0.8892, 0.5176, 0.3852, 0.5240, 0.6640, 0.5635],
  6327. [0.6112, 0.4214, 0.7873, 0.2553, 0.3898, 0.2494, 0.6111, 0.5374],
  6328. [0.5689, 0.3956, 0.8435, 0.4460, 0.3896, 0.5423, 0.5555, 0.5107],
  6329. [0.6235, 0.4064, 0.7775, 0.2624, 0.4293, 0.2045, 0.6141, 0.5317],
  6330. [0.5731, 0.3912, 0.8131, 0.3550, 0.3671, 0.3545, 0.5398, 0.5531],
  6331. [0.5109, 0.3331, 0.8373, 0.3655, 0.3707, 0.5275, 0.5966, 0.5326],
  6332. [0.6127, 0.4086, 0.8259, 0.5347, 0.4222, 0.4452, 0.5156, 0.5748]],
  6333. device='cuda:0', grad_fn=<AddmmBackward>)
  6334. landmarks are: tensor([[[0.6275, 0.4050, 0.9038, 0.3767, 0.3837, 0.3533, 0.7074, 0.5575],
  6335. [0.6339, 0.4149, 0.8800, 0.5000, 0.3900, 0.5283, 0.7541, 0.5424],
  6336. [0.6257, 0.4034, 0.8288, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  6337. [0.6178, 0.4012, 0.8900, 0.4495, 0.3891, 0.5617, 0.5972, 0.4879],
  6338. [0.6277, 0.4056, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  6339. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667],
  6340. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6363, 0.5200],
  6341. [0.6130, 0.4063, 0.8737, 0.5250, 0.4313, 0.4733, 0.5230, 0.5874]]],
  6342. device='cuda:0')
  6343. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  6344. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  6345. loss_train: 0.009910384309478104
  6346. step: 5
  6347. running loss: 0.001982076861895621
  6348. Train Steps: 5/90 Loss: 0.0020 torch.Size([8, 600, 800])
  6349. torch.Size([8, 8])
  6350. tensor([[0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  6351. [0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767],
  6352. [0.6193, 0.3930, 0.8949, 0.4437, 0.3852, 0.5435, 0.6263, 0.5263],
  6353. [0.6200, 0.3961, 0.8461, 0.5497, 0.4142, 0.4577, 0.5892, 0.5402],
  6354. [0.6275, 0.4024, 0.8500, 0.5383, 0.3912, 0.4883, 0.6288, 0.5100],
  6355. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  6356. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  6357. [0.6182, 0.4099, 0.7812, 0.3000, 0.3937, 0.2367, 0.5325, 0.5750]],
  6358. device='cuda:0', dtype=torch.float64)
  6359. predictions are: tensor([[0.6028, 0.4081, 0.8787, 0.4410, 0.3613, 0.3906, 0.6559, 0.5075],
  6360. [0.5579, 0.3736, 0.8574, 0.4347, 0.3937, 0.4630, 0.5563, 0.5520],
  6361. [0.6174, 0.4114, 0.8715, 0.4025, 0.3697, 0.5316, 0.6222, 0.5028],
  6362. [0.5251, 0.3515, 0.8400, 0.4720, 0.3916, 0.4560, 0.5975, 0.5433],
  6363. [0.6194, 0.4004, 0.8417, 0.4841, 0.3773, 0.4897, 0.6316, 0.4948],
  6364. [0.4978, 0.3611, 0.7664, 0.2651, 0.3495, 0.2716, 0.5151, 0.5461],
  6365. [0.6735, 0.4743, 0.7903, 0.2001, 0.4035, 0.2606, 0.6323, 0.5188],
  6366. [0.5516, 0.3580, 0.7681, 0.2637, 0.4117, 0.2454, 0.5465, 0.5668]],
  6367. device='cuda:0', grad_fn=<AddmmBackward>)
  6368. landmarks are: tensor([[[0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  6369. [0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767],
  6370. [0.6193, 0.3930, 0.8949, 0.4437, 0.3852, 0.5435, 0.6263, 0.5263],
  6371. [0.6200, 0.3961, 0.8461, 0.5497, 0.4142, 0.4577, 0.5892, 0.5402],
  6372. [0.6275, 0.4024, 0.8500, 0.5383, 0.3913, 0.4883, 0.6288, 0.5100],
  6373. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  6374. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  6375. [0.6182, 0.4099, 0.7812, 0.3000, 0.3938, 0.2367, 0.5325, 0.5750]]],
  6376. device='cuda:0')
  6377. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  6378. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  6379. loss_train: 0.011184405419044197
  6380. step: 6
  6381. running loss: 0.0018640675698406994
  6382. Train Steps: 6/90 Loss: 0.0019 torch.Size([8, 600, 800])
  6383. torch.Size([8, 8])
  6384. tensor([[0.6152, 0.4131, 0.6863, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  6385. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5463, 0.5800],
  6386. [0.6154, 0.4112, 0.7037, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  6387. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  6388. [0.6185, 0.4042, 0.7700, 0.2250, 0.4062, 0.2117, 0.5763, 0.5150],
  6389. [0.6234, 0.4179, 0.7825, 0.3450, 0.3813, 0.2867, 0.5675, 0.5617],
  6390. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  6391. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672]],
  6392. device='cuda:0', dtype=torch.float64)
  6393. predictions are: tensor([[0.6380, 0.4439, 0.7222, 0.2509, 0.3482, 0.3219, 0.5899, 0.5464],
  6394. [0.5228, 0.3732, 0.7454, 0.2522, 0.3416, 0.2843, 0.5357, 0.5352],
  6395. [0.6708, 0.4600, 0.7047, 0.2072, 0.4197, 0.1954, 0.5537, 0.5195],
  6396. [0.6443, 0.4543, 0.8250, 0.2313, 0.3901, 0.2928, 0.6525, 0.5063],
  6397. [0.6395, 0.4302, 0.7451, 0.2097, 0.3914, 0.2084, 0.6203, 0.5268],
  6398. [0.5345, 0.3629, 0.7980, 0.3231, 0.3868, 0.2772, 0.6029, 0.5457],
  6399. [0.4908, 0.3311, 0.9332, 0.4952, 0.3968, 0.5405, 0.6285, 0.5112],
  6400. [0.5857, 0.3792, 0.9133, 0.4411, 0.3189, 0.3980, 0.6537, 0.4978]],
  6401. device='cuda:0', grad_fn=<AddmmBackward>)
  6402. landmarks are: tensor([[[0.6152, 0.4131, 0.6862, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  6403. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5462, 0.5800],
  6404. [0.6154, 0.4112, 0.7038, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  6405. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  6406. [0.6184, 0.4042, 0.7700, 0.2250, 0.4062, 0.2117, 0.5763, 0.5150],
  6407. [0.6234, 0.4179, 0.7825, 0.3450, 0.3812, 0.2867, 0.5675, 0.5617],
  6408. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  6409. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672]]],
  6410. device='cuda:0')
  6411. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  6412. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  6413. loss_train: 0.012649990734644234
  6414. step: 7
  6415. running loss: 0.0018071415335206048
  6416. Train Steps: 7/90 Loss: 0.0018 torch.Size([8, 600, 800])
  6417. torch.Size([8, 8])
  6418. tensor([[0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  6419. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  6420. [0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6138, 0.5400],
  6421. [0.6254, 0.3993, 0.8988, 0.4767, 0.3987, 0.5517, 0.6955, 0.5285],
  6422. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  6423. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791],
  6424. [0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  6425. [0.6261, 0.3987, 0.8688, 0.4917, 0.4300, 0.5333, 0.7010, 0.5309]],
  6426. device='cuda:0', dtype=torch.float64)
  6427. predictions are: tensor([[0.6730, 0.4779, 0.7829, 0.1793, 0.3983, 0.2202, 0.6320, 0.5176],
  6428. [0.6574, 0.4479, 0.8553, 0.2471, 0.3966, 0.2033, 0.6424, 0.5357],
  6429. [0.5816, 0.3918, 0.8557, 0.3419, 0.3306, 0.4048, 0.6079, 0.5148],
  6430. [0.5688, 0.3792, 0.8812, 0.4301, 0.3675, 0.5155, 0.6587, 0.5246],
  6431. [0.6893, 0.4635, 0.6720, 0.2585, 0.3296, 0.2493, 0.5315, 0.5631],
  6432. [0.5993, 0.3999, 0.8540, 0.4222, 0.3700, 0.4330, 0.6404, 0.4731],
  6433. [0.5385, 0.3690, 0.8455, 0.4344, 0.3897, 0.4744, 0.5790, 0.5234],
  6434. [0.6363, 0.4504, 0.8473, 0.4180, 0.4055, 0.4664, 0.6760, 0.5129]],
  6435. device='cuda:0', grad_fn=<AddmmBackward>)
  6436. landmarks are: tensor([[[0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  6437. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  6438. [0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6137, 0.5400],
  6439. [0.6254, 0.3993, 0.8988, 0.4767, 0.3988, 0.5517, 0.6955, 0.5285],
  6440. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  6441. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791],
  6442. [0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  6443. [0.6261, 0.3987, 0.8687, 0.4917, 0.4300, 0.5333, 0.7010, 0.5309]]],
  6444. device='cuda:0')
  6445. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  6446. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  6447. loss_train: 0.014307374716736376
  6448. step: 8
  6449. running loss: 0.001788421839592047
  6450.  
  6451. Train Steps: 8/90 Loss: 0.0018 torch.Size([8, 600, 800])
  6452. torch.Size([8, 8])
  6453. tensor([[0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  6454. [0.6270, 0.4267, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  6455. [0.6192, 0.4128, 0.8513, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  6456. [0.6197, 0.4118, 0.8688, 0.5517, 0.4037, 0.5233, 0.5875, 0.5600],
  6457. [0.6336, 0.4191, 0.8938, 0.5167, 0.3937, 0.3517, 0.7343, 0.5748],
  6458. [0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5413, 0.5433],
  6459. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  6460. [0.6126, 0.3954, 0.8538, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350]],
  6461. device='cuda:0', dtype=torch.float64)
  6462. predictions are: tensor([[0.6075, 0.4180, 0.8209, 0.1754, 0.4414, 0.2094, 0.7048, 0.5392],
  6463. [0.5856, 0.4044, 0.6639, 0.1650, 0.3894, 0.1484, 0.5538, 0.5401],
  6464. [0.6596, 0.4292, 0.8553, 0.4984, 0.3863, 0.5089, 0.6441, 0.4961],
  6465. [0.6291, 0.4204, 0.8378, 0.4888, 0.3686, 0.5141, 0.6246, 0.5025],
  6466. [0.6524, 0.4352, 0.8420, 0.4286, 0.3649, 0.2985, 0.6814, 0.5026],
  6467. [0.6159, 0.4232, 0.7951, 0.2934, 0.3295, 0.2710, 0.5499, 0.5150],
  6468. [0.6264, 0.4401, 0.7368, 0.1997, 0.3638, 0.2359, 0.5624, 0.5183],
  6469. [0.5910, 0.3714, 0.8494, 0.4495, 0.3998, 0.4437, 0.6143, 0.4924]],
  6470. device='cuda:0', grad_fn=<AddmmBackward>)
  6471. landmarks are: tensor([[[0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  6472. [0.6270, 0.4266, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  6473. [0.6192, 0.4128, 0.8512, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  6474. [0.6197, 0.4118, 0.8687, 0.5517, 0.4038, 0.5233, 0.5875, 0.5600],
  6475. [0.6336, 0.4191, 0.8938, 0.5167, 0.3938, 0.3517, 0.7343, 0.5748],
  6476. [0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5412, 0.5433],
  6477. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  6478. [0.6126, 0.3954, 0.8537, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350]]],
  6479. device='cuda:0')
  6480. loss_train_step before backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  6481. loss_train_step after backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  6482. loss_train: 0.016634083236567676
  6483. step: 9
  6484. running loss: 0.0018482314707297417
  6485. Train Steps: 9/90 Loss: 0.0018 torch.Size([8, 600, 800])
  6486. torch.Size([8, 8])
  6487. tensor([[0.6185, 0.4080, 0.8625, 0.3483, 0.3788, 0.2650, 0.5320, 0.5272],
  6488. [0.6219, 0.3934, 0.8688, 0.5267, 0.4313, 0.4967, 0.5988, 0.4983],
  6489. [0.6140, 0.4070, 0.8700, 0.5000, 0.4612, 0.4900, 0.5260, 0.5852],
  6490. [0.6100, 0.4016, 0.8600, 0.5067, 0.4612, 0.5233, 0.5086, 0.5519],
  6491. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650],
  6492. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  6493. [0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050],
  6494. [0.6182, 0.4099, 0.7812, 0.3000, 0.3937, 0.2367, 0.5325, 0.5750]],
  6495. device='cuda:0', dtype=torch.float64)
  6496. predictions are: tensor([[0.6524, 0.4339, 0.8155, 0.3268, 0.3706, 0.2373, 0.5624, 0.5438],
  6497. [0.5945, 0.3936, 0.8238, 0.4429, 0.4082, 0.4306, 0.6347, 0.4983],
  6498. [0.5720, 0.4040, 0.8387, 0.4393, 0.4390, 0.4357, 0.5802, 0.5505],
  6499. [0.6188, 0.4136, 0.8323, 0.4499, 0.4364, 0.4505, 0.5917, 0.5264],
  6500. [0.6634, 0.4575, 0.7849, 0.3285, 0.3516, 0.2794, 0.5788, 0.5534],
  6501. [0.5991, 0.3866, 0.8214, 0.4122, 0.3813, 0.4463, 0.6662, 0.5194],
  6502. [0.6810, 0.4477, 0.8417, 0.3520, 0.3407, 0.4455, 0.6495, 0.5353],
  6503. [0.6106, 0.3994, 0.7506, 0.2554, 0.3855, 0.2089, 0.5507, 0.5628]],
  6504. device='cuda:0', grad_fn=<AddmmBackward>)
  6505. landmarks are: tensor([[[0.6186, 0.4080, 0.8625, 0.3483, 0.3787, 0.2650, 0.5320, 0.5272],
  6506. [0.6219, 0.3934, 0.8687, 0.5267, 0.4313, 0.4967, 0.5987, 0.4983],
  6507. [0.6140, 0.4070, 0.8700, 0.5000, 0.4613, 0.4900, 0.5260, 0.5852],
  6508. [0.6100, 0.4016, 0.8600, 0.5067, 0.4613, 0.5233, 0.5086, 0.5519],
  6509. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650],
  6510. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  6511. [0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050],
  6512. [0.6182, 0.4099, 0.7812, 0.3000, 0.3938, 0.2367, 0.5325, 0.5750]]],
  6513. device='cuda:0')
  6514. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  6515. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  6516. loss_train: 0.018363903160206974
  6517. step: 10
  6518. running loss: 0.0018363903160206973
  6519. Train Steps: 10/90 Loss: 0.0018 torch.Size([8, 600, 800])
  6520. torch.Size([8, 8])
  6521. tensor([[0.6317, 0.4038, 0.8287, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  6522. [0.6179, 0.4118, 0.7278, 0.4237, 0.3588, 0.3400, 0.5675, 0.5917],
  6523. [0.6064, 0.4019, 0.8650, 0.4517, 0.4037, 0.5367, 0.5703, 0.5609],
  6524. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  6525. [0.6286, 0.4086, 0.8408, 0.2801, 0.4163, 0.2800, 0.6725, 0.5393],
  6526. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  6527. [0.6127, 0.4084, 0.8700, 0.4467, 0.3987, 0.4317, 0.5013, 0.5471],
  6528. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466]],
  6529. device='cuda:0', dtype=torch.float64)
  6530. predictions are: tensor([[0.6670, 0.4386, 0.8303, 0.5567, 0.4043, 0.4091, 0.6370, 0.5482],
  6531. [0.6776, 0.4518, 0.7636, 0.4052, 0.3734, 0.2846, 0.5586, 0.5967],
  6532. [0.6549, 0.4292, 0.8408, 0.4363, 0.4181, 0.5187, 0.6062, 0.5413],
  6533. [0.6972, 0.4272, 0.8284, 0.4156, 0.3770, 0.4023, 0.5834, 0.5589],
  6534. [0.6584, 0.4397, 0.8381, 0.2707, 0.4301, 0.2266, 0.6898, 0.5920],
  6535. [0.6660, 0.4326, 0.8762, 0.4353, 0.4584, 0.5287, 0.6234, 0.5476],
  6536. [0.5939, 0.3899, 0.8410, 0.4262, 0.3975, 0.3439, 0.5154, 0.5441],
  6537. [0.5964, 0.3956, 0.8295, 0.4933, 0.4590, 0.4670, 0.5879, 0.5507]],
  6538. device='cuda:0', grad_fn=<AddmmBackward>)
  6539. landmarks are: tensor([[[0.6317, 0.4038, 0.8288, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  6540. [0.6179, 0.4118, 0.7278, 0.4237, 0.3587, 0.3400, 0.5675, 0.5917],
  6541. [0.6064, 0.4019, 0.8650, 0.4517, 0.4038, 0.5367, 0.5703, 0.5609],
  6542. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  6543. [0.6286, 0.4086, 0.8408, 0.2801, 0.4162, 0.2800, 0.6725, 0.5393],
  6544. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  6545. [0.6127, 0.4084, 0.8700, 0.4467, 0.3988, 0.4317, 0.5013, 0.5471],
  6546. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466]]],
  6547. device='cuda:0')
  6548. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  6549. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  6550. loss_train: 0.019579105195589364
  6551. step: 11
  6552. running loss: 0.0017799186541444876
  6553. Train Steps: 11/90 Loss: 0.0018 torch.Size([8, 600, 800])
  6554. torch.Size([8, 8])
  6555. tensor([[0.6200, 0.4098, 0.8237, 0.2917, 0.4012, 0.2967, 0.6000, 0.5683],
  6556. [0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500],
  6557. [0.6053, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  6558. [0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  6559. [0.6147, 0.4026, 0.6600, 0.2467, 0.4088, 0.2150, 0.5489, 0.5773],
  6560. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  6561. [0.6137, 0.4038, 0.8563, 0.4050, 0.3813, 0.2550, 0.5106, 0.4954],
  6562. [0.6200, 0.3999, 0.8653, 0.5207, 0.4100, 0.5125, 0.5975, 0.5103]],
  6563. device='cuda:0', dtype=torch.float64)
  6564. predictions are: tensor([[0.6627, 0.4582, 0.8446, 0.3431, 0.4621, 0.3512, 0.6252, 0.6020],
  6565. [0.6581, 0.4269, 0.7875, 0.3111, 0.3887, 0.4321, 0.5941, 0.5865],
  6566. [0.6116, 0.3716, 0.7028, 0.2405, 0.4561, 0.2085, 0.5293, 0.5720],
  6567. [0.6217, 0.4010, 0.9149, 0.5132, 0.4004, 0.4406, 0.5760, 0.5551],
  6568. [0.7454, 0.4856, 0.7083, 0.3050, 0.4564, 0.2597, 0.4888, 0.6232],
  6569. [0.6429, 0.3958, 0.9150, 0.5988, 0.4104, 0.5090, 0.6470, 0.5615],
  6570. [0.6573, 0.4318, 0.8670, 0.4660, 0.4384, 0.3084, 0.5093, 0.5605],
  6571. [0.6276, 0.4006, 0.9061, 0.5831, 0.4815, 0.5650, 0.5835, 0.5617]],
  6572. device='cuda:0', grad_fn=<AddmmBackward>)
  6573. landmarks are: tensor([[[0.6200, 0.4098, 0.8238, 0.2917, 0.4013, 0.2967, 0.6000, 0.5683],
  6574. [0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500],
  6575. [0.6054, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  6576. [0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  6577. [0.6147, 0.4026, 0.6600, 0.2467, 0.4087, 0.2150, 0.5489, 0.5773],
  6578. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  6579. [0.6137, 0.4038, 0.8562, 0.4050, 0.3812, 0.2550, 0.5106, 0.4954],
  6580. [0.6200, 0.3999, 0.8653, 0.5207, 0.4100, 0.5125, 0.5975, 0.5103]]],
  6581. device='cuda:0')
  6582. loss_train_step before backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  6583. loss_train_step after backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  6584. loss_train: 0.021589155890978873
  6585. step: 12
  6586. running loss: 0.0017990963242482394
  6587.  
  6588. Train Steps: 12/90 Loss: 0.0018 torch.Size([8, 600, 800])
  6589. torch.Size([8, 8])
  6590. tensor([[0.6043, 0.4022, 0.6887, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136],
  6591. [0.6273, 0.4143, 0.8750, 0.5700, 0.3987, 0.4717, 0.6013, 0.5467],
  6592. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  6593. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5837, 0.5500],
  6594. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  6595. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  6596. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  6597. [0.6118, 0.4052, 0.8463, 0.3917, 0.3538, 0.3450, 0.5053, 0.5593]],
  6598. device='cuda:0', dtype=torch.float64)
  6599. predictions are: tensor([[0.6850, 0.4247, 0.7308, 0.2875, 0.4270, 0.2783, 0.5280, 0.5645],
  6600. [0.6193, 0.3962, 0.8800, 0.6018, 0.4218, 0.5049, 0.5980, 0.5801],
  6601. [0.6418, 0.3852, 0.8744, 0.3716, 0.3916, 0.4108, 0.6206, 0.5730],
  6602. [0.6372, 0.4026, 0.9109, 0.5178, 0.4621, 0.5772, 0.5763, 0.5733],
  6603. [0.6267, 0.3887, 0.8751, 0.5489, 0.4650, 0.5104, 0.5475, 0.5908],
  6604. [0.6339, 0.4063, 0.8848, 0.5980, 0.4351, 0.4774, 0.5554, 0.5471],
  6605. [0.6724, 0.4247, 0.7273, 0.3193, 0.4394, 0.2667, 0.5483, 0.5913],
  6606. [0.6507, 0.4327, 0.8642, 0.4348, 0.3746, 0.3852, 0.4934, 0.6010]],
  6607. device='cuda:0', grad_fn=<AddmmBackward>)
  6608. landmarks are: tensor([[[0.6043, 0.4022, 0.6888, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136],
  6609. [0.6273, 0.4143, 0.8750, 0.5700, 0.3988, 0.4717, 0.6012, 0.5467],
  6610. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  6611. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5838, 0.5500],
  6612. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  6613. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  6614. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  6615. [0.6118, 0.4052, 0.8462, 0.3917, 0.3537, 0.3450, 0.5053, 0.5593]]],
  6616. device='cuda:0')
  6617. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  6618. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  6619. loss_train: 0.022897822898812592
  6620. step: 13
  6621. running loss: 0.0017613709922163533
  6622. Train Steps: 13/90 Loss: 0.0018 torch.Size([8, 600, 800])
  6623. torch.Size([8, 8])
  6624. tensor([[0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  6625. [ nan, nan, 0.7648, 0.2722, 0.3962, 0.2183, 0.5060, 0.5422],
  6626. [ nan, nan, 0.7612, 0.3250, 0.4037, 0.2533, 0.5438, 0.5767],
  6627. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  6628. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  6629. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  6630. [0.6262, 0.4052, 0.8888, 0.4700, 0.3675, 0.5117, 0.6350, 0.5233],
  6631. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209]],
  6632. device='cuda:0', dtype=torch.float64)
  6633. predictions are: tensor([[0.7103, 0.4486, 0.8747, 0.4249, 0.3902, 0.5114, 0.6438, 0.5376],
  6634. [0.3599, 0.2209, 0.7311, 0.3070, 0.3967, 0.2738, 0.4546, 0.5411],
  6635. [0.4195, 0.2813, 0.7883, 0.3753, 0.4167, 0.2952, 0.4734, 0.5861],
  6636. [0.6465, 0.4145, 0.8845, 0.5313, 0.3612, 0.4709, 0.6170, 0.5398],
  6637. [0.6977, 0.4486, 0.8552, 0.6218, 0.4135, 0.4887, 0.5198, 0.5217],
  6638. [0.6612, 0.4368, 0.8415, 0.3068, 0.4765, 0.3075, 0.6446, 0.5897],
  6639. [0.7176, 0.4319, 0.9126, 0.5278, 0.3973, 0.5687, 0.6204, 0.5488],
  6640. [0.7001, 0.4696, 0.8438, 0.5102, 0.4606, 0.3406, 0.4802, 0.6126]],
  6641. device='cuda:0', grad_fn=<AddmmBackward>)
  6642. landmarks are: tensor([[[0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  6643. [0.0000, 0.0000, 0.7648, 0.2722, 0.3963, 0.2183, 0.5060, 0.5422],
  6644. [0.0000, 0.0000, 0.7613, 0.3250, 0.4038, 0.2533, 0.5437, 0.5767],
  6645. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  6646. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  6647. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  6648. [0.6262, 0.4052, 0.8888, 0.4700, 0.3675, 0.5117, 0.6350, 0.5233],
  6649. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209]]],
  6650. device='cuda:0')
  6651. loss_train_step before backward: tensor(0.0086, device='cuda:0', grad_fn=<MseLossBackward>)
  6652. loss_train_step after backward: tensor(0.0086, device='cuda:0', grad_fn=<MseLossBackward>)
  6653. loss_train: 0.031523438286967576
  6654. step: 14
  6655. running loss: 0.002251674163354827
  6656. Train Steps: 14/90 Loss: 0.0023 torch.Size([8, 600, 800])
  6657. torch.Size([8, 8])
  6658. tensor([[0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  6659. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  6660. [0.6192, 0.4128, 0.8513, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  6661. [0.6200, 0.4098, 0.8237, 0.2917, 0.4012, 0.2967, 0.6000, 0.5683],
  6662. [ nan, nan, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  6663. [0.6124, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  6664. [0.6147, 0.4112, 0.7988, 0.3200, 0.3775, 0.2767, 0.5150, 0.5550],
  6665. [0.6190, 0.4135, 0.8000, 0.4883, 0.3566, 0.3647, 0.5613, 0.5900]],
  6666. device='cuda:0', dtype=torch.float64)
  6667. predictions are: tensor([[0.6369, 0.3770, 0.8405, 0.3821, 0.3651, 0.3864, 0.5791, 0.5156],
  6668. [0.6178, 0.3775, 0.9148, 0.4095, 0.3825, 0.3501, 0.5726, 0.5146],
  6669. [0.7046, 0.4203, 0.8782, 0.6498, 0.4341, 0.5928, 0.5846, 0.5494],
  6670. [0.6253, 0.4038, 0.8356, 0.3593, 0.4174, 0.3532, 0.6002, 0.5456],
  6671. [0.2293, 0.1415, 0.8490, 0.2963, 0.5242, 0.2863, 0.6327, 0.5384],
  6672. [0.6541, 0.4069, 0.7483, 0.3571, 0.3859, 0.3337, 0.5057, 0.5481],
  6673. [0.5698, 0.3742, 0.8092, 0.4008, 0.3791, 0.3166, 0.4940, 0.5553],
  6674. [0.6750, 0.4287, 0.8318, 0.5494, 0.3564, 0.4079, 0.5192, 0.5380]],
  6675. device='cuda:0', grad_fn=<AddmmBackward>)
  6676. landmarks are: tensor([[[0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  6677. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  6678. [0.6192, 0.4128, 0.8512, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  6679. [0.6200, 0.4098, 0.8238, 0.2917, 0.4013, 0.2967, 0.6000, 0.5683],
  6680. [0.0000, 0.0000, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  6681. [0.6123, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  6682. [0.6147, 0.4112, 0.7987, 0.3200, 0.3775, 0.2767, 0.5150, 0.5550],
  6683. [0.6190, 0.4135, 0.8000, 0.4883, 0.3566, 0.3647, 0.5612, 0.5900]]],
  6684. device='cuda:0')
  6685. loss_train_step before backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  6686. loss_train_step after backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  6687. loss_train: 0.03430542966816574
  6688. step: 15
  6689. running loss: 0.0022870286445443827
  6690. Train Steps: 15/90 Loss: 0.0023 torch.Size([8, 600, 800])
  6691. torch.Size([8, 8])
  6692. tensor([[0.6264, 0.4049, 0.8988, 0.4633, 0.3813, 0.4983, 0.6326, 0.4843],
  6693. [0.6079, 0.3964, 0.7420, 0.2958, 0.3563, 0.2917, 0.5351, 0.4980],
  6694. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  6695. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  6696. [0.6140, 0.4070, 0.8700, 0.5000, 0.4612, 0.4900, 0.5260, 0.5852],
  6697. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  6698. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  6699. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290]],
  6700. device='cuda:0', dtype=torch.float64)
  6701. predictions are: tensor([[0.5944, 0.3454, 0.9136, 0.5033, 0.3575, 0.5314, 0.6047, 0.5082],
  6702. [0.5365, 0.3196, 0.7885, 0.3145, 0.3401, 0.3181, 0.5186, 0.5181],
  6703. [0.6247, 0.3771, 0.9237, 0.5689, 0.3381, 0.4124, 0.5901, 0.5049],
  6704. [0.5429, 0.3474, 0.8863, 0.5152, 0.4129, 0.5361, 0.5636, 0.5641],
  6705. [0.5060, 0.3406, 0.8971, 0.5473, 0.4520, 0.5284, 0.5180, 0.5835],
  6706. [0.5774, 0.3656, 0.8471, 0.3874, 0.3530, 0.4066, 0.5703, 0.5504],
  6707. [0.5550, 0.3419, 0.7782, 0.2975, 0.4284, 0.1834, 0.5793, 0.5391],
  6708. [0.4954, 0.2894, 0.7917, 0.2573, 0.4166, 0.2622, 0.6149, 0.5335]],
  6709. device='cuda:0', grad_fn=<AddmmBackward>)
  6710. landmarks are: tensor([[[0.6264, 0.4049, 0.8988, 0.4633, 0.3812, 0.4983, 0.6326, 0.4843],
  6711. [0.6079, 0.3964, 0.7420, 0.2958, 0.3562, 0.2917, 0.5351, 0.4980],
  6712. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  6713. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  6714. [0.6140, 0.4070, 0.8700, 0.5000, 0.4613, 0.4900, 0.5260, 0.5852],
  6715. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  6716. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  6717. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290]]],
  6718. device='cuda:0')
  6719. loss_train_step before backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  6720. loss_train_step after backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  6721. loss_train: 0.03636402601841837
  6722. step: 16
  6723. running loss: 0.0022727516261511482
  6724.  
  6725. Train Steps: 16/90 Loss: 0.0023 torch.Size([8, 600, 800])
  6726. torch.Size([8, 8])
  6727. tensor([[0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  6728. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  6729. [0.6199, 0.4093, 0.7913, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700],
  6730. [0.6276, 0.4095, 0.8237, 0.2250, 0.4662, 0.1783, 0.6171, 0.4869],
  6731. [0.6198, 0.4130, 0.8762, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  6732. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  6733. [0.6275, 0.4024, 0.8500, 0.5383, 0.3912, 0.4883, 0.6288, 0.5100],
  6734. [0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767]],
  6735. device='cuda:0', dtype=torch.float64)
  6736. predictions are: tensor([[0.4330, 0.2798, 0.8786, 0.2619, 0.4194, 0.2778, 0.6735, 0.5339],
  6737. [0.5555, 0.3388, 0.8658, 0.5031, 0.3932, 0.5309, 0.6476, 0.5368],
  6738. [0.4834, 0.3136, 0.8026, 0.2625, 0.3900, 0.2671, 0.5784, 0.5424],
  6739. [0.5196, 0.3330, 0.8274, 0.2463, 0.4491, 0.2297, 0.6057, 0.4950],
  6740. [0.5648, 0.3605, 0.8913, 0.4223, 0.3230, 0.4654, 0.5457, 0.5211],
  6741. [0.5265, 0.3438, 0.8746, 0.5002, 0.3539, 0.3281, 0.5949, 0.5066],
  6742. [0.5507, 0.3276, 0.8567, 0.5455, 0.3664, 0.5009, 0.5833, 0.4898],
  6743. [0.5677, 0.3474, 0.8651, 0.5104, 0.3851, 0.4639, 0.5004, 0.5649]],
  6744. device='cuda:0', grad_fn=<AddmmBackward>)
  6745. landmarks are: tensor([[[0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  6746. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  6747. [0.6198, 0.4093, 0.7912, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700],
  6748. [0.6276, 0.4095, 0.8238, 0.2250, 0.4663, 0.1783, 0.6171, 0.4869],
  6749. [0.6198, 0.4130, 0.8763, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  6750. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  6751. [0.6275, 0.4024, 0.8500, 0.5383, 0.3913, 0.4883, 0.6288, 0.5100],
  6752. [0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767]]],
  6753. device='cuda:0')
  6754. loss_train_step before backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  6755. loss_train_step after backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  6756. loss_train: 0.03901701339054853
  6757. step: 17
  6758. running loss: 0.0022951184347381488
  6759. Train Steps: 17/90 Loss: 0.0023 torch.Size([8, 600, 800])
  6760. torch.Size([8, 8])
  6761. tensor([[0.6311, 0.3998, 0.7975, 0.5767, 0.3838, 0.4850, 0.7327, 0.5343],
  6762. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  6763. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  6764. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  6765. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  6766. [0.6250, 0.4131, 0.8688, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  6767. [0.6246, 0.4028, 0.8738, 0.4867, 0.4088, 0.5667, 0.6362, 0.5200],
  6768. [ nan, nan, 0.7525, 0.2291, 0.3838, 0.3017, 0.6050, 0.5667]],
  6769. device='cuda:0', dtype=torch.float64)
  6770. predictions are: tensor([[0.5370, 0.3631, 0.8025, 0.4733, 0.3694, 0.4301, 0.6613, 0.5039],
  6771. [0.5555, 0.3662, 0.8576, 0.4481, 0.3509, 0.3076, 0.5604, 0.5309],
  6772. [0.5755, 0.3779, 0.8860, 0.4477, 0.3929, 0.5262, 0.6205, 0.5168],
  6773. [0.5609, 0.3712, 0.8877, 0.3128, 0.3627, 0.2935, 0.6108, 0.4896],
  6774. [0.4438, 0.3181, 0.7941, 0.2994, 0.3447, 0.2704, 0.5179, 0.5191],
  6775. [0.5578, 0.3616, 0.8624, 0.2708, 0.4258, 0.2182, 0.6350, 0.5023],
  6776. [0.6400, 0.3798, 0.8733, 0.4429, 0.4123, 0.5609, 0.6441, 0.5230],
  6777. [0.2633, 0.2200, 0.7622, 0.2249, 0.3862, 0.2525, 0.5937, 0.5282]],
  6778. device='cuda:0', grad_fn=<AddmmBackward>)
  6779. landmarks are: tensor([[[0.6311, 0.3998, 0.7975, 0.5767, 0.3837, 0.4850, 0.7327, 0.5343],
  6780. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  6781. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  6782. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  6783. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  6784. [0.6250, 0.4131, 0.8687, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  6785. [0.6246, 0.4028, 0.8737, 0.4867, 0.4087, 0.5667, 0.6363, 0.5200],
  6786. [0.0000, 0.0000, 0.7525, 0.2291, 0.3837, 0.3017, 0.6050, 0.5667]]],
  6787. device='cuda:0')
  6788. loss_train_step before backward: tensor(0.0036, device='cuda:0', grad_fn=<MseLossBackward>)
  6789. loss_train_step after backward: tensor(0.0036, device='cuda:0', grad_fn=<MseLossBackward>)
  6790. loss_train: 0.04258008219767362
  6791. step: 18
  6792. running loss: 0.002365560122092979
  6793. Train Steps: 18/90 Loss: 0.0024 torch.Size([8, 600, 800])
  6794. torch.Size([8, 8])
  6795. tensor([[0.6131, 0.4064, 0.8638, 0.5200, 0.4788, 0.4783, 0.5258, 0.5867],
  6796. [0.6197, 0.3986, 0.8800, 0.4617, 0.4188, 0.4783, 0.5687, 0.5550],
  6797. [ nan, nan, 0.8938, 0.2850, 0.4662, 0.3117, 0.7406, 0.5528],
  6798. [0.6201, 0.4151, 0.8588, 0.5467, 0.3700, 0.3950, 0.5637, 0.5933],
  6799. [0.6229, 0.4198, 0.7662, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783],
  6800. [0.6286, 0.3977, 0.9038, 0.4733, 0.3900, 0.4150, 0.7074, 0.5320],
  6801. [0.6296, 0.4045, 0.9138, 0.4100, 0.4232, 0.4242, 0.7422, 0.5297],
  6802. [0.6143, 0.4034, 0.8800, 0.4833, 0.4512, 0.5367, 0.5289, 0.5097]],
  6803. device='cuda:0', dtype=torch.float64)
  6804. predictions are: tensor([[0.5918, 0.3952, 0.8584, 0.4755, 0.4334, 0.4507, 0.5884, 0.5322],
  6805. [0.5844, 0.3977, 0.8591, 0.4129, 0.3728, 0.4645, 0.6018, 0.5118],
  6806. [0.2210, 0.1637, 0.9241, 0.2472, 0.4548, 0.2745, 0.7776, 0.5157],
  6807. [0.6087, 0.4235, 0.8215, 0.5093, 0.3536, 0.3738, 0.5898, 0.5271],
  6808. [0.5043, 0.3472, 0.7739, 0.2174, 0.4279, 0.2316, 0.6006, 0.5307],
  6809. [0.5409, 0.3545, 0.8611, 0.3833, 0.3235, 0.4136, 0.6816, 0.4781],
  6810. [0.6830, 0.4531, 0.8903, 0.3539, 0.3679, 0.4203, 0.7136, 0.4970],
  6811. [0.5697, 0.3720, 0.8571, 0.4214, 0.4092, 0.4769, 0.5778, 0.4975]],
  6812. device='cuda:0', grad_fn=<AddmmBackward>)
  6813. landmarks are: tensor([[[0.6132, 0.4063, 0.8637, 0.5200, 0.4787, 0.4783, 0.5258, 0.5867],
  6814. [0.6197, 0.3986, 0.8800, 0.4617, 0.4187, 0.4783, 0.5688, 0.5550],
  6815. [0.0000, 0.0000, 0.8938, 0.2850, 0.4663, 0.3117, 0.7406, 0.5528],
  6816. [0.6202, 0.4151, 0.8587, 0.5467, 0.3700, 0.3950, 0.5638, 0.5933],
  6817. [0.6229, 0.4198, 0.7663, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783],
  6818. [0.6286, 0.3977, 0.9038, 0.4733, 0.3900, 0.4150, 0.7074, 0.5320],
  6819. [0.6296, 0.4045, 0.9137, 0.4100, 0.4232, 0.4242, 0.7422, 0.5297],
  6820. [0.6143, 0.4034, 0.8800, 0.4833, 0.4512, 0.5367, 0.5289, 0.5097]]],
  6821. device='cuda:0')
  6822. loss_train_step before backward: tensor(0.0031, device='cuda:0', grad_fn=<MseLossBackward>)
  6823. loss_train_step after backward: tensor(0.0031, device='cuda:0', grad_fn=<MseLossBackward>)
  6824. loss_train: 0.045656067435629666
  6825. step: 19
  6826. running loss: 0.0024029509176647194
  6827. Train Steps: 19/90 Loss: 0.0024 torch.Size([8, 600, 800])
  6828. torch.Size([8, 8])
  6829. tensor([[ nan, nan, 0.8850, 0.2817, 0.5112, 0.2183, 0.7184, 0.5436],
  6830. [0.6277, 0.4083, 0.8350, 0.2717, 0.4562, 0.1800, 0.5918, 0.4878],
  6831. [0.6300, 0.4133, 0.8538, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  6832. [ nan, nan, 0.7648, 0.2722, 0.3962, 0.2183, 0.5060, 0.5422],
  6833. [0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921],
  6834. [0.6136, 0.4117, 0.8700, 0.5167, 0.4188, 0.5083, 0.5147, 0.5495],
  6835. [0.6134, 0.4090, 0.6926, 0.2819, 0.3538, 0.3233, 0.5563, 0.5667],
  6836. [0.6260, 0.4120, 0.8013, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895]],
  6837. device='cuda:0', dtype=torch.float64)
  6838. predictions are: tensor([[0.3317, 0.2437, 0.8838, 0.2707, 0.5007, 0.2329, 0.7101, 0.5134],
  6839. [0.4226, 0.2902, 0.8046, 0.2608, 0.4476, 0.2214, 0.6137, 0.5088],
  6840. [0.5430, 0.3784, 0.8516, 0.2332, 0.5131, 0.2774, 0.7556, 0.5394],
  6841. [0.1304, 0.1043, 0.7471, 0.2489, 0.3800, 0.2012, 0.5221, 0.5210],
  6842. [0.6030, 0.3979, 0.8176, 0.2325, 0.4119, 0.2649, 0.6792, 0.5080],
  6843. [0.7374, 0.4957, 0.8782, 0.5724, 0.4039, 0.4957, 0.5789, 0.5219],
  6844. [0.6925, 0.4588, 0.7116, 0.2866, 0.3439, 0.3203, 0.5630, 0.5454],
  6845. [0.6044, 0.3941, 0.7845, 0.2106, 0.4687, 0.1488, 0.6680, 0.5050]],
  6846. device='cuda:0', grad_fn=<AddmmBackward>)
  6847. landmarks are: tensor([[[0.0000, 0.0000, 0.8850, 0.2817, 0.5113, 0.2183, 0.7184, 0.5436],
  6848. [0.6277, 0.4083, 0.8350, 0.2717, 0.4563, 0.1800, 0.5918, 0.4878],
  6849. [0.6300, 0.4133, 0.8537, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  6850. [0.0000, 0.0000, 0.7648, 0.2722, 0.3963, 0.2183, 0.5060, 0.5422],
  6851. [0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921],
  6852. [0.6136, 0.4117, 0.8700, 0.5167, 0.4187, 0.5083, 0.5147, 0.5495],
  6853. [0.6134, 0.4090, 0.6926, 0.2819, 0.3537, 0.3233, 0.5562, 0.5667],
  6854. [0.6259, 0.4120, 0.8012, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895]]],
  6855. device='cuda:0')
  6856. loss_train_step before backward: tensor(0.0051, device='cuda:0', grad_fn=<MseLossBackward>)
  6857. loss_train_step after backward: tensor(0.0051, device='cuda:0', grad_fn=<MseLossBackward>)
  6858. loss_train: 0.050726136774756014
  6859. step: 20
  6860. running loss: 0.002536306838737801
  6861.  
  6862. Train Steps: 20/90 Loss: 0.0025 torch.Size([8, 600, 800])
  6863. torch.Size([8, 8])
  6864. tensor([[0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6138, 0.5333],
  6865. [0.6207, 0.4081, 0.7662, 0.2067, 0.3962, 0.3200, 0.6312, 0.5300],
  6866. [0.6115, 0.4081, 0.6725, 0.2433, 0.4088, 0.1933, 0.5167, 0.5544],
  6867. [0.6128, 0.4084, 0.8738, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  6868. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  6869. [0.6201, 0.4151, 0.8588, 0.5467, 0.3700, 0.3950, 0.5637, 0.5933],
  6870. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  6871. [0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131]],
  6872. device='cuda:0', dtype=torch.float64)
  6873. predictions are: tensor([[0.5901, 0.4298, 0.8875, 0.4076, 0.3987, 0.4621, 0.6725, 0.5381],
  6874. [0.5370, 0.3682, 0.7592, 0.1734, 0.4341, 0.2684, 0.6666, 0.5472],
  6875. [0.4572, 0.3172, 0.7158, 0.2226, 0.4405, 0.1772, 0.5711, 0.5436],
  6876. [0.5756, 0.3987, 0.8808, 0.4527, 0.3977, 0.3508, 0.5557, 0.5388],
  6877. [0.5095, 0.3654, 0.8806, 0.4264, 0.4335, 0.4443, 0.6113, 0.5361],
  6878. [0.5600, 0.4047, 0.8448, 0.5110, 0.4104, 0.3563, 0.6075, 0.5725],
  6879. [0.4382, 0.3084, 0.8828, 0.2070, 0.4856, 0.2378, 0.7775, 0.5535],
  6880. [0.6409, 0.4133, 0.8676, 0.3542, 0.4018, 0.4103, 0.6507, 0.5465]],
  6881. device='cuda:0', grad_fn=<AddmmBackward>)
  6882. landmarks are: tensor([[[0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6137, 0.5333],
  6883. [0.6207, 0.4081, 0.7663, 0.2067, 0.3963, 0.3200, 0.6313, 0.5300],
  6884. [0.6115, 0.4081, 0.6725, 0.2433, 0.4087, 0.1933, 0.5167, 0.5544],
  6885. [0.6127, 0.4084, 0.8737, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  6886. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  6887. [0.6202, 0.4151, 0.8587, 0.5467, 0.3700, 0.3950, 0.5638, 0.5933],
  6888. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  6889. [0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131]]],
  6890. device='cuda:0')
  6891. loss_train_step before backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  6892. loss_train_step after backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  6893. loss_train: 0.053505434771068394
  6894. step: 21
  6895. running loss: 0.002547877846241352
  6896. Train Steps: 21/90 Loss: 0.0025 torch.Size([8, 600, 800])
  6897. torch.Size([8, 8])
  6898. tensor([[0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  6899. [0.6268, 0.4061, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  6900. [0.6069, 0.3975, 0.8625, 0.5083, 0.4388, 0.5483, 0.5650, 0.4967],
  6901. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  6902. [0.6204, 0.4013, 0.8075, 0.2400, 0.4313, 0.2050, 0.5800, 0.5150],
  6903. [0.6346, 0.4144, 0.9088, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  6904. [0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332],
  6905. [0.6257, 0.4024, 0.8672, 0.5422, 0.4196, 0.5198, 0.6694, 0.5204]],
  6906. device='cuda:0', dtype=torch.float64)
  6907. predictions are: tensor([[0.5056, 0.3462, 0.8709, 0.2943, 0.4799, 0.2926, 0.6936, 0.5203],
  6908. [0.5236, 0.3603, 0.8550, 0.1726, 0.4822, 0.1785, 0.6520, 0.5347],
  6909. [0.5216, 0.3837, 0.8388, 0.4630, 0.4735, 0.4920, 0.5264, 0.5561],
  6910. [0.5107, 0.3523, 0.8397, 0.4314, 0.4522, 0.4588, 0.5876, 0.5638],
  6911. [0.4940, 0.3262, 0.7804, 0.2140, 0.4554, 0.2161, 0.5925, 0.5326],
  6912. [0.5580, 0.3879, 0.8799, 0.4109, 0.4304, 0.3769, 0.6884, 0.5656],
  6913. [0.6593, 0.4375, 0.8602, 0.4501, 0.4064, 0.3761, 0.6315, 0.5474],
  6914. [0.5514, 0.3732, 0.8135, 0.4741, 0.4225, 0.4865, 0.6201, 0.5624]],
  6915. device='cuda:0', grad_fn=<AddmmBackward>)
  6916. landmarks are: tensor([[[0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  6917. [0.6268, 0.4060, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  6918. [0.6069, 0.3975, 0.8625, 0.5083, 0.4387, 0.5483, 0.5650, 0.4967],
  6919. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  6920. [0.6204, 0.4013, 0.8075, 0.2400, 0.4313, 0.2050, 0.5800, 0.5150],
  6921. [0.6346, 0.4144, 0.9087, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  6922. [0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332],
  6923. [0.6257, 0.4024, 0.8672, 0.5422, 0.4196, 0.5198, 0.6694, 0.5204]]],
  6924. device='cuda:0')
  6925. loss_train_step before backward: tensor(0.0025, device='cuda:0', grad_fn=<MseLossBackward>)
  6926. loss_train_step after backward: tensor(0.0025, device='cuda:0', grad_fn=<MseLossBackward>)
  6927. loss_train: 0.05599151493515819
  6928. step: 22
  6929. running loss: 0.002545068860689009
  6930. Train Steps: 22/90 Loss: 0.0025 torch.Size([8, 600, 800])
  6931. torch.Size([8, 8])
  6932. tensor([[0.6200, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  6933. [ nan, nan, 0.6688, 0.2513, 0.4113, 0.2117, 0.5193, 0.5933],
  6934. [0.6124, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  6935. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5413, 0.5717],
  6936. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  6937. [0.6122, 0.4006, 0.8850, 0.4217, 0.4088, 0.5517, 0.6063, 0.5517],
  6938. [0.6196, 0.4088, 0.8888, 0.4583, 0.4500, 0.5683, 0.6138, 0.5883],
  6939. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395]],
  6940. device='cuda:0', dtype=torch.float64)
  6941. predictions are: tensor([[0.6189, 0.3997, 0.8820, 0.4238, 0.4160, 0.3840, 0.6119, 0.5490],
  6942. [0.0977, 0.0911, 0.6965, 0.2213, 0.4232, 0.1823, 0.5793, 0.5771],
  6943. [0.6315, 0.4235, 0.7317, 0.2649, 0.4093, 0.2834, 0.5647, 0.5624],
  6944. [0.6621, 0.4304, 0.8660, 0.4649, 0.4866, 0.4302, 0.5786, 0.5840],
  6945. [0.6096, 0.3989, 0.8899, 0.3080, 0.4147, 0.2769, 0.6552, 0.5774],
  6946. [0.6882, 0.4586, 0.8791, 0.4039, 0.4599, 0.5236, 0.6363, 0.5646],
  6947. [0.6097, 0.4281, 0.8857, 0.4110, 0.4833, 0.5131, 0.6350, 0.5820],
  6948. [0.6308, 0.4299, 0.8597, 0.4899, 0.3927, 0.3489, 0.5909, 0.5397]],
  6949. device='cuda:0', grad_fn=<AddmmBackward>)
  6950. landmarks are: tensor([[[0.6201, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  6951. [0.0000, 0.0000, 0.6688, 0.2513, 0.4112, 0.2117, 0.5193, 0.5933],
  6952. [0.6123, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  6953. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5412, 0.5717],
  6954. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  6955. [0.6122, 0.4006, 0.8850, 0.4217, 0.4087, 0.5517, 0.6062, 0.5517],
  6956. [0.6196, 0.4088, 0.8888, 0.4583, 0.4500, 0.5683, 0.6137, 0.5883],
  6957. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395]]],
  6958. device='cuda:0')
  6959. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  6960. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  6961. loss_train: 0.05741637130267918
  6962. step: 23
  6963. running loss: 0.0024963639696817036
  6964. Train Steps: 23/90 Loss: 0.0025 torch.Size([8, 600, 800])
  6965. torch.Size([8, 8])
  6966. tensor([[0.6212, 0.4033, 0.8938, 0.4167, 0.3813, 0.4267, 0.5613, 0.5583],
  6967. [0.6170, 0.4102, 0.7468, 0.3695, 0.3463, 0.3767, 0.5238, 0.5823],
  6968. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  6969. [ nan, nan, 0.6469, 0.1943, 0.4025, 0.2000, 0.5125, 0.5533],
  6970. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323],
  6971. [0.6102, 0.4005, 0.8688, 0.5100, 0.4813, 0.5400, 0.5404, 0.5064],
  6972. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309],
  6973. [0.6299, 0.4008, 0.8450, 0.5350, 0.4213, 0.5000, 0.6350, 0.5100]],
  6974. device='cuda:0', dtype=torch.float64)
  6975. predictions are: tensor([[0.6483, 0.4329, 0.8978, 0.4326, 0.3863, 0.4186, 0.5454, 0.5792],
  6976. [0.6430, 0.4231, 0.7608, 0.3607, 0.3721, 0.4013, 0.5512, 0.5972],
  6977. [0.6352, 0.3953, 0.7954, 0.2319, 0.4659, 0.2090, 0.6230, 0.5244],
  6978. [0.0224, 0.0161, 0.6969, 0.2110, 0.4282, 0.1571, 0.5344, 0.5740],
  6979. [0.6293, 0.4141, 0.7977, 0.2335, 0.4426, 0.1968, 0.6254, 0.5558],
  6980. [0.6616, 0.4534, 0.8593, 0.5196, 0.4908, 0.4956, 0.5336, 0.5511],
  6981. [0.6865, 0.4557, 0.8721, 0.4978, 0.3912, 0.3721, 0.6095, 0.5410],
  6982. [0.7057, 0.4689, 0.8426, 0.5325, 0.4386, 0.5152, 0.6351, 0.5462]],
  6983. device='cuda:0', grad_fn=<AddmmBackward>)
  6984. landmarks are: tensor([[[0.6212, 0.4033, 0.8938, 0.4167, 0.3812, 0.4267, 0.5612, 0.5583],
  6985. [0.6170, 0.4102, 0.7468, 0.3695, 0.3462, 0.3767, 0.5238, 0.5823],
  6986. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  6987. [0.0000, 0.0000, 0.6469, 0.1943, 0.4025, 0.2000, 0.5125, 0.5533],
  6988. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323],
  6989. [0.6102, 0.4005, 0.8687, 0.5100, 0.4812, 0.5400, 0.5404, 0.5064],
  6990. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309],
  6991. [0.6299, 0.4008, 0.8450, 0.5350, 0.4212, 0.5000, 0.6350, 0.5100]]],
  6992. device='cuda:0')
  6993. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  6994. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  6995. loss_train: 0.058127752621658146
  6996. step: 24
  6997. running loss: 0.0024219896925690896
  6998.  
  6999. Train Steps: 24/90 Loss: 0.0024 torch.Size([8, 600, 800])
  7000. torch.Size([8, 8])
  7001. tensor([[0.6182, 0.4058, 0.8738, 0.4350, 0.3563, 0.3400, 0.5290, 0.5822],
  7002. [0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050],
  7003. [ nan, nan, 0.7512, 0.2117, 0.4288, 0.2000, 0.5600, 0.5617],
  7004. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  7005. [0.6308, 0.3990, 0.8688, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  7006. [0.6128, 0.4022, 0.8738, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  7007. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  7008. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450]],
  7009. device='cuda:0', dtype=torch.float64)
  7010. predictions are: tensor([[0.6093, 0.3999, 0.8337, 0.4284, 0.3559, 0.3779, 0.5022, 0.5830],
  7011. [0.6382, 0.4220, 0.8517, 0.4154, 0.3702, 0.5035, 0.5867, 0.5758],
  7012. [0.2702, 0.1768, 0.7247, 0.2521, 0.4284, 0.2276, 0.5442, 0.5621],
  7013. [0.6540, 0.4113, 0.8690, 0.3864, 0.4371, 0.2669, 0.6174, 0.5398],
  7014. [0.6239, 0.3889, 0.8392, 0.5560, 0.4058, 0.5212, 0.6302, 0.5345],
  7015. [0.6113, 0.4132, 0.8346, 0.5003, 0.4610, 0.4800, 0.5369, 0.5320],
  7016. [0.7170, 0.4505, 0.8821, 0.4155, 0.3609, 0.4126, 0.6495, 0.5369],
  7017. [0.6013, 0.3870, 0.8461, 0.4729, 0.3706, 0.4457, 0.5826, 0.5565]],
  7018. device='cuda:0', grad_fn=<AddmmBackward>)
  7019. landmarks are: tensor([[[0.6182, 0.4058, 0.8737, 0.4350, 0.3562, 0.3400, 0.5290, 0.5822],
  7020. [0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050],
  7021. [0.0000, 0.0000, 0.7513, 0.2117, 0.4288, 0.2000, 0.5600, 0.5617],
  7022. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  7023. [0.6308, 0.3990, 0.8687, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  7024. [0.6128, 0.4022, 0.8737, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  7025. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  7026. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450]]],
  7027. device='cuda:0')
  7028. loss_train_step before backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  7029. loss_train_step after backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  7030. loss_train: 0.06033143226522952
  7031. step: 25
  7032. running loss: 0.002413257290609181
  7033. Train Steps: 25/90 Loss: 0.0024 torch.Size([8, 600, 800])
  7034. torch.Size([8, 8])
  7035. tensor([[0.6196, 0.4094, 0.7562, 0.2817, 0.3937, 0.3183, 0.6013, 0.6183],
  7036. [0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717],
  7037. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  7038. [0.6148, 0.4076, 0.8666, 0.4820, 0.4138, 0.5067, 0.5250, 0.5767],
  7039. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  7040. [0.6127, 0.4115, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  7041. [0.6333, 0.4037, 0.8638, 0.5733, 0.4012, 0.4717, 0.6369, 0.4938],
  7042. [0.6159, 0.4085, 0.6900, 0.2283, 0.4088, 0.1950, 0.5123, 0.5397]],
  7043. device='cuda:0', dtype=torch.float64)
  7044. predictions are: tensor([[0.6160, 0.3895, 0.7928, 0.3331, 0.3901, 0.3177, 0.5873, 0.5922],
  7045. [0.5629, 0.3442, 0.7577, 0.2975, 0.3681, 0.2849, 0.5655, 0.5591],
  7046. [0.6375, 0.3715, 0.9214, 0.4802, 0.3835, 0.5911, 0.6241, 0.5429],
  7047. [0.6171, 0.3862, 0.8805, 0.5191, 0.4360, 0.5156, 0.5317, 0.5459],
  7048. [0.6700, 0.4098, 0.8710, 0.5518, 0.3930, 0.5352, 0.5776, 0.5149],
  7049. [0.5614, 0.3484, 0.7337, 0.3151, 0.3581, 0.3145, 0.5202, 0.5507],
  7050. [0.7045, 0.4279, 0.8837, 0.5943, 0.3764, 0.4633, 0.6147, 0.5073],
  7051. [0.4651, 0.2801, 0.6989, 0.2835, 0.3888, 0.2057, 0.5165, 0.5298]],
  7052. device='cuda:0', grad_fn=<AddmmBackward>)
  7053. landmarks are: tensor([[[0.6196, 0.4094, 0.7563, 0.2817, 0.3938, 0.3183, 0.6012, 0.6183],
  7054. [0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717],
  7055. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  7056. [0.6148, 0.4076, 0.8666, 0.4820, 0.4137, 0.5067, 0.5250, 0.5767],
  7057. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  7058. [0.6127, 0.4114, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  7059. [0.6334, 0.4037, 0.8637, 0.5733, 0.4013, 0.4717, 0.6369, 0.4938],
  7060. [0.6159, 0.4085, 0.6900, 0.2283, 0.4087, 0.1950, 0.5123, 0.5397]]],
  7061. device='cuda:0')
  7062. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  7063. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  7064. loss_train: 0.06175887619610876
  7065. step: 26
  7066. running loss: 0.002375341392158029
  7067. Train Steps: 26/90 Loss: 0.0024 torch.Size([8, 600, 800])
  7068. torch.Size([8, 8])
  7069. tensor([[0.6141, 0.4038, 0.8650, 0.4833, 0.4839, 0.5176, 0.5787, 0.5600],
  7070. [0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  7071. [ nan, nan, 0.7981, 0.3194, 0.3625, 0.3167, 0.5040, 0.5563],
  7072. [0.6304, 0.4024, 0.8925, 0.4800, 0.3937, 0.4817, 0.7485, 0.5297],
  7073. [0.6034, 0.4011, 0.7350, 0.2533, 0.3438, 0.3367, 0.5516, 0.5084],
  7074. [0.6185, 0.4080, 0.8625, 0.3483, 0.3788, 0.2650, 0.5320, 0.5272],
  7075. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  7076. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383]],
  7077. device='cuda:0', dtype=torch.float64)
  7078. predictions are: tensor([[0.6436, 0.3993, 0.8663, 0.5342, 0.4515, 0.5277, 0.5575, 0.5408],
  7079. [0.6770, 0.3979, 0.8363, 0.4462, 0.3444, 0.5367, 0.5881, 0.5528],
  7080. [0.2007, 0.1157, 0.7863, 0.3516, 0.3207, 0.3140, 0.5153, 0.5435],
  7081. [0.6610, 0.3867, 0.8835, 0.5234, 0.3765, 0.5157, 0.6412, 0.5174],
  7082. [0.6181, 0.3727, 0.7298, 0.2919, 0.3445, 0.3732, 0.5756, 0.5017],
  7083. [0.6510, 0.3927, 0.8464, 0.4056, 0.3709, 0.3140, 0.5137, 0.5286],
  7084. [0.6249, 0.3743, 0.8593, 0.4903, 0.3432, 0.3430, 0.5064, 0.5652],
  7085. [0.6487, 0.4039, 0.7184, 0.2811, 0.4340, 0.1724, 0.5875, 0.5331]],
  7086. device='cuda:0', grad_fn=<AddmmBackward>)
  7087. landmarks are: tensor([[[0.6141, 0.4038, 0.8650, 0.4833, 0.4839, 0.5176, 0.5788, 0.5600],
  7088. [0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  7089. [0.0000, 0.0000, 0.7981, 0.3194, 0.3625, 0.3167, 0.5040, 0.5563],
  7090. [0.6304, 0.4024, 0.8925, 0.4800, 0.3938, 0.4817, 0.7485, 0.5297],
  7091. [0.6033, 0.4011, 0.7350, 0.2533, 0.3438, 0.3367, 0.5516, 0.5084],
  7092. [0.6186, 0.4080, 0.8625, 0.3483, 0.3787, 0.2650, 0.5320, 0.5272],
  7093. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  7094. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383]]],
  7095. device='cuda:0')
  7096. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  7097. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  7098. loss_train: 0.06347480788826942
  7099. step: 27
  7100. running loss: 0.002350918810676645
  7101. Train Steps: 27/90 Loss: 0.0024 torch.Size([8, 600, 800])
  7102. torch.Size([8, 8])
  7103. tensor([[0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  7104. [0.6185, 0.4079, 0.8838, 0.4617, 0.4838, 0.5650, 0.6175, 0.5850],
  7105. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  7106. [0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  7107. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  7108. [0.6085, 0.4008, 0.8588, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  7109. [0.6117, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103],
  7110. [0.6201, 0.4151, 0.8588, 0.5467, 0.3700, 0.3950, 0.5637, 0.5933]],
  7111. device='cuda:0', dtype=torch.float64)
  7112. predictions are: tensor([[0.6822, 0.4314, 0.8651, 0.4856, 0.3992, 0.4735, 0.5274, 0.5404],
  7113. [0.5400, 0.3283, 0.8774, 0.4733, 0.4195, 0.5492, 0.6072, 0.5681],
  7114. [0.4843, 0.2985, 0.8924, 0.5053, 0.3795, 0.5752, 0.6132, 0.5527],
  7115. [0.6219, 0.3712, 0.8523, 0.3708, 0.3658, 0.2955, 0.6313, 0.5004],
  7116. [0.6180, 0.4010, 0.8046, 0.3480, 0.3300, 0.3867, 0.5747, 0.5360],
  7117. [0.6175, 0.3946, 0.8366, 0.5188, 0.4460, 0.4738, 0.5096, 0.5268],
  7118. [0.5767, 0.3839, 0.6640, 0.2650, 0.3513, 0.2447, 0.5245, 0.5126],
  7119. [0.6322, 0.4174, 0.8397, 0.5665, 0.3124, 0.4103, 0.5321, 0.5726]],
  7120. device='cuda:0', grad_fn=<AddmmBackward>)
  7121. landmarks are: tensor([[[0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  7122. [0.6184, 0.4079, 0.8838, 0.4617, 0.4837, 0.5650, 0.6175, 0.5850],
  7123. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  7124. [0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  7125. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  7126. [0.6084, 0.4008, 0.8587, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  7127. [0.6116, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103],
  7128. [0.6202, 0.4151, 0.8587, 0.5467, 0.3700, 0.3950, 0.5638, 0.5933]]],
  7129. device='cuda:0')
  7130. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  7131. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  7132. loss_train: 0.06485963379964232
  7133. step: 28
  7134. running loss: 0.0023164154928443687
  7135.  
  7136. Train Steps: 28/90 Loss: 0.0023 torch.Size([8, 600, 800])
  7137. torch.Size([8, 8])
  7138. tensor([[0.6200, 0.3993, 0.8519, 0.4923, 0.3962, 0.4717, 0.6013, 0.5433],
  7139. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  7140. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  7141. [ nan, nan, 0.7240, 0.2722, 0.3900, 0.2567, 0.5168, 0.5933],
  7142. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  7143. [0.6118, 0.4052, 0.8463, 0.3917, 0.3538, 0.3450, 0.5053, 0.5593],
  7144. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  7145. [0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947]],
  7146. device='cuda:0', dtype=torch.float64)
  7147. predictions are: tensor([[ 6.7090e-01, 4.0646e-01, 8.8374e-01, 5.4438e-01, 3.8827e-01,
  7148. 5.3380e-01, 5.8710e-01, 5.2107e-01],
  7149. [ 6.3156e-01, 3.8918e-01, 8.8924e-01, 3.9016e-01, 4.5429e-01,
  7150. 3.5994e-01, 7.0014e-01, 5.0003e-01],
  7151. [ 7.1235e-01, 4.4388e-01, 8.7769e-01, 4.9250e-01, 3.6733e-01,
  7152. 4.6729e-01, 5.2955e-01, 5.4011e-01],
  7153. [-3.6269e-04, -8.9101e-03, 7.3422e-01, 3.1684e-01, 3.8392e-01,
  7154. 2.9073e-01, 4.9513e-01, 5.4069e-01],
  7155. [ 7.0830e-01, 4.4306e-01, 8.0668e-01, 3.2193e-01, 3.6984e-01,
  7156. 3.3750e-01, 5.6709e-01, 5.1859e-01],
  7157. [ 6.9812e-01, 4.5714e-01, 8.6463e-01, 4.2777e-01, 3.4170e-01,
  7158. 3.8019e-01, 4.5121e-01, 5.2652e-01],
  7159. [ 6.6152e-01, 4.4307e-01, 7.3589e-01, 3.0837e-01, 4.2011e-01,
  7160. 2.6642e-01, 5.2988e-01, 5.5355e-01],
  7161. [ 5.7453e-01, 3.4652e-01, 7.8571e-01, 2.3260e-01, 4.2730e-01,
  7162. 2.6989e-01, 6.5486e-01, 5.0269e-01]], device='cuda:0',
  7163. grad_fn=<AddmmBackward>)
  7164. landmarks are: tensor([[[0.6200, 0.3993, 0.8519, 0.4923, 0.3963, 0.4717, 0.6012, 0.5433],
  7165. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  7166. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  7167. [0.0000, 0.0000, 0.7240, 0.2722, 0.3900, 0.2567, 0.5168, 0.5933],
  7168. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  7169. [0.6118, 0.4052, 0.8462, 0.3917, 0.3537, 0.3450, 0.5053, 0.5593],
  7170. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  7171. [0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947]]],
  7172. device='cuda:0')
  7173. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  7174. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  7175. loss_train: 0.066403744276613
  7176. step: 29
  7177. running loss: 0.002289784285400448
  7178. Train Steps: 29/90 Loss: 0.0023 torch.Size([8, 600, 800])
  7179. torch.Size([8, 8])
  7180. tensor([[0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  7181. [0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  7182. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  7183. [0.6317, 0.4038, 0.8287, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  7184. [0.6085, 0.4008, 0.8588, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  7185. [0.6214, 0.4112, 0.7838, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  7186. [ nan, nan, 0.7525, 0.2291, 0.3838, 0.3017, 0.6050, 0.5667],
  7187. [0.6163, 0.4114, 0.7650, 0.2017, 0.3763, 0.2867, 0.5631, 0.5071]],
  7188. device='cuda:0', dtype=torch.float64)
  7189. predictions are: tensor([[0.6553, 0.4516, 0.8909, 0.4744, 0.3987, 0.3983, 0.5979, 0.5874],
  7190. [0.6033, 0.3995, 0.9001, 0.3978, 0.3655, 0.4665, 0.5997, 0.5057],
  7191. [0.6109, 0.3870, 0.8913, 0.4282, 0.3730, 0.3778, 0.5874, 0.5565],
  7192. [0.6447, 0.4530, 0.8713, 0.6013, 0.3892, 0.4725, 0.6208, 0.5390],
  7193. [0.6460, 0.4327, 0.8558, 0.5071, 0.4999, 0.4846, 0.5085, 0.5388],
  7194. [0.6930, 0.4552, 0.7825, 0.2199, 0.3805, 0.2937, 0.5852, 0.5319],
  7195. [0.1483, 0.1368, 0.7527, 0.2525, 0.4189, 0.2928, 0.5746, 0.5629],
  7196. [0.6174, 0.3917, 0.7474, 0.2506, 0.4010, 0.2601, 0.5830, 0.5165]],
  7197. device='cuda:0', grad_fn=<AddmmBackward>)
  7198. landmarks are: tensor([[[0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  7199. [0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  7200. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  7201. [0.6317, 0.4038, 0.8288, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  7202. [0.6084, 0.4008, 0.8587, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  7203. [0.6214, 0.4112, 0.7837, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  7204. [0.0000, 0.0000, 0.7525, 0.2291, 0.3837, 0.3017, 0.6050, 0.5667],
  7205. [0.6163, 0.4114, 0.7650, 0.2017, 0.3762, 0.2867, 0.5631, 0.5071]]],
  7206. device='cuda:0')
  7207. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  7208. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  7209. loss_train: 0.06759074423462152
  7210. step: 30
  7211. running loss: 0.0022530248078207176
  7212. Train Steps: 30/90 Loss: 0.0023 torch.Size([8, 600, 800])
  7213. torch.Size([8, 8])
  7214. tensor([[0.6293, 0.4097, 0.8800, 0.2517, 0.5262, 0.2600, 0.7430, 0.5378],
  7215. [0.6265, 0.4251, 0.7113, 0.3550, 0.4375, 0.2117, 0.5587, 0.6118],
  7216. [0.6145, 0.4008, 0.8750, 0.5383, 0.3975, 0.4650, 0.5563, 0.5533],
  7217. [0.6275, 0.4157, 0.8337, 0.5800, 0.3763, 0.4200, 0.5547, 0.6125],
  7218. [0.6246, 0.4028, 0.8738, 0.4867, 0.4088, 0.5667, 0.6362, 0.5200],
  7219. [0.6125, 0.3999, 0.8750, 0.4883, 0.4750, 0.4700, 0.5533, 0.5617],
  7220. [0.6142, 0.3982, 0.8650, 0.4883, 0.3912, 0.4317, 0.5315, 0.5350],
  7221. [0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153]],
  7222. device='cuda:0', dtype=torch.float64)
  7223. predictions are: tensor([[0.5891, 0.4129, 0.8819, 0.1990, 0.5162, 0.2382, 0.7617, 0.5226],
  7224. [0.5691, 0.4271, 0.7440, 0.2872, 0.3934, 0.2544, 0.5267, 0.5962],
  7225. [0.5291, 0.3691, 0.8776, 0.4669, 0.3964, 0.4697, 0.5507, 0.5498],
  7226. [0.5685, 0.3836, 0.8484, 0.5101, 0.3768, 0.4332, 0.5894, 0.5859],
  7227. [0.5461, 0.3610, 0.8937, 0.4030, 0.4117, 0.5739, 0.6300, 0.5297],
  7228. [0.6028, 0.4109, 0.8958, 0.4430, 0.4658, 0.4558, 0.5568, 0.5362],
  7229. [0.5917, 0.3764, 0.8950, 0.4554, 0.3693, 0.4597, 0.5308, 0.4896],
  7230. [0.5577, 0.3902, 0.8747, 0.4571, 0.4388, 0.5051, 0.5908, 0.5223]],
  7231. device='cuda:0', grad_fn=<AddmmBackward>)
  7232. landmarks are: tensor([[[0.6293, 0.4097, 0.8800, 0.2517, 0.5263, 0.2600, 0.7430, 0.5378],
  7233. [0.6265, 0.4251, 0.7113, 0.3550, 0.4375, 0.2117, 0.5587, 0.6118],
  7234. [0.6145, 0.4008, 0.8750, 0.5383, 0.3975, 0.4650, 0.5562, 0.5533],
  7235. [0.6275, 0.4157, 0.8338, 0.5800, 0.3762, 0.4200, 0.5547, 0.6125],
  7236. [0.6246, 0.4028, 0.8737, 0.4867, 0.4087, 0.5667, 0.6363, 0.5200],
  7237. [0.6125, 0.3999, 0.8750, 0.4883, 0.4750, 0.4700, 0.5533, 0.5617],
  7238. [0.6143, 0.3982, 0.8650, 0.4883, 0.3913, 0.4317, 0.5315, 0.5350],
  7239. [0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153]]],
  7240. device='cuda:0')
  7241. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  7242. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  7243. loss_train: 0.06881934218108654
  7244. step: 31
  7245. running loss: 0.0022199787800350496
  7246. Train Steps: 31/90 Loss: 0.0022 torch.Size([8, 600, 800])
  7247. torch.Size([8, 8])
  7248. tensor([[0.6193, 0.4108, 0.7438, 0.2700, 0.3650, 0.3683, 0.6238, 0.5717],
  7249. [0.6111, 0.4019, 0.8350, 0.3867, 0.3500, 0.4283, 0.5480, 0.5719],
  7250. [0.6189, 0.4049, 0.8888, 0.4417, 0.4213, 0.5200, 0.5988, 0.5633],
  7251. [0.6205, 0.4016, 0.8350, 0.2717, 0.3987, 0.2550, 0.5787, 0.5133],
  7252. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  7253. [ nan, nan, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600],
  7254. [0.6196, 0.4094, 0.7562, 0.2817, 0.3937, 0.3183, 0.6013, 0.6183],
  7255. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383]],
  7256. device='cuda:0', dtype=torch.float64)
  7257. predictions are: tensor([[0.6275, 0.4265, 0.7716, 0.2781, 0.3984, 0.3497, 0.6382, 0.5523],
  7258. [0.5203, 0.3401, 0.8799, 0.4272, 0.3907, 0.4228, 0.5961, 0.5414],
  7259. [0.5463, 0.3887, 0.9214, 0.4432, 0.4414, 0.5566, 0.6011, 0.5751],
  7260. [0.6827, 0.4398, 0.8401, 0.2632, 0.4333, 0.2702, 0.5930, 0.5047],
  7261. [0.6210, 0.4231, 0.8930, 0.4080, 0.3893, 0.3765, 0.5890, 0.5142],
  7262. [0.1107, 0.0948, 0.7827, 0.2740, 0.4395, 0.2777, 0.5472, 0.5458],
  7263. [0.5985, 0.4182, 0.7871, 0.2891, 0.4246, 0.3038, 0.5919, 0.6043],
  7264. [0.6370, 0.4413, 0.7423, 0.2375, 0.4879, 0.1454, 0.5941, 0.5261]],
  7265. device='cuda:0', grad_fn=<AddmmBackward>)
  7266. landmarks are: tensor([[[0.6193, 0.4108, 0.7437, 0.2700, 0.3650, 0.3683, 0.6237, 0.5717],
  7267. [0.6111, 0.4019, 0.8350, 0.3867, 0.3500, 0.4283, 0.5480, 0.5719],
  7268. [0.6189, 0.4049, 0.8888, 0.4417, 0.4212, 0.5200, 0.5987, 0.5633],
  7269. [0.6205, 0.4015, 0.8350, 0.2717, 0.3988, 0.2550, 0.5788, 0.5133],
  7270. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  7271. [0.0000, 0.0000, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600],
  7272. [0.6196, 0.4094, 0.7563, 0.2817, 0.3938, 0.3183, 0.6012, 0.6183],
  7273. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383]]],
  7274. device='cuda:0')
  7275. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  7276. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  7277. loss_train: 0.06997097912244499
  7278. step: 32
  7279. running loss: 0.002186593097576406
  7280.  
  7281. Train Steps: 32/90 Loss: 0.0022 torch.Size([8, 600, 800])
  7282. torch.Size([8, 8])
  7283. tensor([[0.6239, 0.4123, 0.8313, 0.2550, 0.4500, 0.2050, 0.6175, 0.5400],
  7284. [0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  7285. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050],
  7286. [0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  7287. [0.6126, 0.3954, 0.8538, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  7288. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  7289. [0.6170, 0.4102, 0.7468, 0.3695, 0.3463, 0.3767, 0.5238, 0.5823],
  7290. [0.6097, 0.3988, 0.8650, 0.5250, 0.4213, 0.5200, 0.5675, 0.5050]],
  7291. device='cuda:0', dtype=torch.float64)
  7292. predictions are: tensor([[0.6632, 0.4778, 0.8182, 0.1933, 0.4617, 0.1867, 0.6462, 0.5282],
  7293. [0.5586, 0.3796, 0.8881, 0.3963, 0.3744, 0.3900, 0.5972, 0.5318],
  7294. [0.5068, 0.3652, 0.8657, 0.4687, 0.4695, 0.4946, 0.6065, 0.5649],
  7295. [0.5647, 0.3709, 0.8731, 0.3147, 0.4259, 0.2559, 0.6572, 0.5229],
  7296. [0.5499, 0.3703, 0.8740, 0.4694, 0.4646, 0.4603, 0.5615, 0.5666],
  7297. [0.5191, 0.3648, 0.8962, 0.3705, 0.3918, 0.3294, 0.5904, 0.5462],
  7298. [0.5249, 0.3715, 0.7566, 0.3213, 0.3711, 0.3640, 0.5613, 0.6067],
  7299. [0.5703, 0.3967, 0.8596, 0.4826, 0.4569, 0.4958, 0.5331, 0.5503]],
  7300. device='cuda:0', grad_fn=<AddmmBackward>)
  7301. landmarks are: tensor([[[0.6239, 0.4123, 0.8313, 0.2550, 0.4500, 0.2050, 0.6175, 0.5400],
  7302. [0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  7303. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050],
  7304. [0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  7305. [0.6126, 0.3954, 0.8537, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  7306. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  7307. [0.6170, 0.4102, 0.7468, 0.3695, 0.3462, 0.3767, 0.5238, 0.5823],
  7308. [0.6097, 0.3988, 0.8650, 0.5250, 0.4212, 0.5200, 0.5675, 0.5050]]],
  7309. device='cuda:0')
  7310. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  7311. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  7312. loss_train: 0.0716163640609011
  7313. step: 33
  7314. running loss: 0.0021701928503303366
  7315. Train Steps: 33/90 Loss: 0.0022 torch.Size([8, 600, 800])
  7316. torch.Size([8, 8])
  7317. tensor([[0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  7318. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  7319. [0.6200, 0.4049, 0.8638, 0.5617, 0.4125, 0.5100, 0.6013, 0.5317],
  7320. [0.6185, 0.4079, 0.8838, 0.4617, 0.4838, 0.5650, 0.6175, 0.5850],
  7321. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  7322. [0.6197, 0.4118, 0.8688, 0.5517, 0.4037, 0.5233, 0.5875, 0.5600],
  7323. [0.6261, 0.3987, 0.9045, 0.4208, 0.3600, 0.4633, 0.6570, 0.5162],
  7324. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378]],
  7325. device='cuda:0', dtype=torch.float64)
  7326. predictions are: tensor([[0.5498, 0.3801, 0.8887, 0.4040, 0.3866, 0.3922, 0.5649, 0.5455],
  7327. [0.5553, 0.3956, 0.8445, 0.2139, 0.4880, 0.1568, 0.6218, 0.5161],
  7328. [0.5328, 0.3725, 0.8469, 0.5384, 0.4104, 0.4539, 0.5756, 0.5725],
  7329. [0.6221, 0.4063, 0.8781, 0.4012, 0.4715, 0.5037, 0.6208, 0.5813],
  7330. [0.5698, 0.3907, 0.8111, 0.2811, 0.3746, 0.3330, 0.5808, 0.5588],
  7331. [0.5970, 0.4106, 0.8380, 0.5187, 0.4095, 0.5040, 0.5817, 0.5924],
  7332. [0.6124, 0.4144, 0.8904, 0.4044, 0.3616, 0.4242, 0.6757, 0.5401],
  7333. [0.5840, 0.3941, 0.8406, 0.4465, 0.4335, 0.4914, 0.6647, 0.5600]],
  7334. device='cuda:0', grad_fn=<AddmmBackward>)
  7335. landmarks are: tensor([[[0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  7336. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  7337. [0.6199, 0.4049, 0.8637, 0.5617, 0.4125, 0.5100, 0.6012, 0.5317],
  7338. [0.6184, 0.4079, 0.8838, 0.4617, 0.4837, 0.5650, 0.6175, 0.5850],
  7339. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  7340. [0.6197, 0.4118, 0.8687, 0.5517, 0.4038, 0.5233, 0.5875, 0.5600],
  7341. [0.6261, 0.3987, 0.9045, 0.4208, 0.3600, 0.4633, 0.6570, 0.5162],
  7342. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378]]],
  7343. device='cuda:0')
  7344. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  7345. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  7346. loss_train: 0.07260472874622792
  7347. step: 34
  7348. running loss: 0.0021354331984184682
  7349. Train Steps: 34/90 Loss: 0.0021 torch.Size([8, 600, 800])
  7350. torch.Size([8, 8])
  7351. tensor([[0.6245, 0.4100, 0.7762, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  7352. [ nan, nan, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  7353. [0.6263, 0.4039, 0.9000, 0.4400, 0.3625, 0.4667, 0.6424, 0.4804],
  7354. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  7355. [0.6126, 0.4039, 0.8237, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138],
  7356. [0.6212, 0.4171, 0.7875, 0.3633, 0.3813, 0.2933, 0.5675, 0.5700],
  7357. [0.6114, 0.4018, 0.7213, 0.1967, 0.3763, 0.2700, 0.5875, 0.5533],
  7358. [0.6179, 0.4008, 0.7505, 0.2678, 0.4368, 0.1891, 0.5831, 0.5263]],
  7359. device='cuda:0', dtype=torch.float64)
  7360. predictions are: tensor([[0.5184, 0.3429, 0.7628, 0.2571, 0.4802, 0.1860, 0.6074, 0.5515],
  7361. [0.0977, 0.0458, 0.8108, 0.3233, 0.3749, 0.3155, 0.5274, 0.5635],
  7362. [0.6940, 0.4294, 0.9553, 0.4917, 0.3541, 0.5029, 0.6596, 0.4610],
  7363. [0.5996, 0.3737, 0.8759, 0.3087, 0.5299, 0.2640, 0.7339, 0.5147],
  7364. [0.6866, 0.4262, 0.8580, 0.4324, 0.3571, 0.4155, 0.6056, 0.5862],
  7365. [0.7120, 0.4739, 0.8066, 0.3997, 0.3782, 0.3399, 0.5830, 0.5756],
  7366. [0.6466, 0.4199, 0.7163, 0.2580, 0.3951, 0.2921, 0.5666, 0.5387],
  7367. [0.7154, 0.4815, 0.7369, 0.2808, 0.4385, 0.2124, 0.5525, 0.5242]],
  7368. device='cuda:0', grad_fn=<AddmmBackward>)
  7369. landmarks are: tensor([[[0.6245, 0.4100, 0.7763, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  7370. [0.0000, 0.0000, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  7371. [0.6263, 0.4038, 0.9000, 0.4400, 0.3625, 0.4667, 0.6424, 0.4804],
  7372. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  7373. [0.6126, 0.4038, 0.8238, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138],
  7374. [0.6212, 0.4171, 0.7875, 0.3633, 0.3812, 0.2933, 0.5675, 0.5700],
  7375. [0.6114, 0.4018, 0.7212, 0.1967, 0.3762, 0.2700, 0.5875, 0.5533],
  7376. [0.6179, 0.4008, 0.7505, 0.2678, 0.4368, 0.1891, 0.5831, 0.5263]]],
  7377. device='cuda:0')
  7378. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  7379. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  7380. loss_train: 0.07426053285598755
  7381. step: 35
  7382. running loss: 0.0021217295101710726
  7383. Train Steps: 35/90 Loss: 0.0021 torch.Size([8, 600, 800])
  7384. torch.Size([8, 8])
  7385. tensor([[0.6274, 0.4003, 0.8638, 0.5967, 0.3688, 0.4900, 0.6108, 0.4661],
  7386. [0.6219, 0.3934, 0.8688, 0.5267, 0.4313, 0.4967, 0.5988, 0.4983],
  7387. [ nan, nan, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  7388. [0.6161, 0.4024, 0.8838, 0.4583, 0.3688, 0.3733, 0.5311, 0.5344],
  7389. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  7390. [0.6087, 0.3951, 0.8387, 0.5833, 0.4188, 0.4933, 0.5146, 0.4830],
  7391. [0.6255, 0.4017, 0.8688, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  7392. [0.6248, 0.4185, 0.8500, 0.5767, 0.4463, 0.4550, 0.5613, 0.5917]],
  7393. device='cuda:0', dtype=torch.float64)
  7394. predictions are: tensor([[0.6699, 0.4275, 0.8024, 0.5265, 0.3900, 0.4471, 0.6596, 0.5230],
  7395. [0.6794, 0.4385, 0.8418, 0.4789, 0.4136, 0.4619, 0.6039, 0.4783],
  7396. [0.1114, 0.0595, 0.8397, 0.2376, 0.5027, 0.1908, 0.6890, 0.5289],
  7397. [0.6909, 0.4345, 0.8839, 0.4388, 0.3304, 0.3869, 0.5642, 0.5191],
  7398. [0.6291, 0.3867, 0.8446, 0.4202, 0.3504, 0.3528, 0.5102, 0.5479],
  7399. [0.6586, 0.4093, 0.8303, 0.5168, 0.3792, 0.4555, 0.5674, 0.5253],
  7400. [0.7494, 0.4655, 0.8382, 0.2952, 0.3746, 0.3281, 0.6696, 0.5156],
  7401. [0.7090, 0.4590, 0.8363, 0.5264, 0.4275, 0.4275, 0.5747, 0.5815]],
  7402. device='cuda:0', grad_fn=<AddmmBackward>)
  7403. landmarks are: tensor([[[0.6274, 0.4003, 0.8637, 0.5967, 0.3688, 0.4900, 0.6108, 0.4661],
  7404. [0.6219, 0.3934, 0.8687, 0.5267, 0.4313, 0.4967, 0.5987, 0.4983],
  7405. [0.0000, 0.0000, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  7406. [0.6161, 0.4024, 0.8838, 0.4583, 0.3688, 0.3733, 0.5311, 0.5344],
  7407. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  7408. [0.6087, 0.3951, 0.8388, 0.5833, 0.4187, 0.4933, 0.5146, 0.4830],
  7409. [0.6255, 0.4017, 0.8687, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  7410. [0.6248, 0.4185, 0.8500, 0.5767, 0.4462, 0.4550, 0.5612, 0.5917]]],
  7411. device='cuda:0')
  7412. loss_train_step before backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  7413. loss_train_step after backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  7414. loss_train: 0.07608700543642044
  7415. step: 36
  7416. running loss: 0.002113527928789457
  7417.  
  7418. Train Steps: 36/90 Loss: 0.0021 torch.Size([8, 600, 800])
  7419. torch.Size([8, 8])
  7420. tensor([[0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  7421. [0.6289, 0.4081, 0.8720, 0.3487, 0.3900, 0.3183, 0.6703, 0.5376],
  7422. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  7423. [0.6162, 0.4014, 0.8800, 0.5333, 0.3750, 0.4817, 0.5988, 0.5283],
  7424. [0.6157, 0.4102, 0.8513, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  7425. [0.6257, 0.4034, 0.8287, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  7426. [0.6261, 0.4045, 0.8865, 0.5369, 0.3895, 0.4859, 0.6683, 0.5249],
  7427. [0.6259, 0.4133, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947]],
  7428. device='cuda:0', dtype=torch.float64)
  7429. predictions are: tensor([[0.6112, 0.3707, 0.7490, 0.3077, 0.3605, 0.2968, 0.5415, 0.5408],
  7430. [0.5955, 0.3761, 0.8708, 0.3528, 0.3607, 0.2691, 0.6453, 0.5391],
  7431. [0.5341, 0.3475, 0.8463, 0.5252, 0.4797, 0.4976, 0.5357, 0.5157],
  7432. [0.5937, 0.3833, 0.8467, 0.5804, 0.3934, 0.4725, 0.5941, 0.5159],
  7433. [0.5596, 0.3415, 0.8215, 0.3929, 0.3491, 0.3478, 0.5140, 0.5521],
  7434. [0.5897, 0.3710, 0.8079, 0.2786, 0.3989, 0.2553, 0.6312, 0.4956],
  7435. [0.6347, 0.4102, 0.8645, 0.5367, 0.3555, 0.4859, 0.6540, 0.5185],
  7436. [0.5740, 0.3737, 0.8167, 0.2516, 0.4687, 0.1609, 0.6137, 0.5002]],
  7437. device='cuda:0', grad_fn=<AddmmBackward>)
  7438. landmarks are: tensor([[[0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  7439. [0.6289, 0.4081, 0.8720, 0.3487, 0.3900, 0.3183, 0.6703, 0.5376],
  7440. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  7441. [0.6162, 0.4014, 0.8800, 0.5333, 0.3750, 0.4817, 0.5987, 0.5283],
  7442. [0.6157, 0.4102, 0.8512, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  7443. [0.6257, 0.4034, 0.8288, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  7444. [0.6261, 0.4045, 0.8865, 0.5369, 0.3895, 0.4859, 0.6683, 0.5249],
  7445. [0.6259, 0.4132, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947]]],
  7446. device='cuda:0')
  7447. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  7448. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  7449. loss_train: 0.07694085594266653
  7450. step: 37
  7451. running loss: 0.0020794825930450416
  7452. Train Steps: 37/90 Loss: 0.0021 torch.Size([8, 600, 800])
  7453. torch.Size([8, 8])
  7454. tensor([[0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  7455. [0.6111, 0.4033, 0.8300, 0.3267, 0.3588, 0.3333, 0.5444, 0.5637],
  7456. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6188, 0.5283],
  7457. [ nan, nan, 0.7412, 0.2200, 0.4450, 0.1517, 0.5312, 0.4983],
  7458. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  7459. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5363, 0.5550],
  7460. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  7461. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586]],
  7462. device='cuda:0', dtype=torch.float64)
  7463. predictions are: tensor([[ 0.5779, 0.3774, 0.8839, 0.4152, 0.4410, 0.2534, 0.5820, 0.5221],
  7464. [ 0.6797, 0.4455, 0.8260, 0.3675, 0.3468, 0.3362, 0.5157, 0.5417],
  7465. [ 0.6687, 0.4174, 0.8940, 0.4060, 0.3847, 0.2705, 0.5886, 0.5112],
  7466. [-0.0912, -0.0921, 0.6975, 0.2365, 0.4422, 0.1867, 0.5306, 0.5320],
  7467. [ 0.6604, 0.4093, 0.8252, 0.6504, 0.3691, 0.5073, 0.6017, 0.4858],
  7468. [ 0.6601, 0.4308, 0.6697, 0.2957, 0.3913, 0.2468, 0.5048, 0.5545],
  7469. [ 0.7503, 0.4870, 0.8212, 0.3324, 0.4210, 0.2805, 0.6182, 0.4864],
  7470. [ 0.6137, 0.3834, 0.8276, 0.2994, 0.4318, 0.2749, 0.6676, 0.5525]],
  7471. device='cuda:0', grad_fn=<AddmmBackward>)
  7472. landmarks are: tensor([[[0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  7473. [0.6111, 0.4033, 0.8300, 0.3267, 0.3587, 0.3333, 0.5444, 0.5637],
  7474. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6187, 0.5283],
  7475. [0.0000, 0.0000, 0.7412, 0.2200, 0.4450, 0.1517, 0.5312, 0.4983],
  7476. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  7477. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5362, 0.5550],
  7478. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  7479. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586]]],
  7480. device='cuda:0')
  7481. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  7482. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  7483. loss_train: 0.07825841219164431
  7484. step: 38
  7485. running loss: 0.0020594318997801133
  7486. Train Steps: 38/90 Loss: 0.0021 torch.Size([8, 600, 800])
  7487. torch.Size([8, 8])
  7488. tensor([[0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  7489. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  7490. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  7491. [0.6275, 0.4050, 0.9038, 0.3767, 0.3838, 0.3533, 0.7074, 0.5575],
  7492. [0.6289, 0.4019, 0.8113, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332],
  7493. [0.6364, 0.4165, 0.9088, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  7494. [0.6185, 0.4079, 0.8838, 0.4617, 0.4838, 0.5650, 0.6175, 0.5850],
  7495. [0.6140, 0.4070, 0.8700, 0.5000, 0.4612, 0.4900, 0.5260, 0.5852]],
  7496. device='cuda:0', dtype=torch.float64)
  7497. predictions are: tensor([[0.5947, 0.3805, 0.8456, 0.4897, 0.4121, 0.4539, 0.5597, 0.4984],
  7498. [0.3858, 0.2530, 0.7886, 0.3215, 0.3442, 0.2716, 0.4741, 0.5347],
  7499. [0.5989, 0.3812, 0.8558, 0.5530, 0.3637, 0.4137, 0.6011, 0.5037],
  7500. [0.6297, 0.4055, 0.8894, 0.3635, 0.3809, 0.3036, 0.6904, 0.5203],
  7501. [0.6403, 0.4136, 0.8121, 0.4900, 0.3675, 0.4450, 0.6480, 0.5108],
  7502. [0.5778, 0.3763, 0.8999, 0.4514, 0.4168, 0.2672, 0.6207, 0.5142],
  7503. [0.6699, 0.4190, 0.8753, 0.4437, 0.4606, 0.5210, 0.6239, 0.5325],
  7504. [0.5892, 0.3982, 0.8542, 0.4954, 0.4509, 0.4645, 0.5348, 0.5564]],
  7505. device='cuda:0', grad_fn=<AddmmBackward>)
  7506. landmarks are: tensor([[[0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  7507. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  7508. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  7509. [0.6275, 0.4050, 0.9038, 0.3767, 0.3837, 0.3533, 0.7074, 0.5575],
  7510. [0.6289, 0.4019, 0.8112, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332],
  7511. [0.6364, 0.4165, 0.9087, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  7512. [0.6184, 0.4079, 0.8838, 0.4617, 0.4837, 0.5650, 0.6175, 0.5850],
  7513. [0.6140, 0.4070, 0.8700, 0.5000, 0.4613, 0.4900, 0.5260, 0.5852]]],
  7514. device='cuda:0')
  7515. loss_train_step before backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  7516. loss_train_step after backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  7517. loss_train: 0.08022535732015967
  7518. step: 39
  7519. running loss: 0.0020570604441066584
  7520. Train Steps: 39/90 Loss: 0.0021 torch.Size([8, 600, 800])
  7521. torch.Size([8, 8])
  7522. tensor([[0.6277, 0.4118, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  7523. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791],
  7524. [ nan, nan, 0.7981, 0.3194, 0.3625, 0.3167, 0.5040, 0.5563],
  7525. [0.6159, 0.4085, 0.6900, 0.2283, 0.4088, 0.1950, 0.5123, 0.5397],
  7526. [0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038],
  7527. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5837, 0.5583],
  7528. [0.6143, 0.4040, 0.8237, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  7529. [0.6143, 0.4034, 0.8800, 0.4833, 0.4512, 0.5367, 0.5289, 0.5097]],
  7530. device='cuda:0', dtype=torch.float64)
  7531. predictions are: tensor([[ 0.6195, 0.4118, 0.9066, 0.4095, 0.3811, 0.2271, 0.6301, 0.5351],
  7532. [ 0.6931, 0.4617, 0.8825, 0.5103, 0.3938, 0.5106, 0.6760, 0.4836],
  7533. [-0.0069, -0.0022, 0.7917, 0.3328, 0.3496, 0.2844, 0.5340, 0.5845],
  7534. [ 0.5912, 0.3992, 0.6751, 0.2480, 0.3963, 0.1762, 0.5067, 0.5832],
  7535. [ 0.6294, 0.4174, 0.8644, 0.5082, 0.4266, 0.4852, 0.5873, 0.5055],
  7536. [ 0.7166, 0.4977, 0.8992, 0.4628, 0.4537, 0.5101, 0.6015, 0.5534],
  7537. [ 0.4912, 0.3237, 0.8256, 0.3268, 0.4155, 0.2062, 0.5291, 0.5372],
  7538. [ 0.7153, 0.4560, 0.8673, 0.5064, 0.4377, 0.4808, 0.5551, 0.5160]],
  7539. device='cuda:0', grad_fn=<AddmmBackward>)
  7540. landmarks are: tensor([[[0.6277, 0.4117, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  7541. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791],
  7542. [0.0000, 0.0000, 0.7981, 0.3194, 0.3625, 0.3167, 0.5040, 0.5563],
  7543. [0.6159, 0.4085, 0.6900, 0.2283, 0.4087, 0.1950, 0.5123, 0.5397],
  7544. [0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038],
  7545. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5838, 0.5583],
  7546. [0.6143, 0.4040, 0.8238, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  7547. [0.6143, 0.4034, 0.8800, 0.4833, 0.4512, 0.5367, 0.5289, 0.5097]]],
  7548. device='cuda:0')
  7549. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  7550. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  7551. loss_train: 0.08172035915777087
  7552. step: 40
  7553. running loss: 0.002043008978944272
  7554.  
  7555. Train Steps: 40/90 Loss: 0.0020 torch.Size([8, 600, 800])
  7556. torch.Size([8, 8])
  7557. tensor([[0.6229, 0.4107, 0.8137, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  7558. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5837, 0.5583],
  7559. [0.6200, 0.4098, 0.8237, 0.2917, 0.4012, 0.2967, 0.6000, 0.5683],
  7560. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  7561. [0.6274, 0.4003, 0.8638, 0.5967, 0.3688, 0.4900, 0.6108, 0.4661],
  7562. [0.6125, 0.3974, 0.7725, 0.2517, 0.3538, 0.3317, 0.5887, 0.5500],
  7563. [0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726],
  7564. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6063, 0.6083]],
  7565. device='cuda:0', dtype=torch.float64)
  7566. predictions are: tensor([[0.5758, 0.3786, 0.8423, 0.3171, 0.4899, 0.2170, 0.5905, 0.5425],
  7567. [0.5740, 0.3879, 0.9308, 0.4643, 0.4837, 0.5307, 0.5876, 0.5373],
  7568. [0.6391, 0.4176, 0.8629, 0.3083, 0.4227, 0.3055, 0.6264, 0.5473],
  7569. [0.5625, 0.3663, 0.8680, 0.4720, 0.3909, 0.4157, 0.5760, 0.5533],
  7570. [0.5354, 0.3332, 0.8488, 0.6208, 0.4072, 0.4879, 0.6330, 0.5155],
  7571. [0.5616, 0.3440, 0.7892, 0.2749, 0.3800, 0.3443, 0.5898, 0.5582],
  7572. [0.6584, 0.4165, 0.7801, 0.2906, 0.4001, 0.2840, 0.5664, 0.5037],
  7573. [0.5920, 0.4034, 0.8760, 0.4120, 0.3817, 0.4544, 0.5925, 0.5546]],
  7574. device='cuda:0', grad_fn=<AddmmBackward>)
  7575. landmarks are: tensor([[[0.6229, 0.4107, 0.8138, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  7576. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5838, 0.5583],
  7577. [0.6200, 0.4098, 0.8238, 0.2917, 0.4013, 0.2967, 0.6000, 0.5683],
  7578. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  7579. [0.6274, 0.4003, 0.8637, 0.5967, 0.3688, 0.4900, 0.6108, 0.4661],
  7580. [0.6125, 0.3974, 0.7725, 0.2517, 0.3537, 0.3317, 0.5888, 0.5500],
  7581. [0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726],
  7582. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6062, 0.6083]]],
  7583. device='cuda:0')
  7584. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  7585. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  7586. loss_train: 0.08272404491435736
  7587. step: 41
  7588. running loss: 0.002017659632057497
  7589. Train Steps: 41/90 Loss: 0.0020 torch.Size([8, 600, 800])
  7590. torch.Size([8, 8])
  7591. tensor([[0.6084, 0.3981, 0.8588, 0.5233, 0.4600, 0.5367, 0.5680, 0.5006],
  7592. [0.6200, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  7593. [0.6179, 0.4008, 0.8600, 0.4015, 0.3932, 0.2515, 0.5711, 0.5438],
  7594. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  7595. [0.6239, 0.4174, 0.8425, 0.5733, 0.4825, 0.4500, 0.5625, 0.5933],
  7596. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  7597. [0.6275, 0.4157, 0.8337, 0.5800, 0.3763, 0.4200, 0.5547, 0.6125],
  7598. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5963, 0.6217]],
  7599. device='cuda:0', dtype=torch.float64)
  7600. predictions are: tensor([[0.5543, 0.3742, 0.8943, 0.5196, 0.4794, 0.5638, 0.5621, 0.4876],
  7601. [0.6037, 0.3794, 0.9385, 0.4418, 0.4021, 0.4450, 0.5992, 0.5035],
  7602. [0.6467, 0.4319, 0.8756, 0.3346, 0.4146, 0.2846, 0.5714, 0.5388],
  7603. [0.5918, 0.3820, 0.8937, 0.4119, 0.3816, 0.4555, 0.5682, 0.5513],
  7604. [0.6275, 0.4428, 0.8985, 0.5641, 0.4904, 0.4356, 0.6108, 0.5810],
  7605. [0.5539, 0.3666, 0.8488, 0.2932, 0.3606, 0.4177, 0.6532, 0.5217],
  7606. [0.6402, 0.4327, 0.8611, 0.5682, 0.3874, 0.4667, 0.5907, 0.5768],
  7607. [0.5660, 0.3770, 0.7327, 0.2319, 0.4440, 0.2731, 0.5796, 0.6056]],
  7608. device='cuda:0', grad_fn=<AddmmBackward>)
  7609. landmarks are: tensor([[[0.6084, 0.3981, 0.8587, 0.5233, 0.4600, 0.5367, 0.5680, 0.5006],
  7610. [0.6201, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  7611. [0.6179, 0.4008, 0.8600, 0.4015, 0.3932, 0.2515, 0.5711, 0.5438],
  7612. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  7613. [0.6239, 0.4174, 0.8425, 0.5733, 0.4825, 0.4500, 0.5625, 0.5933],
  7614. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  7615. [0.6275, 0.4157, 0.8338, 0.5800, 0.3762, 0.4200, 0.5547, 0.6125],
  7616. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5962, 0.6217]]],
  7617. device='cuda:0')
  7618. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  7619. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  7620. loss_train: 0.08353073458420113
  7621. step: 42
  7622. running loss: 0.0019888270139095505
  7623. Train Steps: 42/90 Loss: 0.0020 torch.Size([8, 600, 800])
  7624. torch.Size([8, 8])
  7625. tensor([[0.6137, 0.4038, 0.8563, 0.4050, 0.3813, 0.2550, 0.5106, 0.4954],
  7626. [0.6199, 0.4060, 0.8888, 0.4667, 0.3800, 0.5050, 0.6188, 0.5433],
  7627. [0.6102, 0.4001, 0.7738, 0.3583, 0.3463, 0.3800, 0.5524, 0.5689],
  7628. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395],
  7629. [0.6293, 0.3982, 0.8700, 0.5300, 0.3763, 0.4717, 0.7050, 0.5297],
  7630. [0.6204, 0.4110, 0.7913, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367],
  7631. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  7632. [0.6263, 0.4065, 0.9038, 0.4317, 0.3588, 0.4550, 0.6325, 0.5250]],
  7633. device='cuda:0', dtype=torch.float64)
  7634. predictions are: tensor([[0.6746, 0.4319, 0.8910, 0.3957, 0.4231, 0.3041, 0.4968, 0.5360],
  7635. [0.6110, 0.4048, 0.9118, 0.4690, 0.4132, 0.5844, 0.6055, 0.5478],
  7636. [0.5277, 0.3620, 0.8001, 0.3551, 0.3683, 0.4127, 0.5339, 0.5777],
  7637. [0.6432, 0.4046, 0.7831, 0.2371, 0.4205, 0.2584, 0.5683, 0.5560],
  7638. [0.5422, 0.3705, 0.8758, 0.5354, 0.3853, 0.5174, 0.6447, 0.5465],
  7639. [0.6417, 0.4331, 0.8154, 0.2463, 0.4465, 0.2785, 0.5989, 0.5654],
  7640. [0.5830, 0.4069, 0.8817, 0.5367, 0.4828, 0.5881, 0.5900, 0.5317],
  7641. [0.6231, 0.4210, 0.9309, 0.4411, 0.3923, 0.4972, 0.6554, 0.5362]],
  7642. device='cuda:0', grad_fn=<AddmmBackward>)
  7643. landmarks are: tensor([[[0.6137, 0.4038, 0.8562, 0.4050, 0.3812, 0.2550, 0.5106, 0.4954],
  7644. [0.6199, 0.4060, 0.8888, 0.4667, 0.3800, 0.5050, 0.6187, 0.5433],
  7645. [0.6102, 0.4001, 0.7738, 0.3583, 0.3462, 0.3800, 0.5524, 0.5689],
  7646. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395],
  7647. [0.6293, 0.3982, 0.8700, 0.5300, 0.3762, 0.4717, 0.7050, 0.5297],
  7648. [0.6204, 0.4110, 0.7912, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367],
  7649. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  7650. [0.6263, 0.4065, 0.9038, 0.4317, 0.3587, 0.4550, 0.6325, 0.5250]]],
  7651. device='cuda:0')
  7652. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  7653. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  7654. loss_train: 0.08457283378811553
  7655. step: 43
  7656. running loss: 0.0019668100880957102
  7657. Train Steps: 43/90 Loss: 0.0020 torch.Size([8, 600, 800])
  7658. torch.Size([8, 8])
  7659. tensor([[0.6132, 0.4118, 0.8200, 0.3633, 0.3563, 0.5400, 0.5787, 0.5136],
  7660. [0.6272, 0.4045, 0.8538, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  7661. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  7662. [0.6246, 0.4008, 0.8757, 0.5088, 0.4101, 0.5392, 0.6644, 0.5133],
  7663. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6038, 0.6167],
  7664. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  7665. [0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050],
  7666. [0.6201, 0.4151, 0.8588, 0.5467, 0.3700, 0.3950, 0.5637, 0.5933]],
  7667. device='cuda:0', dtype=torch.float64)
  7668. predictions are: tensor([[0.5829, 0.3921, 0.8264, 0.3159, 0.3731, 0.5240, 0.5861, 0.5414],
  7669. [0.6574, 0.4408, 0.8602, 0.5335, 0.3847, 0.4092, 0.5730, 0.5130],
  7670. [0.6363, 0.4418, 0.8349, 0.5432, 0.3740, 0.5134, 0.6518, 0.5425],
  7671. [0.5697, 0.3814, 0.8596, 0.4887, 0.4200, 0.5548, 0.6287, 0.5410],
  7672. [0.6425, 0.4493, 0.8188, 0.2990, 0.3808, 0.4217, 0.5833, 0.5963],
  7673. [0.6742, 0.4422, 0.8950, 0.3099, 0.3901, 0.3213, 0.6032, 0.5729],
  7674. [0.6124, 0.4264, 0.8670, 0.3953, 0.3919, 0.5078, 0.5886, 0.5685],
  7675. [0.5960, 0.4155, 0.8576, 0.5290, 0.3820, 0.4110, 0.5455, 0.5905]],
  7676. device='cuda:0', grad_fn=<AddmmBackward>)
  7677. landmarks are: tensor([[[0.6132, 0.4118, 0.8200, 0.3633, 0.3562, 0.5400, 0.5787, 0.5136],
  7678. [0.6271, 0.4045, 0.8537, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  7679. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  7680. [0.6246, 0.4008, 0.8757, 0.5088, 0.4101, 0.5392, 0.6644, 0.5133],
  7681. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6037, 0.6167],
  7682. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  7683. [0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050],
  7684. [0.6202, 0.4151, 0.8587, 0.5467, 0.3700, 0.3950, 0.5638, 0.5933]]],
  7685. device='cuda:0')
  7686. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  7687. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  7688. loss_train: 0.08523785881698132
  7689. step: 44
  7690. running loss: 0.0019372240640223026
  7691.  
  7692. Train Steps: 44/90 Loss: 0.0019 torch.Size([8, 600, 800])
  7693. torch.Size([8, 8])
  7694. tensor([[0.6133, 0.4066, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358],
  7695. [0.6239, 0.4061, 0.8850, 0.4600, 0.4225, 0.5200, 0.6138, 0.5450],
  7696. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  7697. [0.6125, 0.3974, 0.7725, 0.2517, 0.3538, 0.3317, 0.5887, 0.5500],
  7698. [0.6267, 0.4065, 0.8313, 0.2467, 0.4788, 0.1733, 0.6312, 0.5133],
  7699. [0.6124, 0.4075, 0.7696, 0.4153, 0.3475, 0.3767, 0.5157, 0.5427],
  7700. [0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  7701. [0.6058, 0.3978, 0.8287, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461]],
  7702. device='cuda:0', dtype=torch.float64)
  7703. predictions are: tensor([[0.5574, 0.3708, 0.7304, 0.2770, 0.3952, 0.2549, 0.5315, 0.5325],
  7704. [0.5851, 0.4072, 0.8954, 0.4923, 0.4134, 0.6094, 0.6331, 0.5524],
  7705. [0.7421, 0.4717, 0.8658, 0.4165, 0.3477, 0.4763, 0.6099, 0.5323],
  7706. [0.6102, 0.3866, 0.7828, 0.2630, 0.3653, 0.3681, 0.5750, 0.5749],
  7707. [0.6527, 0.4293, 0.8477, 0.2623, 0.4776, 0.2003, 0.6449, 0.5496],
  7708. [0.7032, 0.4611, 0.8211, 0.4001, 0.3383, 0.4056, 0.5390, 0.5727],
  7709. [0.6002, 0.3951, 0.8682, 0.5551, 0.3725, 0.5488, 0.6113, 0.5342],
  7710. [0.5485, 0.3695, 0.8289, 0.3966, 0.3556, 0.4300, 0.5648, 0.5475]],
  7711. device='cuda:0', grad_fn=<AddmmBackward>)
  7712. landmarks are: tensor([[[0.6133, 0.4065, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358],
  7713. [0.6239, 0.4061, 0.8850, 0.4600, 0.4225, 0.5200, 0.6137, 0.5450],
  7714. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  7715. [0.6125, 0.3974, 0.7725, 0.2517, 0.3537, 0.3317, 0.5888, 0.5500],
  7716. [0.6266, 0.4065, 0.8313, 0.2467, 0.4787, 0.1733, 0.6313, 0.5133],
  7717. [0.6124, 0.4075, 0.7696, 0.4153, 0.3475, 0.3767, 0.5157, 0.5427],
  7718. [0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  7719. [0.6058, 0.3978, 0.8288, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461]]],
  7720. device='cuda:0')
  7721. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  7722. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  7723. loss_train: 0.08670229942072183
  7724. step: 45
  7725. running loss: 0.0019267177649049294
  7726. Train Steps: 45/90 Loss: 0.0019 torch.Size([8, 600, 800])
  7727. torch.Size([8, 8])
  7728. tensor([[0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  7729. [0.6265, 0.4071, 0.8875, 0.3367, 0.3975, 0.3350, 0.6312, 0.5250],
  7730. [0.6100, 0.4016, 0.8600, 0.5067, 0.4612, 0.5233, 0.5086, 0.5519],
  7731. [0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153],
  7732. [0.6250, 0.4103, 0.8950, 0.4400, 0.3912, 0.5650, 0.6050, 0.5133],
  7733. [ nan, nan, 0.7412, 0.2200, 0.4450, 0.1517, 0.5312, 0.4983],
  7734. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  7735. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633]],
  7736. device='cuda:0', dtype=torch.float64)
  7737. predictions are: tensor([[0.6801, 0.4410, 0.8683, 0.5659, 0.3392, 0.4688, 0.5803, 0.5495],
  7738. [0.6600, 0.4212, 0.8819, 0.3557, 0.3762, 0.3373, 0.6782, 0.5341],
  7739. [0.6769, 0.4397, 0.8591, 0.5386, 0.4188, 0.5211, 0.5393, 0.5377],
  7740. [0.6893, 0.4491, 0.8465, 0.5392, 0.4218, 0.5473, 0.6143, 0.5291],
  7741. [0.7045, 0.4556, 0.8790, 0.4567, 0.3647, 0.5870, 0.6346, 0.5346],
  7742. [0.2349, 0.1266, 0.7190, 0.2019, 0.4143, 0.1636, 0.5454, 0.5321],
  7743. [0.6809, 0.4381, 0.6721, 0.2559, 0.3333, 0.2282, 0.5364, 0.5570],
  7744. [0.6265, 0.4215, 0.8322, 0.3383, 0.3416, 0.3138, 0.5845, 0.5514]],
  7745. device='cuda:0', grad_fn=<AddmmBackward>)
  7746. landmarks are: tensor([[[0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  7747. [0.6265, 0.4071, 0.8875, 0.3367, 0.3975, 0.3350, 0.6313, 0.5250],
  7748. [0.6100, 0.4016, 0.8600, 0.5067, 0.4613, 0.5233, 0.5086, 0.5519],
  7749. [0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153],
  7750. [0.6250, 0.4103, 0.8950, 0.4400, 0.3913, 0.5650, 0.6050, 0.5133],
  7751. [0.0000, 0.0000, 0.7412, 0.2200, 0.4450, 0.1517, 0.5312, 0.4983],
  7752. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  7753. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633]]],
  7754. device='cuda:0')
  7755. loss_train_step before backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  7756. loss_train_step after backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  7757. loss_train: 0.08878360188100487
  7758. step: 46
  7759. running loss: 0.0019300783017609754
  7760. Train Steps: 46/90 Loss: 0.0019 torch.Size([8, 600, 800])
  7761. torch.Size([8, 8])
  7762. tensor([[0.6276, 0.4235, 0.8888, 0.5333, 0.3800, 0.3117, 0.5427, 0.6164],
  7763. [0.6307, 0.4045, 0.8025, 0.5833, 0.3775, 0.4867, 0.6892, 0.5459],
  7764. [ nan, nan, 0.6793, 0.2110, 0.4012, 0.2167, 0.5112, 0.5583],
  7765. [0.6132, 0.4118, 0.8200, 0.3633, 0.3563, 0.5400, 0.5787, 0.5136],
  7766. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  7767. [0.6164, 0.4119, 0.7913, 0.2650, 0.3538, 0.3500, 0.5614, 0.5038],
  7768. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  7769. [0.6277, 0.4083, 0.8350, 0.2717, 0.4562, 0.1800, 0.5918, 0.4878]],
  7770. device='cuda:0', dtype=torch.float64)
  7771. predictions are: tensor([[0.7055, 0.4578, 0.8519, 0.5879, 0.3761, 0.3916, 0.5511, 0.5944],
  7772. [0.7292, 0.4616, 0.8460, 0.5709, 0.3622, 0.4798, 0.6770, 0.4983],
  7773. [0.0664, 0.0312, 0.7004, 0.2303, 0.4082, 0.1988, 0.5215, 0.5506],
  7774. [0.6563, 0.4132, 0.8240, 0.3713, 0.3363, 0.5313, 0.5989, 0.5189],
  7775. [0.6488, 0.4153, 0.8015, 0.3351, 0.3190, 0.3517, 0.6047, 0.5187],
  7776. [0.7151, 0.4625, 0.7876, 0.2750, 0.3320, 0.3584, 0.5924, 0.4999],
  7777. [0.6842, 0.4461, 0.6953, 0.2469, 0.3832, 0.1930, 0.5175, 0.5420],
  7778. [0.6422, 0.3944, 0.8310, 0.2808, 0.4452, 0.2402, 0.5940, 0.5098]],
  7779. device='cuda:0', grad_fn=<AddmmBackward>)
  7780. landmarks are: tensor([[[0.6276, 0.4235, 0.8888, 0.5333, 0.3800, 0.3117, 0.5427, 0.6164],
  7781. [0.6307, 0.4045, 0.8025, 0.5833, 0.3775, 0.4867, 0.6892, 0.5459],
  7782. [0.0000, 0.0000, 0.6793, 0.2110, 0.4013, 0.2167, 0.5113, 0.5583],
  7783. [0.6132, 0.4118, 0.8200, 0.3633, 0.3562, 0.5400, 0.5787, 0.5136],
  7784. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  7785. [0.6164, 0.4119, 0.7912, 0.2650, 0.3537, 0.3500, 0.5614, 0.5038],
  7786. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  7787. [0.6277, 0.4083, 0.8350, 0.2717, 0.4563, 0.1800, 0.5918, 0.4878]]],
  7788. device='cuda:0')
  7789. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  7790. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  7791. loss_train: 0.09005057823378593
  7792. step: 47
  7793. running loss: 0.0019159697496550197
  7794. Train Steps: 47/90 Loss: 0.0019 torch.Size([8, 600, 800])
  7795. torch.Size([8, 8])
  7796. tensor([[0.6204, 0.4013, 0.8075, 0.2400, 0.4313, 0.2050, 0.5800, 0.5150],
  7797. [0.6185, 0.4129, 0.8900, 0.4567, 0.3937, 0.5417, 0.5734, 0.5110],
  7798. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167],
  7799. [0.6204, 0.4055, 0.8438, 0.5733, 0.4574, 0.4801, 0.5487, 0.5617],
  7800. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791],
  7801. [0.6264, 0.4248, 0.8938, 0.4183, 0.3875, 0.4100, 0.5400, 0.5350],
  7802. [0.6078, 0.4033, 0.8019, 0.3055, 0.3450, 0.4200, 0.6025, 0.5550],
  7803. [0.6113, 0.4088, 0.6859, 0.2208, 0.4363, 0.1700, 0.5188, 0.5533]],
  7804. device='cuda:0', dtype=torch.float64)
  7805. predictions are: tensor([[0.6486, 0.4020, 0.7514, 0.2730, 0.4087, 0.2175, 0.5757, 0.5366],
  7806. [0.6411, 0.4125, 0.8790, 0.4814, 0.3593, 0.5086, 0.5643, 0.5240],
  7807. [0.6816, 0.4637, 0.7844, 0.3338, 0.3328, 0.2747, 0.5542, 0.5338],
  7808. [0.6323, 0.3974, 0.7978, 0.5710, 0.4151, 0.4028, 0.5774, 0.5467],
  7809. [0.5792, 0.3522, 0.8565, 0.4896, 0.3740, 0.4974, 0.6258, 0.4819],
  7810. [0.6051, 0.3964, 0.8591, 0.4106, 0.3487, 0.3718, 0.5399, 0.5469],
  7811. [0.5822, 0.3830, 0.7570, 0.2944, 0.3506, 0.4014, 0.6029, 0.5421],
  7812. [0.3315, 0.2145, 0.6800, 0.2316, 0.4172, 0.1543, 0.5413, 0.5392]],
  7813. device='cuda:0', grad_fn=<AddmmBackward>)
  7814. landmarks are: tensor([[[0.6204, 0.4013, 0.8075, 0.2400, 0.4313, 0.2050, 0.5800, 0.5150],
  7815. [0.6186, 0.4129, 0.8900, 0.4567, 0.3938, 0.5417, 0.5734, 0.5110],
  7816. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167],
  7817. [0.6204, 0.4055, 0.8438, 0.5733, 0.4574, 0.4801, 0.5487, 0.5617],
  7818. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791],
  7819. [0.6264, 0.4248, 0.8938, 0.4183, 0.3875, 0.4100, 0.5400, 0.5350],
  7820. [0.6078, 0.4033, 0.8019, 0.3055, 0.3450, 0.4200, 0.6025, 0.5550],
  7821. [0.6113, 0.4088, 0.6859, 0.2208, 0.4363, 0.1700, 0.5188, 0.5533]]],
  7822. device='cuda:0')
  7823. loss_train_step before backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  7824. loss_train_step after backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  7825. loss_train: 0.09267542359884828
  7826. step: 48
  7827. running loss: 0.0019307379916426726
  7828.  
  7829. Train Steps: 48/90 Loss: 0.0019 torch.Size([8, 600, 800])
  7830. torch.Size([8, 8])
  7831. tensor([[0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  7832. [0.6143, 0.4040, 0.8237, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  7833. [0.6198, 0.4164, 0.8700, 0.5067, 0.4625, 0.5650, 0.5464, 0.5197],
  7834. [0.6339, 0.4102, 0.9088, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  7835. [0.6030, 0.3969, 0.7988, 0.3917, 0.3450, 0.3667, 0.5266, 0.4700],
  7836. [0.6192, 0.4128, 0.8513, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  7837. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5513, 0.5750],
  7838. [0.6308, 0.3990, 0.8688, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133]],
  7839. device='cuda:0', dtype=torch.float64)
  7840. predictions are: tensor([[0.6052, 0.3889, 0.7422, 0.3402, 0.3384, 0.3641, 0.5236, 0.5260],
  7841. [0.6399, 0.4004, 0.7969, 0.3118, 0.4094, 0.1803, 0.5158, 0.5109],
  7842. [0.5728, 0.3562, 0.8510, 0.4912, 0.4429, 0.4912, 0.5814, 0.5052],
  7843. [0.5621, 0.3503, 0.8764, 0.4556, 0.4044, 0.5215, 0.6876, 0.5352],
  7844. [0.6308, 0.4028, 0.8012, 0.3729, 0.3408, 0.3155, 0.5477, 0.5014],
  7845. [0.6521, 0.4160, 0.8504, 0.5561, 0.4178, 0.4928, 0.5914, 0.5374],
  7846. [0.5205, 0.3282, 0.6716, 0.2172, 0.4256, 0.1459, 0.5180, 0.5464],
  7847. [0.6104, 0.3535, 0.8501, 0.5325, 0.3972, 0.4680, 0.6415, 0.4927]],
  7848. device='cuda:0', grad_fn=<AddmmBackward>)
  7849. landmarks are: tensor([[[0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  7850. [0.6143, 0.4040, 0.8238, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  7851. [0.6198, 0.4164, 0.8700, 0.5067, 0.4625, 0.5650, 0.5464, 0.5197],
  7852. [0.6339, 0.4102, 0.9087, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  7853. [0.6030, 0.3969, 0.7987, 0.3917, 0.3450, 0.3667, 0.5266, 0.4700],
  7854. [0.6192, 0.4128, 0.8512, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  7855. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5512, 0.5750],
  7856. [0.6308, 0.3990, 0.8687, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133]]],
  7857. device='cuda:0')
  7858. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  7859. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  7860. loss_train: 0.09376236831303686
  7861. step: 49
  7862. running loss: 0.0019135177206742217
  7863. Train Steps: 49/90 Loss: 0.0019 torch.Size([8, 600, 800])
  7864. torch.Size([8, 8])
  7865. tensor([[0.6147, 0.4112, 0.7988, 0.3200, 0.3775, 0.2767, 0.5150, 0.5550],
  7866. [0.6265, 0.4088, 0.8025, 0.1850, 0.4163, 0.2500, 0.6290, 0.4947],
  7867. [0.6068, 0.3963, 0.8650, 0.4317, 0.4037, 0.5083, 0.5253, 0.4999],
  7868. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  7869. [0.6329, 0.4055, 0.9050, 0.4783, 0.3613, 0.3917, 0.6464, 0.5019],
  7870. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  7871. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  7872. [0.6199, 0.4102, 0.8950, 0.4417, 0.4012, 0.5367, 0.6112, 0.5967]],
  7873. device='cuda:0', dtype=torch.float64)
  7874. predictions are: tensor([[0.6403, 0.4320, 0.7649, 0.3199, 0.3983, 0.2335, 0.4917, 0.5320],
  7875. [0.5692, 0.3730, 0.7705, 0.2126, 0.4442, 0.1903, 0.5995, 0.5071],
  7876. [0.5403, 0.3438, 0.8224, 0.4247, 0.4037, 0.4621, 0.5542, 0.4882],
  7877. [0.5514, 0.3663, 0.8637, 0.4957, 0.3862, 0.3813, 0.6040, 0.5017],
  7878. [0.6259, 0.3877, 0.8745, 0.4758, 0.3988, 0.3611, 0.6208, 0.5046],
  7879. [0.5511, 0.3389, 0.8570, 0.5071, 0.4225, 0.5230, 0.6351, 0.5395],
  7880. [0.5870, 0.3975, 0.7331, 0.3451, 0.3606, 0.3595, 0.5102, 0.5234],
  7881. [0.6117, 0.3787, 0.8476, 0.4176, 0.4175, 0.4941, 0.5977, 0.5509]],
  7882. device='cuda:0', grad_fn=<AddmmBackward>)
  7883. landmarks are: tensor([[[0.6147, 0.4112, 0.7987, 0.3200, 0.3775, 0.2767, 0.5150, 0.5550],
  7884. [0.6265, 0.4088, 0.8025, 0.1850, 0.4162, 0.2500, 0.6290, 0.4947],
  7885. [0.6068, 0.3963, 0.8650, 0.4317, 0.4038, 0.5083, 0.5253, 0.4999],
  7886. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  7887. [0.6329, 0.4055, 0.9050, 0.4783, 0.3613, 0.3917, 0.6464, 0.5019],
  7888. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  7889. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  7890. [0.6199, 0.4102, 0.8950, 0.4417, 0.4013, 0.5367, 0.6112, 0.5967]]],
  7891. device='cuda:0')
  7892. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  7893. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  7894. loss_train: 0.09513336268719286
  7895. step: 50
  7896. running loss: 0.0019026672537438571
  7897. Train Steps: 50/90 Loss: 0.0019 torch.Size([8, 600, 800])
  7898. torch.Size([8, 8])
  7899. tensor([[0.6142, 0.4127, 0.7575, 0.3067, 0.3438, 0.4383, 0.5778, 0.5207],
  7900. [0.6198, 0.4101, 0.8838, 0.5283, 0.3763, 0.5267, 0.5913, 0.5567],
  7901. [0.6264, 0.4248, 0.8938, 0.4183, 0.3875, 0.4100, 0.5400, 0.5350],
  7902. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  7903. [0.6339, 0.4118, 0.7988, 0.5800, 0.3912, 0.4583, 0.7343, 0.5760],
  7904. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385],
  7905. [0.6102, 0.3999, 0.8750, 0.5133, 0.3825, 0.4750, 0.5637, 0.5083],
  7906. [0.6293, 0.4097, 0.8800, 0.2517, 0.5262, 0.2600, 0.7430, 0.5378]],
  7907. device='cuda:0', dtype=torch.float64)
  7908. predictions are: tensor([[0.5291, 0.3672, 0.7294, 0.2730, 0.3650, 0.3765, 0.5397, 0.5277],
  7909. [0.5886, 0.3764, 0.8469, 0.5366, 0.3838, 0.5122, 0.5506, 0.5424],
  7910. [0.5200, 0.3578, 0.8842, 0.4014, 0.3746, 0.3781, 0.5141, 0.5359],
  7911. [0.5578, 0.3773, 0.7889, 0.2673, 0.4389, 0.2098, 0.5802, 0.5160],
  7912. [0.5594, 0.3759, 0.8057, 0.5207, 0.3739, 0.4169, 0.6299, 0.5353],
  7913. [0.6054, 0.4064, 0.7891, 0.2735, 0.4530, 0.2321, 0.6273, 0.5279],
  7914. [0.6244, 0.3972, 0.8628, 0.5094, 0.3892, 0.4647, 0.5318, 0.5040],
  7915. [0.5981, 0.3899, 0.8339, 0.2679, 0.5203, 0.2410, 0.6810, 0.5113]],
  7916. device='cuda:0', grad_fn=<AddmmBackward>)
  7917. landmarks are: tensor([[[0.6142, 0.4127, 0.7575, 0.3067, 0.3438, 0.4383, 0.5778, 0.5207],
  7918. [0.6198, 0.4101, 0.8838, 0.5283, 0.3762, 0.5267, 0.5913, 0.5567],
  7919. [0.6264, 0.4248, 0.8938, 0.4183, 0.3875, 0.4100, 0.5400, 0.5350],
  7920. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  7921. [0.6339, 0.4118, 0.7987, 0.5800, 0.3913, 0.4583, 0.7343, 0.5760],
  7922. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385],
  7923. [0.6102, 0.3999, 0.8750, 0.5133, 0.3825, 0.4750, 0.5638, 0.5083],
  7924. [0.6293, 0.4097, 0.8800, 0.2517, 0.5263, 0.2600, 0.7430, 0.5378]]],
  7925. device='cuda:0')
  7926. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  7927. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  7928. loss_train: 0.09657937183510512
  7929. step: 51
  7930. running loss: 0.0018937131732373552
  7931. Train Steps: 51/90 Loss: 0.0019 torch.Size([8, 600, 800])
  7932. torch.Size([8, 8])
  7933. tensor([[0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367],
  7934. [0.6207, 0.4081, 0.7662, 0.2067, 0.3962, 0.3200, 0.6312, 0.5300],
  7935. [0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  7936. [0.6243, 0.4128, 0.7762, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417],
  7937. [0.6353, 0.4128, 0.9138, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  7938. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  7939. [ nan, nan, 0.7525, 0.2291, 0.3838, 0.3017, 0.6050, 0.5667],
  7940. [0.6030, 0.3969, 0.7988, 0.3917, 0.3450, 0.3667, 0.5266, 0.4700]],
  7941. device='cuda:0', dtype=torch.float64)
  7942. predictions are: tensor([[0.6311, 0.4221, 0.8473, 0.2995, 0.4195, 0.2622, 0.6039, 0.5202],
  7943. [0.7261, 0.5016, 0.7322, 0.2343, 0.3933, 0.3053, 0.5926, 0.5256],
  7944. [0.6551, 0.4545, 0.8713, 0.4405, 0.3823, 0.5575, 0.5866, 0.5082],
  7945. [0.6100, 0.4321, 0.7779, 0.2898, 0.4315, 0.2631, 0.6073, 0.5497],
  7946. [0.4372, 0.2929, 0.9085, 0.3609, 0.4803, 0.3184, 0.6927, 0.5634],
  7947. [0.6183, 0.4245, 0.8867, 0.5138, 0.3870, 0.4361, 0.6212, 0.5193],
  7948. [0.1437, 0.1349, 0.7505, 0.2220, 0.3988, 0.2872, 0.5788, 0.5707],
  7949. [0.6677, 0.4697, 0.8175, 0.4028, 0.3731, 0.3748, 0.5594, 0.5090]],
  7950. device='cuda:0', grad_fn=<AddmmBackward>)
  7951. landmarks are: tensor([[[0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367],
  7952. [0.6207, 0.4081, 0.7663, 0.2067, 0.3963, 0.3200, 0.6313, 0.5300],
  7953. [0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  7954. [0.6243, 0.4128, 0.7763, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417],
  7955. [0.6353, 0.4128, 0.9137, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  7956. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  7957. [0.0000, 0.0000, 0.7525, 0.2291, 0.3837, 0.3017, 0.6050, 0.5667],
  7958. [0.6030, 0.3969, 0.7987, 0.3917, 0.3450, 0.3667, 0.5266, 0.4700]]],
  7959. device='cuda:0')
  7960. loss_train_step before backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  7961. loss_train_step after backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  7962. loss_train: 0.0988936327630654
  7963. step: 52
  7964. running loss: 0.00190180063005895
  7965.  
  7966. Train Steps: 52/90 Loss: 0.0019 torch.Size([8, 600, 800])
  7967. torch.Size([8, 8])
  7968. tensor([[0.6202, 0.4079, 0.8025, 0.2500, 0.3763, 0.3217, 0.6125, 0.5533],
  7969. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  7970. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  7971. [0.6200, 0.4098, 0.8237, 0.2917, 0.4012, 0.2967, 0.6000, 0.5683],
  7972. [0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332],
  7973. [0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367],
  7974. [0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274],
  7975. [0.6091, 0.3997, 0.8314, 0.4334, 0.3788, 0.4550, 0.5213, 0.5656]],
  7976. device='cuda:0', dtype=torch.float64)
  7977. predictions are: tensor([[0.5353, 0.3823, 0.7938, 0.2598, 0.3850, 0.3384, 0.6041, 0.5397],
  7978. [0.5435, 0.3924, 0.8487, 0.3284, 0.3678, 0.4744, 0.6533, 0.5549],
  7979. [0.6516, 0.4569, 0.8807, 0.4752, 0.3808, 0.4926, 0.6206, 0.5455],
  7980. [0.5597, 0.3985, 0.8237, 0.2700, 0.4093, 0.3204, 0.6297, 0.5742],
  7981. [0.6297, 0.4217, 0.9052, 0.4837, 0.4001, 0.4403, 0.6905, 0.5351],
  7982. [0.5590, 0.3859, 0.8605, 0.2797, 0.4072, 0.2892, 0.6317, 0.5315],
  7983. [0.5734, 0.4131, 0.8632, 0.3005, 0.4720, 0.2418, 0.6259, 0.5304],
  7984. [0.6172, 0.4364, 0.8233, 0.4241, 0.3716, 0.4831, 0.6026, 0.5529]],
  7985. device='cuda:0', grad_fn=<AddmmBackward>)
  7986. landmarks are: tensor([[[0.6202, 0.4079, 0.8025, 0.2500, 0.3762, 0.3217, 0.6125, 0.5533],
  7987. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  7988. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  7989. [0.6200, 0.4098, 0.8238, 0.2917, 0.4013, 0.2967, 0.6000, 0.5683],
  7990. [0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332],
  7991. [0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367],
  7992. [0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274],
  7993. [0.6091, 0.3997, 0.8314, 0.4334, 0.3787, 0.4550, 0.5213, 0.5656]]],
  7994. device='cuda:0')
  7995. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  7996. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  7997. loss_train: 0.09974715590942651
  7998. step: 53
  7999. running loss: 0.001882021809611821
  8000. Train Steps: 53/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8001. torch.Size([8, 8])
  8002. tensor([[0.6147, 0.4081, 0.8538, 0.3400, 0.3663, 0.3150, 0.5142, 0.4875],
  8003. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  8004. [0.6115, 0.4081, 0.6725, 0.2433, 0.4088, 0.1933, 0.5167, 0.5544],
  8005. [0.6275, 0.4050, 0.9038, 0.3767, 0.3838, 0.3533, 0.7074, 0.5575],
  8006. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  8007. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600],
  8008. [0.6293, 0.3982, 0.8700, 0.5300, 0.3763, 0.4717, 0.7050, 0.5297],
  8009. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483]],
  8010. device='cuda:0', dtype=torch.float64)
  8011. predictions are: tensor([[0.6275, 0.4306, 0.8911, 0.3299, 0.3758, 0.3486, 0.5929, 0.5505],
  8012. [0.5658, 0.4011, 0.9082, 0.4369, 0.4359, 0.5329, 0.6209, 0.5843],
  8013. [0.7188, 0.4968, 0.7221, 0.2397, 0.4145, 0.2572, 0.5564, 0.5696],
  8014. [0.6036, 0.4003, 0.9182, 0.3324, 0.4068, 0.3903, 0.7308, 0.5581],
  8015. [0.5868, 0.3989, 0.9414, 0.4409, 0.3893, 0.4048, 0.6734, 0.5355],
  8016. [0.6512, 0.4482, 0.9158, 0.5223, 0.4085, 0.5194, 0.6330, 0.5632],
  8017. [0.6254, 0.4223, 0.8987, 0.4838, 0.3649, 0.5039, 0.7165, 0.5501],
  8018. [0.5276, 0.3468, 0.8923, 0.3737, 0.3886, 0.5233, 0.6415, 0.5414]],
  8019. device='cuda:0', grad_fn=<AddmmBackward>)
  8020. landmarks are: tensor([[[0.6147, 0.4081, 0.8537, 0.3400, 0.3663, 0.3150, 0.5142, 0.4875],
  8021. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  8022. [0.6115, 0.4081, 0.6725, 0.2433, 0.4087, 0.1933, 0.5167, 0.5544],
  8023. [0.6275, 0.4050, 0.9038, 0.3767, 0.3837, 0.3533, 0.7074, 0.5575],
  8024. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  8025. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600],
  8026. [0.6293, 0.3982, 0.8700, 0.5300, 0.3762, 0.4717, 0.7050, 0.5297],
  8027. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483]]],
  8028. device='cuda:0')
  8029. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  8030. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  8031. loss_train: 0.10123581916559488
  8032. step: 54
  8033. running loss: 0.0018747373919554607
  8034. Train Steps: 54/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8035. torch.Size([8, 8])
  8036. tensor([[ nan, nan, 0.7192, 0.2346, 0.4037, 0.2050, 0.5138, 0.5650],
  8037. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6063, 0.6083],
  8038. [ nan, nan, 0.7425, 0.2117, 0.3937, 0.2433, 0.5438, 0.5567],
  8039. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167],
  8040. [0.6333, 0.4037, 0.8638, 0.5733, 0.4012, 0.4717, 0.6369, 0.4938],
  8041. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  8042. [0.6275, 0.4048, 0.8488, 0.2883, 0.4463, 0.2033, 0.6321, 0.5155],
  8043. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343]],
  8044. device='cuda:0', dtype=torch.float64)
  8045. predictions are: tensor([[0.2724, 0.1738, 0.7532, 0.2123, 0.3912, 0.2749, 0.5379, 0.5518],
  8046. [0.7120, 0.4843, 0.8970, 0.3871, 0.3452, 0.4813, 0.6276, 0.5857],
  8047. [0.2392, 0.1629, 0.7932, 0.2384, 0.3899, 0.2904, 0.5817, 0.5716],
  8048. [0.7815, 0.5277, 0.8028, 0.2982, 0.3764, 0.3326, 0.6129, 0.6341],
  8049. [0.7014, 0.4486, 0.9404, 0.5730, 0.3902, 0.5327, 0.6732, 0.5283],
  8050. [0.6658, 0.4433, 0.8476, 0.2791, 0.4124, 0.2763, 0.6565, 0.5462],
  8051. [0.7444, 0.4732, 0.8730, 0.2678, 0.4320, 0.2968, 0.6825, 0.5032],
  8052. [0.6798, 0.4357, 0.8951, 0.2959, 0.4816, 0.3000, 0.7251, 0.5187]],
  8053. device='cuda:0', grad_fn=<AddmmBackward>)
  8054. landmarks are: tensor([[[0.0000, 0.0000, 0.7192, 0.2346, 0.4038, 0.2050, 0.5138, 0.5650],
  8055. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6062, 0.6083],
  8056. [0.0000, 0.0000, 0.7425, 0.2117, 0.3938, 0.2433, 0.5437, 0.5567],
  8057. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167],
  8058. [0.6334, 0.4037, 0.8637, 0.5733, 0.4013, 0.4717, 0.6369, 0.4938],
  8059. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  8060. [0.6275, 0.4048, 0.8487, 0.2883, 0.4462, 0.2033, 0.6321, 0.5155],
  8061. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343]]],
  8062. device='cuda:0')
  8063. loss_train_step before backward: tensor(0.0053, device='cuda:0', grad_fn=<MseLossBackward>)
  8064. loss_train_step after backward: tensor(0.0053, device='cuda:0', grad_fn=<MseLossBackward>)
  8065. loss_train: 0.1065472379559651
  8066. step: 55
  8067. running loss: 0.0019372225082902746
  8068. Train Steps: 55/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8069. torch.Size([8, 8])
  8070. tensor([[0.6128, 0.4022, 0.8738, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  8071. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  8072. [0.6132, 0.4066, 0.7259, 0.2402, 0.3588, 0.3300, 0.6000, 0.5600],
  8073. [0.6289, 0.4024, 0.9088, 0.4567, 0.3937, 0.5633, 0.7058, 0.5609],
  8074. [0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332],
  8075. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  8076. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  8077. [0.6336, 0.4191, 0.8938, 0.5167, 0.3937, 0.3517, 0.7343, 0.5748]],
  8078. device='cuda:0', dtype=torch.float64)
  8079. predictions are: tensor([[0.5691, 0.3698, 0.9165, 0.4752, 0.4648, 0.4836, 0.5714, 0.5409],
  8080. [0.5765, 0.3781, 0.8772, 0.3890, 0.3343, 0.3800, 0.5563, 0.5370],
  8081. [0.6562, 0.4239, 0.7415, 0.2279, 0.3311, 0.3317, 0.6071, 0.5846],
  8082. [0.6207, 0.3919, 0.9110, 0.4470, 0.3838, 0.5880, 0.6730, 0.5643],
  8083. [0.6225, 0.3718, 0.9335, 0.4698, 0.3647, 0.3998, 0.6734, 0.5408],
  8084. [0.6486, 0.4084, 0.9166, 0.4715, 0.3560, 0.4715, 0.6107, 0.5510],
  8085. [0.6746, 0.4125, 0.8632, 0.2376, 0.4625, 0.2011, 0.6717, 0.5324],
  8086. [0.5819, 0.3607, 0.9126, 0.4763, 0.3888, 0.3742, 0.6958, 0.5375]],
  8087. device='cuda:0', grad_fn=<AddmmBackward>)
  8088. landmarks are: tensor([[[0.6128, 0.4022, 0.8737, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  8089. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  8090. [0.6132, 0.4066, 0.7259, 0.2402, 0.3587, 0.3300, 0.6000, 0.5600],
  8091. [0.6289, 0.4024, 0.9087, 0.4567, 0.3938, 0.5633, 0.7058, 0.5609],
  8092. [0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332],
  8093. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  8094. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  8095. [0.6336, 0.4191, 0.8938, 0.5167, 0.3938, 0.3517, 0.7343, 0.5748]]],
  8096. device='cuda:0')
  8097. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  8098. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  8099. loss_train: 0.10731903073610738
  8100. step: 56
  8101. running loss: 0.0019164112631447747
  8102.  
  8103. Train Steps: 56/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8104. torch.Size([8, 8])
  8105. tensor([[0.6307, 0.4045, 0.8025, 0.5833, 0.3775, 0.4867, 0.6892, 0.5459],
  8106. [0.6114, 0.4018, 0.7213, 0.1967, 0.3763, 0.2700, 0.5875, 0.5533],
  8107. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  8108. [0.6165, 0.4106, 0.7575, 0.1733, 0.3838, 0.2650, 0.5680, 0.5116],
  8109. [0.6111, 0.4033, 0.8300, 0.3267, 0.3588, 0.3333, 0.5444, 0.5637],
  8110. [0.6218, 0.4185, 0.7338, 0.2650, 0.4625, 0.1950, 0.5687, 0.5800],
  8111. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  8112. [0.6102, 0.3999, 0.8750, 0.5133, 0.3825, 0.4750, 0.5637, 0.5083]],
  8113. device='cuda:0', dtype=torch.float64)
  8114. predictions are: tensor([[0.6403, 0.3795, 0.8980, 0.5622, 0.3909, 0.4747, 0.7285, 0.5281],
  8115. [0.6636, 0.4230, 0.7843, 0.2408, 0.3759, 0.2624, 0.5990, 0.5570],
  8116. [0.4426, 0.2650, 0.7819, 0.2717, 0.4223, 0.2305, 0.5698, 0.5799],
  8117. [0.6475, 0.3757, 0.8123, 0.2576, 0.3823, 0.2330, 0.6141, 0.4856],
  8118. [0.5844, 0.3666, 0.9030, 0.3338, 0.3551, 0.3270, 0.5759, 0.5615],
  8119. [0.5405, 0.3511, 0.7870, 0.2629, 0.4318, 0.2087, 0.6067, 0.5856],
  8120. [0.6199, 0.3592, 0.9353, 0.5180, 0.4166, 0.4974, 0.6105, 0.5446],
  8121. [0.6197, 0.3734, 0.9563, 0.5340, 0.3956, 0.4882, 0.6105, 0.5443]],
  8122. device='cuda:0', grad_fn=<AddmmBackward>)
  8123. landmarks are: tensor([[[0.6307, 0.4045, 0.8025, 0.5833, 0.3775, 0.4867, 0.6892, 0.5459],
  8124. [0.6114, 0.4018, 0.7212, 0.1967, 0.3762, 0.2700, 0.5875, 0.5533],
  8125. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  8126. [0.6165, 0.4106, 0.7575, 0.1733, 0.3837, 0.2650, 0.5680, 0.5116],
  8127. [0.6111, 0.4033, 0.8300, 0.3267, 0.3587, 0.3333, 0.5444, 0.5637],
  8128. [0.6218, 0.4185, 0.7337, 0.2650, 0.4625, 0.1950, 0.5688, 0.5800],
  8129. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  8130. [0.6102, 0.3999, 0.8750, 0.5133, 0.3825, 0.4750, 0.5638, 0.5083]]],
  8131. device='cuda:0')
  8132. loss_train_step before backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  8133. loss_train_step after backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  8134. loss_train: 0.10952302097575739
  8135. step: 57
  8136. running loss: 0.0019214565083466209
  8137. Train Steps: 57/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8138. torch.Size([8, 8])
  8139. tensor([[0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  8140. [0.6127, 0.4084, 0.8700, 0.4467, 0.3987, 0.4317, 0.5013, 0.5471],
  8141. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  8142. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6188, 0.5283],
  8143. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  8144. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  8145. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  8146. [0.6125, 0.3999, 0.8750, 0.4883, 0.4750, 0.4700, 0.5533, 0.5617]],
  8147. device='cuda:0', dtype=torch.float64)
  8148. predictions are: tensor([[0.6365, 0.3743, 0.8675, 0.4952, 0.4117, 0.5146, 0.6097, 0.5063],
  8149. [0.5609, 0.3361, 0.8477, 0.4395, 0.3576, 0.3986, 0.4763, 0.5326],
  8150. [0.6198, 0.3910, 0.8201, 0.2905, 0.4859, 0.1939, 0.6222, 0.5119],
  8151. [0.5869, 0.3545, 0.8819, 0.3543, 0.3917, 0.2430, 0.6047, 0.5221],
  8152. [0.5527, 0.3526, 0.8673, 0.5385, 0.3781, 0.4188, 0.5554, 0.5453],
  8153. [0.5270, 0.3048, 0.7687, 0.2961, 0.3601, 0.3038, 0.5805, 0.5237],
  8154. [0.6488, 0.4108, 0.8584, 0.4499, 0.4041, 0.3312, 0.6700, 0.5362],
  8155. [0.6019, 0.3609, 0.8609, 0.4872, 0.4596, 0.4562, 0.5581, 0.5371]],
  8156. device='cuda:0', grad_fn=<AddmmBackward>)
  8157. landmarks are: tensor([[[0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  8158. [0.6127, 0.4084, 0.8700, 0.4467, 0.3988, 0.4317, 0.5013, 0.5471],
  8159. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  8160. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6187, 0.5283],
  8161. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  8162. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  8163. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  8164. [0.6125, 0.3999, 0.8750, 0.4883, 0.4750, 0.4700, 0.5533, 0.5617]]],
  8165. device='cuda:0')
  8166. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  8167. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  8168. loss_train: 0.11065401375526562
  8169. step: 58
  8170. running loss: 0.0019078278233666487
  8171. Train Steps: 58/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8172. torch.Size([8, 8])
  8173. tensor([[0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090],
  8174. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  8175. [0.6311, 0.3998, 0.7975, 0.5767, 0.3838, 0.4850, 0.7327, 0.5343],
  8176. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  8177. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  8178. [0.6289, 0.4019, 0.8113, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332],
  8179. [0.6257, 0.4167, 0.8775, 0.3433, 0.3563, 0.4133, 0.6200, 0.5667],
  8180. [0.6325, 0.4165, 0.9000, 0.4617, 0.3813, 0.4900, 0.7485, 0.5447]],
  8181. device='cuda:0', dtype=torch.float64)
  8182. predictions are: tensor([[0.6700, 0.4326, 0.8465, 0.3759, 0.4052, 0.2541, 0.5767, 0.4936],
  8183. [0.6088, 0.3693, 0.8447, 0.4587, 0.3971, 0.4752, 0.5430, 0.5230],
  8184. [0.6091, 0.3827, 0.7853, 0.5151, 0.3908, 0.4268, 0.6323, 0.5240],
  8185. [0.6106, 0.3826, 0.8802, 0.4890, 0.3949, 0.3110, 0.5897, 0.5250],
  8186. [0.6589, 0.4166, 0.8583, 0.4883, 0.4730, 0.5148, 0.5638, 0.5646],
  8187. [0.6102, 0.3873, 0.8047, 0.5130, 0.3933, 0.4348, 0.6198, 0.5365],
  8188. [0.5807, 0.3623, 0.8555, 0.3527, 0.3793, 0.3696, 0.5950, 0.5563],
  8189. [0.5650, 0.3652, 0.8887, 0.4821, 0.4247, 0.4604, 0.6303, 0.5394]],
  8190. device='cuda:0', grad_fn=<AddmmBackward>)
  8191. landmarks are: tensor([[[0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090],
  8192. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  8193. [0.6311, 0.3998, 0.7975, 0.5767, 0.3837, 0.4850, 0.7327, 0.5343],
  8194. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  8195. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  8196. [0.6289, 0.4019, 0.8112, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332],
  8197. [0.6257, 0.4167, 0.8775, 0.3433, 0.3562, 0.4133, 0.6200, 0.5667],
  8198. [0.6325, 0.4165, 0.9000, 0.4617, 0.3812, 0.4900, 0.7485, 0.5447]]],
  8199. device='cuda:0')
  8200. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  8201. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  8202. loss_train: 0.11230159673141316
  8203. step: 59
  8204. running loss: 0.0019034168937527654
  8205. Train Steps: 59/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8206. torch.Size([8, 8])
  8207. tensor([[0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385],
  8208. [0.6164, 0.4076, 0.8838, 0.4117, 0.3713, 0.5550, 0.6238, 0.5350],
  8209. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  8210. [0.6239, 0.4123, 0.8313, 0.2550, 0.4500, 0.2050, 0.6175, 0.5400],
  8211. [0.6202, 0.4066, 0.8746, 0.3376, 0.3717, 0.3090, 0.5842, 0.5165],
  8212. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  8213. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  8214. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5637, 0.5633]],
  8215. device='cuda:0', dtype=torch.float64)
  8216. predictions are: tensor([[0.5976, 0.3865, 0.7750, 0.2826, 0.4411, 0.2298, 0.6490, 0.5309],
  8217. [0.5556, 0.3312, 0.8283, 0.3899, 0.3939, 0.5214, 0.6137, 0.5141],
  8218. [0.6032, 0.3724, 0.7987, 0.5199, 0.4078, 0.4533, 0.5771, 0.5130],
  8219. [0.5418, 0.3389, 0.7947, 0.2667, 0.4400, 0.1755, 0.5935, 0.5198],
  8220. [0.5723, 0.3736, 0.8127, 0.3381, 0.3666, 0.2736, 0.5842, 0.5161],
  8221. [0.6249, 0.4152, 0.7927, 0.4392, 0.4423, 0.2585, 0.5467, 0.5873],
  8222. [0.5716, 0.3700, 0.8444, 0.4802, 0.4240, 0.4934, 0.5635, 0.5217],
  8223. [0.5619, 0.3722, 0.7979, 0.4927, 0.3437, 0.3396, 0.5290, 0.5390]],
  8224. device='cuda:0', grad_fn=<AddmmBackward>)
  8225. landmarks are: tensor([[[0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385],
  8226. [0.6164, 0.4076, 0.8838, 0.4117, 0.3713, 0.5550, 0.6237, 0.5350],
  8227. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  8228. [0.6239, 0.4123, 0.8313, 0.2550, 0.4500, 0.2050, 0.6175, 0.5400],
  8229. [0.6202, 0.4066, 0.8746, 0.3376, 0.3717, 0.3090, 0.5842, 0.5165],
  8230. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  8231. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  8232. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5638, 0.5633]]],
  8233. device='cuda:0')
  8234. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  8235. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  8236. loss_train: 0.11372045130701736
  8237. step: 60
  8238. running loss: 0.001895340855116956
  8239.  
  8240. Train Steps: 60/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8241. torch.Size([8, 8])
  8242. tensor([[0.6275, 0.4003, 0.9100, 0.3783, 0.4388, 0.3133, 0.7058, 0.5343],
  8243. [0.6229, 0.4066, 0.8513, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  8244. [0.6268, 0.4052, 0.8175, 0.2250, 0.4688, 0.1917, 0.6375, 0.5267],
  8245. [0.6164, 0.4076, 0.8838, 0.4117, 0.3713, 0.5550, 0.6238, 0.5350],
  8246. [0.6274, 0.4099, 0.8625, 0.3233, 0.4400, 0.1983, 0.5876, 0.4869],
  8247. [0.6098, 0.3991, 0.8638, 0.4717, 0.4263, 0.4967, 0.5212, 0.5650],
  8248. [ nan, nan, 0.7525, 0.2291, 0.3838, 0.3017, 0.6050, 0.5667],
  8249. [0.6085, 0.4008, 0.8588, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283]],
  8250. device='cuda:0', dtype=torch.float64)
  8251. predictions are: tensor([[0.6158, 0.4215, 0.8445, 0.3986, 0.4052, 0.2963, 0.6601, 0.5171],
  8252. [0.6921, 0.4721, 0.8280, 0.5742, 0.4215, 0.4660, 0.6214, 0.5356],
  8253. [0.6141, 0.4159, 0.7788, 0.2511, 0.4412, 0.1787, 0.6674, 0.5213],
  8254. [0.6449, 0.4211, 0.8393, 0.3993, 0.3831, 0.5536, 0.6376, 0.5329],
  8255. [0.6885, 0.4565, 0.8322, 0.3377, 0.4438, 0.2241, 0.6034, 0.5038],
  8256. [0.6514, 0.4302, 0.8149, 0.5266, 0.4088, 0.4930, 0.5682, 0.5626],
  8257. [0.1568, 0.1365, 0.7287, 0.2559, 0.3713, 0.2906, 0.6051, 0.5638],
  8258. [0.6785, 0.4605, 0.8245, 0.5355, 0.4814, 0.4635, 0.5484, 0.5490]],
  8259. device='cuda:0', grad_fn=<AddmmBackward>)
  8260. landmarks are: tensor([[[0.6275, 0.4003, 0.9100, 0.3783, 0.4387, 0.3133, 0.7058, 0.5343],
  8261. [0.6229, 0.4066, 0.8512, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  8262. [0.6268, 0.4052, 0.8175, 0.2250, 0.4688, 0.1917, 0.6375, 0.5267],
  8263. [0.6164, 0.4076, 0.8838, 0.4117, 0.3713, 0.5550, 0.6237, 0.5350],
  8264. [0.6274, 0.4099, 0.8625, 0.3233, 0.4400, 0.1983, 0.5876, 0.4869],
  8265. [0.6098, 0.3991, 0.8637, 0.4717, 0.4263, 0.4967, 0.5213, 0.5650],
  8266. [0.0000, 0.0000, 0.7525, 0.2291, 0.3837, 0.3017, 0.6050, 0.5667],
  8267. [0.6084, 0.4008, 0.8587, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283]]],
  8268. device='cuda:0')
  8269. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  8270. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  8271. loss_train: 0.11536864080699161
  8272. step: 61
  8273. running loss: 0.0018912891935572395
  8274. Train Steps: 61/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8275. torch.Size([8, 8])
  8276. tensor([[0.6218, 0.4137, 0.7263, 0.2233, 0.4075, 0.2650, 0.6212, 0.5783],
  8277. [0.6196, 0.4094, 0.7562, 0.2817, 0.3937, 0.3183, 0.6013, 0.6183],
  8278. [0.6145, 0.4007, 0.8775, 0.4533, 0.4562, 0.5533, 0.6088, 0.5533],
  8279. [0.6125, 0.4035, 0.7825, 0.3100, 0.3463, 0.4900, 0.5832, 0.5637],
  8280. [0.6085, 0.4008, 0.8588, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  8281. [0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  8282. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  8283. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012]],
  8284. device='cuda:0', dtype=torch.float64)
  8285. predictions are: tensor([[0.6308, 0.4368, 0.7588, 0.2792, 0.4103, 0.2608, 0.6189, 0.5617],
  8286. [0.6197, 0.4305, 0.7728, 0.3165, 0.4232, 0.3204, 0.6322, 0.5996],
  8287. [0.6046, 0.4186, 0.8744, 0.4556, 0.4631, 0.5276, 0.6331, 0.5470],
  8288. [0.5291, 0.3804, 0.7644, 0.3500, 0.3569, 0.4627, 0.6394, 0.5231],
  8289. [0.6324, 0.4379, 0.8579, 0.5266, 0.5035, 0.4578, 0.5769, 0.5433],
  8290. [0.6046, 0.4234, 0.8726, 0.4919, 0.3836, 0.4204, 0.5508, 0.5633],
  8291. [0.5945, 0.4069, 0.8636, 0.4195, 0.3478, 0.3993, 0.6374, 0.5284],
  8292. [0.6733, 0.4624, 0.8645, 0.5685, 0.3725, 0.4319, 0.6626, 0.5058]],
  8293. device='cuda:0', grad_fn=<AddmmBackward>)
  8294. landmarks are: tensor([[[0.6218, 0.4137, 0.7262, 0.2233, 0.4075, 0.2650, 0.6212, 0.5783],
  8295. [0.6196, 0.4094, 0.7563, 0.2817, 0.3938, 0.3183, 0.6012, 0.6183],
  8296. [0.6145, 0.4007, 0.8775, 0.4533, 0.4563, 0.5533, 0.6087, 0.5533],
  8297. [0.6125, 0.4035, 0.7825, 0.3100, 0.3462, 0.4900, 0.5832, 0.5637],
  8298. [0.6084, 0.4008, 0.8587, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  8299. [0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  8300. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  8301. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012]]],
  8302. device='cuda:0')
  8303. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  8304. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  8305. loss_train: 0.11608802812406793
  8306. step: 62
  8307. running loss: 0.0018723875503881924
  8308. Train Steps: 62/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8309. torch.Size([8, 8])
  8310. tensor([[0.6273, 0.4100, 0.7137, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633],
  8311. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  8312. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6362, 0.5200],
  8313. [0.6245, 0.4115, 0.8700, 0.4883, 0.4625, 0.5517, 0.6100, 0.5217],
  8314. [0.6271, 0.4040, 0.9138, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  8315. [0.6117, 0.4019, 0.8538, 0.4067, 0.3513, 0.3583, 0.5663, 0.5133],
  8316. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  8317. [0.6170, 0.4102, 0.7468, 0.3695, 0.3463, 0.3767, 0.5238, 0.5823]],
  8318. device='cuda:0', dtype=torch.float64)
  8319. predictions are: tensor([[0.4617, 0.3219, 0.7192, 0.2454, 0.3976, 0.2679, 0.6069, 0.5669],
  8320. [0.6731, 0.4611, 0.7892, 0.2466, 0.4606, 0.2235, 0.6119, 0.5364],
  8321. [0.5985, 0.4165, 0.8816, 0.3927, 0.3924, 0.5346, 0.6360, 0.5497],
  8322. [0.6380, 0.4531, 0.8811, 0.4980, 0.4677, 0.5485, 0.6470, 0.5719],
  8323. [0.5778, 0.3894, 0.9415, 0.3813, 0.4868, 0.3180, 0.7241, 0.5422],
  8324. [0.6515, 0.4434, 0.8502, 0.4100, 0.3677, 0.3639, 0.5602, 0.5561],
  8325. [0.6955, 0.4848, 0.9014, 0.4948, 0.4301, 0.5923, 0.6377, 0.5528],
  8326. [0.5887, 0.4287, 0.7657, 0.3732, 0.3652, 0.4024, 0.5393, 0.5895]],
  8327. device='cuda:0', grad_fn=<AddmmBackward>)
  8328. landmarks are: tensor([[[0.6273, 0.4099, 0.7138, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633],
  8329. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  8330. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6363, 0.5200],
  8331. [0.6245, 0.4115, 0.8700, 0.4883, 0.4625, 0.5517, 0.6100, 0.5217],
  8332. [0.6271, 0.4040, 0.9137, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  8333. [0.6116, 0.4019, 0.8537, 0.4067, 0.3512, 0.3583, 0.5663, 0.5133],
  8334. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  8335. [0.6170, 0.4102, 0.7468, 0.3695, 0.3462, 0.3767, 0.5238, 0.5823]]],
  8336. device='cuda:0')
  8337. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  8338. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  8339. loss_train: 0.11741927900584415
  8340. step: 63
  8341. running loss: 0.0018637980794578436
  8342. Train Steps: 63/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8343. torch.Size([8, 8])
  8344. tensor([[0.6230, 0.4152, 0.7588, 0.2283, 0.4012, 0.2883, 0.6200, 0.5767],
  8345. [0.6275, 0.4003, 0.9100, 0.3783, 0.4388, 0.3133, 0.7058, 0.5343],
  8346. [0.6179, 0.4008, 0.7505, 0.2678, 0.4368, 0.1891, 0.5831, 0.5263],
  8347. [0.6219, 0.4114, 0.8175, 0.2817, 0.3925, 0.2783, 0.5900, 0.5350],
  8348. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  8349. [0.6084, 0.3981, 0.8588, 0.5233, 0.4600, 0.5367, 0.5680, 0.5006],
  8350. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  8351. [0.6346, 0.4144, 0.9088, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899]],
  8352. device='cuda:0', dtype=torch.float64)
  8353. predictions are: tensor([[0.5823, 0.3989, 0.7990, 0.2686, 0.4015, 0.3397, 0.6064, 0.5830],
  8354. [0.5737, 0.3859, 0.9048, 0.3874, 0.4089, 0.3633, 0.6578, 0.5424],
  8355. [0.6790, 0.4570, 0.7580, 0.2675, 0.4338, 0.2390, 0.5607, 0.5454],
  8356. [0.5070, 0.3484, 0.7983, 0.2682, 0.4036, 0.3215, 0.6074, 0.5580],
  8357. [0.4992, 0.3424, 0.7282, 0.2410, 0.4013, 0.2917, 0.5396, 0.5811],
  8358. [0.5697, 0.3986, 0.8845, 0.5299, 0.4589, 0.5847, 0.5493, 0.5481],
  8359. [0.6039, 0.4162, 0.8152, 0.2024, 0.4612, 0.2351, 0.6230, 0.5262],
  8360. [0.6638, 0.4316, 0.9216, 0.4938, 0.4082, 0.4667, 0.6866, 0.5826]],
  8361. device='cuda:0', grad_fn=<AddmmBackward>)
  8362. landmarks are: tensor([[[0.6230, 0.4152, 0.7588, 0.2283, 0.4013, 0.2883, 0.6200, 0.5767],
  8363. [0.6275, 0.4003, 0.9100, 0.3783, 0.4387, 0.3133, 0.7058, 0.5343],
  8364. [0.6179, 0.4008, 0.7505, 0.2678, 0.4368, 0.1891, 0.5831, 0.5263],
  8365. [0.6219, 0.4114, 0.8175, 0.2817, 0.3925, 0.2783, 0.5900, 0.5350],
  8366. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  8367. [0.6084, 0.3981, 0.8587, 0.5233, 0.4600, 0.5367, 0.5680, 0.5006],
  8368. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  8369. [0.6346, 0.4144, 0.9087, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899]]],
  8370. device='cuda:0')
  8371. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  8372. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  8373. loss_train: 0.11882246885215864
  8374. step: 64
  8375. running loss: 0.0018566010758149787
  8376.  
  8377. Train Steps: 64/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8378. torch.Size([8, 8])
  8379. tensor([[0.6203, 0.4021, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  8380. [0.6200, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913],
  8381. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461],
  8382. [0.6230, 0.4113, 0.7213, 0.1983, 0.4325, 0.2367, 0.6262, 0.5400],
  8383. [ nan, nan, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  8384. [0.6148, 0.3996, 0.8488, 0.3867, 0.3488, 0.4067, 0.5863, 0.5000],
  8385. [0.6266, 0.4067, 0.8588, 0.2867, 0.4300, 0.2850, 0.6325, 0.5267],
  8386. [0.6133, 0.4066, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358]],
  8387. device='cuda:0', dtype=torch.float64)
  8388. predictions are: tensor([[0.7349, 0.4675, 0.9156, 0.5079, 0.3783, 0.4450, 0.5984, 0.5317],
  8389. [0.6379, 0.4151, 0.8800, 0.5041, 0.3912, 0.5593, 0.5922, 0.5318],
  8390. [0.6555, 0.4495, 0.8806, 0.5197, 0.4423, 0.5433, 0.5737, 0.5766],
  8391. [0.5777, 0.3901, 0.7510, 0.2130, 0.4300, 0.2646, 0.6502, 0.5579],
  8392. [0.3214, 0.2092, 0.8647, 0.2586, 0.5082, 0.2396, 0.6842, 0.5704],
  8393. [0.6745, 0.4494, 0.8771, 0.3857, 0.3560, 0.4396, 0.5858, 0.5374],
  8394. [0.6192, 0.3965, 0.8787, 0.2707, 0.4508, 0.3048, 0.6362, 0.5435],
  8395. [0.5929, 0.4056, 0.7187, 0.2436, 0.3843, 0.2642, 0.5445, 0.5340]],
  8396. device='cuda:0', grad_fn=<AddmmBackward>)
  8397. landmarks are: tensor([[[0.6203, 0.4020, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  8398. [0.6199, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913],
  8399. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461],
  8400. [0.6230, 0.4113, 0.7212, 0.1983, 0.4325, 0.2367, 0.6263, 0.5400],
  8401. [0.0000, 0.0000, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  8402. [0.6148, 0.3996, 0.8487, 0.3867, 0.3487, 0.4067, 0.5863, 0.5000],
  8403. [0.6266, 0.4067, 0.8587, 0.2867, 0.4300, 0.2850, 0.6325, 0.5267],
  8404. [0.6133, 0.4065, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358]]],
  8405. device='cuda:0')
  8406. loss_train_step before backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  8407. loss_train_step after backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  8408. loss_train: 0.12204406381351873
  8409. step: 65
  8410. running loss: 0.001877600981746442
  8411. Train Steps: 65/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8412. torch.Size([8, 8])
  8413. tensor([[0.6124, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  8414. [0.6162, 0.4134, 0.6700, 0.2467, 0.3962, 0.2533, 0.5737, 0.5467],
  8415. [0.6126, 0.4073, 0.8750, 0.5133, 0.3800, 0.4333, 0.4986, 0.5378],
  8416. [0.6110, 0.3984, 0.8750, 0.4933, 0.4625, 0.4950, 0.5578, 0.5676],
  8417. [0.6156, 0.4125, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  8418. [0.6311, 0.3998, 0.7975, 0.5767, 0.3838, 0.4850, 0.7327, 0.5343],
  8419. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  8420. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285]],
  8421. device='cuda:0', dtype=torch.float64)
  8422. predictions are: tensor([[0.6549, 0.4234, 0.8641, 0.4460, 0.3999, 0.4543, 0.6240, 0.5543],
  8423. [0.5980, 0.3909, 0.7557, 0.2231, 0.3643, 0.2611, 0.6287, 0.5349],
  8424. [0.6213, 0.4000, 0.9014, 0.4747, 0.3929, 0.4499, 0.5685, 0.5125],
  8425. [0.6375, 0.4072, 0.9264, 0.4221, 0.4738, 0.5030, 0.6107, 0.5567],
  8426. [0.6489, 0.4262, 0.9240, 0.4144, 0.4471, 0.5674, 0.6234, 0.5403],
  8427. [0.6403, 0.4147, 0.8330, 0.4742, 0.3766, 0.4779, 0.7220, 0.5145],
  8428. [0.6165, 0.4096, 0.8961, 0.4373, 0.4341, 0.4900, 0.6077, 0.5471],
  8429. [0.6455, 0.4126, 0.8287, 0.3788, 0.3643, 0.4262, 0.5727, 0.5199]],
  8430. device='cuda:0', grad_fn=<AddmmBackward>)
  8431. landmarks are: tensor([[[0.6123, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  8432. [0.6162, 0.4134, 0.6700, 0.2467, 0.3963, 0.2533, 0.5738, 0.5467],
  8433. [0.6126, 0.4073, 0.8750, 0.5133, 0.3800, 0.4333, 0.4986, 0.5378],
  8434. [0.6110, 0.3984, 0.8750, 0.4933, 0.4625, 0.4950, 0.5578, 0.5676],
  8435. [0.6155, 0.4124, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  8436. [0.6311, 0.3998, 0.7975, 0.5767, 0.3837, 0.4850, 0.7327, 0.5343],
  8437. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  8438. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285]]],
  8439. device='cuda:0')
  8440. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  8441. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  8442. loss_train: 0.12342670414363965
  8443. step: 66
  8444. running loss: 0.001870101577933934
  8445. Train Steps: 66/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8446. torch.Size([8, 8])
  8447. tensor([[0.6293, 0.4097, 0.8800, 0.2517, 0.5262, 0.2600, 0.7430, 0.5378],
  8448. [0.6286, 0.4078, 0.8063, 0.2267, 0.4788, 0.1533, 0.5953, 0.4913],
  8449. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  8450. [0.6286, 0.4060, 0.9188, 0.4333, 0.3675, 0.4167, 0.7034, 0.5528],
  8451. [0.6339, 0.4123, 0.8638, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  8452. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  8453. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  8454. [0.6196, 0.4094, 0.7562, 0.2817, 0.3937, 0.3183, 0.6013, 0.6183]],
  8455. device='cuda:0', dtype=torch.float64)
  8456. predictions are: tensor([[0.6517, 0.3950, 0.8952, 0.2544, 0.5097, 0.2601, 0.6829, 0.5222],
  8457. [0.5733, 0.3255, 0.7971, 0.2400, 0.4828, 0.2072, 0.5727, 0.5022],
  8458. [0.5915, 0.3806, 0.7819, 0.2226, 0.3704, 0.3152, 0.5553, 0.5218],
  8459. [0.6712, 0.3939, 0.9119, 0.4217, 0.3616, 0.4310, 0.6332, 0.5328],
  8460. [0.5975, 0.3927, 0.8889, 0.5248, 0.4028, 0.5850, 0.6696, 0.5266],
  8461. [0.6125, 0.3791, 0.8563, 0.5179, 0.4040, 0.5349, 0.5519, 0.5101],
  8462. [0.5716, 0.3439, 0.9131, 0.4123, 0.3465, 0.3948, 0.5630, 0.5008],
  8463. [0.6269, 0.3833, 0.7719, 0.2893, 0.4000, 0.3447, 0.5547, 0.5907]],
  8464. device='cuda:0', grad_fn=<AddmmBackward>)
  8465. landmarks are: tensor([[[0.6293, 0.4097, 0.8800, 0.2517, 0.5263, 0.2600, 0.7430, 0.5378],
  8466. [0.6286, 0.4078, 0.8062, 0.2267, 0.4787, 0.1533, 0.5953, 0.4913],
  8467. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  8468. [0.6286, 0.4060, 0.9187, 0.4333, 0.3675, 0.4167, 0.7034, 0.5528],
  8469. [0.6339, 0.4123, 0.8637, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  8470. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  8471. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  8472. [0.6196, 0.4094, 0.7563, 0.2817, 0.3938, 0.3183, 0.6012, 0.6183]]],
  8473. device='cuda:0')
  8474. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  8475. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  8476. loss_train: 0.12444062257418409
  8477. step: 67
  8478. running loss: 0.0018573227249878222
  8479. Train Steps: 67/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8480. torch.Size([8, 8])
  8481. tensor([[0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  8482. [0.6254, 0.4076, 0.8700, 0.3267, 0.4150, 0.3083, 0.7050, 0.5609],
  8483. [0.6212, 0.4159, 0.8675, 0.5783, 0.4088, 0.4317, 0.5613, 0.5917],
  8484. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  8485. [0.6228, 0.4119, 0.7938, 0.2233, 0.4674, 0.1773, 0.6188, 0.5433],
  8486. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  8487. [0.6178, 0.4012, 0.8900, 0.4495, 0.3891, 0.5617, 0.5972, 0.4879],
  8488. [0.6257, 0.4034, 0.8287, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901]],
  8489. device='cuda:0', dtype=torch.float64)
  8490. predictions are: tensor([[0.6532, 0.3863, 0.8381, 0.2773, 0.4113, 0.2732, 0.5710, 0.5276],
  8491. [0.6314, 0.3613, 0.8673, 0.3157, 0.4232, 0.2928, 0.6520, 0.5310],
  8492. [0.5641, 0.3614, 0.8284, 0.5704, 0.3935, 0.4518, 0.5438, 0.5686],
  8493. [0.6321, 0.3723, 0.8681, 0.2921, 0.5005, 0.2325, 0.6820, 0.5124],
  8494. [0.6510, 0.3914, 0.7841, 0.2322, 0.4774, 0.1835, 0.5655, 0.5282],
  8495. [0.5351, 0.3280, 0.8166, 0.3076, 0.3767, 0.2982, 0.5308, 0.5306],
  8496. [0.5455, 0.3412, 0.8776, 0.4614, 0.3899, 0.5606, 0.5658, 0.4976],
  8497. [0.5498, 0.3304, 0.8194, 0.2611, 0.3887, 0.2648, 0.5999, 0.4929]],
  8498. device='cuda:0', grad_fn=<AddmmBackward>)
  8499. landmarks are: tensor([[[0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  8500. [0.6254, 0.4076, 0.8700, 0.3267, 0.4150, 0.3083, 0.7050, 0.5609],
  8501. [0.6212, 0.4159, 0.8675, 0.5783, 0.4087, 0.4317, 0.5612, 0.5917],
  8502. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  8503. [0.6228, 0.4119, 0.7937, 0.2233, 0.4674, 0.1773, 0.6187, 0.5433],
  8504. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  8505. [0.6178, 0.4012, 0.8900, 0.4495, 0.3891, 0.5617, 0.5972, 0.4879],
  8506. [0.6257, 0.4034, 0.8288, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901]]],
  8507. device='cuda:0')
  8508. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  8509. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  8510. loss_train: 0.1254878489417024
  8511. step: 68
  8512. running loss: 0.0018454095432603294
  8513.  
  8514. Train Steps: 68/90 Loss: 0.0018 torch.Size([8, 600, 800])
  8515. torch.Size([8, 8])
  8516. tensor([[0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500],
  8517. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  8518. [ nan, nan, 0.8938, 0.2850, 0.4662, 0.3117, 0.7406, 0.5528],
  8519. [0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726],
  8520. [0.6138, 0.4101, 0.8800, 0.5083, 0.4637, 0.5950, 0.5587, 0.5077],
  8521. [0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5413, 0.5433],
  8522. [ nan, nan, 0.7240, 0.2722, 0.3900, 0.2567, 0.5168, 0.5933],
  8523. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667]],
  8524. device='cuda:0', dtype=torch.float64)
  8525. predictions are: tensor([[0.6844, 0.4405, 0.7799, 0.3047, 0.3637, 0.3610, 0.6419, 0.5207],
  8526. [0.7517, 0.4813, 0.7875, 0.3211, 0.4087, 0.2684, 0.5113, 0.5277],
  8527. [0.3289, 0.1807, 0.9043, 0.3134, 0.4653, 0.2704, 0.7535, 0.5364],
  8528. [0.6930, 0.4234, 0.7588, 0.2919, 0.3988, 0.2571, 0.5680, 0.4735],
  8529. [0.6735, 0.4149, 0.8795, 0.5366, 0.4842, 0.5513, 0.5983, 0.5389],
  8530. [0.6846, 0.4375, 0.8348, 0.3652, 0.3751, 0.2941, 0.5715, 0.5281],
  8531. [0.2488, 0.1441, 0.7344, 0.2680, 0.4080, 0.2224, 0.5436, 0.5447],
  8532. [0.7038, 0.4378, 0.8358, 0.3522, 0.3714, 0.3495, 0.5815, 0.5219]],
  8533. device='cuda:0', grad_fn=<AddmmBackward>)
  8534. landmarks are: tensor([[[0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500],
  8535. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  8536. [0.0000, 0.0000, 0.8938, 0.2850, 0.4663, 0.3117, 0.7406, 0.5528],
  8537. [0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726],
  8538. [0.6138, 0.4101, 0.8800, 0.5083, 0.4638, 0.5950, 0.5587, 0.5077],
  8539. [0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5412, 0.5433],
  8540. [0.0000, 0.0000, 0.7240, 0.2722, 0.3900, 0.2567, 0.5168, 0.5933],
  8541. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667]]],
  8542. device='cuda:0')
  8543. loss_train_step before backward: tensor(0.0049, device='cuda:0', grad_fn=<MseLossBackward>)
  8544. loss_train_step after backward: tensor(0.0049, device='cuda:0', grad_fn=<MseLossBackward>)
  8545. loss_train: 0.13038400636287406
  8546. step: 69
  8547. running loss: 0.0018896232806213632
  8548. Train Steps: 69/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8549. torch.Size([8, 8])
  8550. tensor([[0.6219, 0.3934, 0.8688, 0.5267, 0.4313, 0.4967, 0.5988, 0.4983],
  8551. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  8552. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  8553. [0.6100, 0.4016, 0.8600, 0.5067, 0.4612, 0.5233, 0.5086, 0.5519],
  8554. [0.6185, 0.4067, 0.8838, 0.4450, 0.4037, 0.4733, 0.5213, 0.5142],
  8555. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  8556. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  8557. [0.6128, 0.4115, 0.7934, 0.3778, 0.3450, 0.4033, 0.5337, 0.5456]],
  8558. device='cuda:0', dtype=torch.float64)
  8559. predictions are: tensor([[0.5951, 0.3895, 0.8416, 0.5235, 0.4252, 0.4983, 0.6261, 0.4952],
  8560. [0.5503, 0.3507, 0.8777, 0.4197, 0.3331, 0.4423, 0.6457, 0.5161],
  8561. [0.6041, 0.3976, 0.7719, 0.2042, 0.4376, 0.1779, 0.6123, 0.5478],
  8562. [0.5624, 0.3707, 0.8516, 0.5060, 0.4581, 0.5141, 0.5622, 0.5226],
  8563. [0.5725, 0.3693, 0.8532, 0.4400, 0.3778, 0.4570, 0.5595, 0.5110],
  8564. [0.5750, 0.3745, 0.7993, 0.2587, 0.4015, 0.2159, 0.6403, 0.5213],
  8565. [0.6475, 0.4099, 0.8714, 0.3639, 0.4452, 0.2523, 0.6694, 0.5656],
  8566. [0.6218, 0.3965, 0.7890, 0.3606, 0.3206, 0.4008, 0.5353, 0.5258]],
  8567. device='cuda:0', grad_fn=<AddmmBackward>)
  8568. landmarks are: tensor([[[0.6219, 0.3934, 0.8687, 0.5267, 0.4313, 0.4967, 0.5987, 0.4983],
  8569. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  8570. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  8571. [0.6100, 0.4016, 0.8600, 0.5067, 0.4613, 0.5233, 0.5086, 0.5519],
  8572. [0.6185, 0.4067, 0.8838, 0.4450, 0.4038, 0.4733, 0.5213, 0.5142],
  8573. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  8574. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  8575. [0.6128, 0.4115, 0.7934, 0.3778, 0.3450, 0.4033, 0.5337, 0.5456]]],
  8576. device='cuda:0')
  8577. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  8578. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  8579. loss_train: 0.13097077485872433
  8580. step: 70
  8581. running loss: 0.0018710110694103476
  8582. Train Steps: 70/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8583. torch.Size([8, 8])
  8584. tensor([[0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  8585. [0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  8586. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  8587. [ nan, nan, 0.6900, 0.1917, 0.3937, 0.2367, 0.5240, 0.5246],
  8588. [0.6069, 0.3975, 0.8625, 0.5083, 0.4388, 0.5483, 0.5650, 0.4967],
  8589. [0.6260, 0.4153, 0.9000, 0.4533, 0.4025, 0.2633, 0.6223, 0.4967],
  8590. [0.6276, 0.4095, 0.8237, 0.2250, 0.4662, 0.1783, 0.6171, 0.4869],
  8591. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309]],
  8592. device='cuda:0', dtype=torch.float64)
  8593. predictions are: tensor([[0.6149, 0.4330, 0.7990, 0.2836, 0.4278, 0.1979, 0.5801, 0.5593],
  8594. [0.6108, 0.4046, 0.8090, 0.4209, 0.3469, 0.4819, 0.6012, 0.5438],
  8595. [0.5945, 0.3862, 0.8863, 0.4450, 0.3350, 0.4499, 0.6335, 0.5160],
  8596. [0.1991, 0.1424, 0.7229, 0.2248, 0.4367, 0.1982, 0.5435, 0.5702],
  8597. [0.5651, 0.3826, 0.8614, 0.5356, 0.4345, 0.5389, 0.5768, 0.5182],
  8598. [0.6056, 0.3988, 0.8797, 0.4604, 0.4088, 0.2651, 0.6134, 0.5259],
  8599. [0.6272, 0.4085, 0.8028, 0.2248, 0.4821, 0.1933, 0.6165, 0.5272],
  8600. [0.5926, 0.4070, 0.8533, 0.4843, 0.3864, 0.3605, 0.6208, 0.5338]],
  8601. device='cuda:0', grad_fn=<AddmmBackward>)
  8602. landmarks are: tensor([[[0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  8603. [0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  8604. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  8605. [0.0000, 0.0000, 0.6900, 0.1917, 0.3938, 0.2367, 0.5240, 0.5246],
  8606. [0.6069, 0.3975, 0.8625, 0.5083, 0.4387, 0.5483, 0.5650, 0.4967],
  8607. [0.6260, 0.4153, 0.9000, 0.4533, 0.4025, 0.2633, 0.6223, 0.4967],
  8608. [0.6276, 0.4095, 0.8238, 0.2250, 0.4663, 0.1783, 0.6171, 0.4869],
  8609. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309]]],
  8610. device='cuda:0')
  8611. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  8612. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  8613. loss_train: 0.1323865795857273
  8614. step: 71
  8615. running loss: 0.0018645997124750325
  8616. Train Steps: 71/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8617. torch.Size([8, 8])
  8618. tensor([[0.6219, 0.3934, 0.8688, 0.5267, 0.4313, 0.4967, 0.5988, 0.4983],
  8619. [0.6249, 0.4142, 0.8350, 0.3283, 0.3613, 0.3700, 0.6188, 0.5400],
  8620. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323],
  8621. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  8622. [0.6189, 0.4049, 0.8888, 0.4417, 0.4213, 0.5200, 0.5988, 0.5633],
  8623. [0.6162, 0.4134, 0.6700, 0.2467, 0.3962, 0.2533, 0.5737, 0.5467],
  8624. [0.6109, 0.4015, 0.7668, 0.3639, 0.3513, 0.3667, 0.5200, 0.5641],
  8625. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284]],
  8626. device='cuda:0', dtype=torch.float64)
  8627. predictions are: tensor([[0.5403, 0.3588, 0.8773, 0.5526, 0.4269, 0.4926, 0.5876, 0.5006],
  8628. [0.5450, 0.3859, 0.8281, 0.3307, 0.3718, 0.3166, 0.6227, 0.5600],
  8629. [0.6058, 0.4103, 0.8098, 0.2198, 0.4384, 0.1634, 0.6116, 0.5619],
  8630. [0.5277, 0.3507, 0.8685, 0.5997, 0.3702, 0.4894, 0.5950, 0.4821],
  8631. [0.5591, 0.3835, 0.9010, 0.4585, 0.4093, 0.5428, 0.6151, 0.5703],
  8632. [0.5242, 0.3608, 0.7226, 0.2736, 0.3663, 0.2387, 0.5913, 0.5539],
  8633. [0.5398, 0.3709, 0.8187, 0.3559, 0.3501, 0.3463, 0.5325, 0.5561],
  8634. [0.5712, 0.3901, 0.7794, 0.2210, 0.4508, 0.1871, 0.6098, 0.5435]],
  8635. device='cuda:0', grad_fn=<AddmmBackward>)
  8636. landmarks are: tensor([[[0.6219, 0.3934, 0.8687, 0.5267, 0.4313, 0.4967, 0.5987, 0.4983],
  8637. [0.6249, 0.4142, 0.8350, 0.3283, 0.3613, 0.3700, 0.6187, 0.5400],
  8638. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323],
  8639. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  8640. [0.6189, 0.4049, 0.8888, 0.4417, 0.4212, 0.5200, 0.5987, 0.5633],
  8641. [0.6162, 0.4134, 0.6700, 0.2467, 0.3963, 0.2533, 0.5738, 0.5467],
  8642. [0.6109, 0.4015, 0.7668, 0.3639, 0.3512, 0.3667, 0.5200, 0.5641],
  8643. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284]]],
  8644. device='cuda:0')
  8645. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  8646. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  8647. loss_train: 0.13352739083347842
  8648. step: 72
  8649. running loss: 0.0018545470949094226
  8650.  
  8651. Train Steps: 72/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8652. torch.Size([8, 8])
  8653. tensor([[0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  8654. [0.6200, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  8655. [0.6182, 0.4058, 0.8738, 0.4350, 0.3563, 0.3400, 0.5290, 0.5822],
  8656. [0.6114, 0.4018, 0.7213, 0.1967, 0.3763, 0.2700, 0.5875, 0.5533],
  8657. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  8658. [0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  8659. [0.6109, 0.4003, 0.8650, 0.4883, 0.4775, 0.4867, 0.5175, 0.5683],
  8660. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103]],
  8661. device='cuda:0', dtype=torch.float64)
  8662. predictions are: tensor([[0.5702, 0.3739, 0.7809, 0.2329, 0.4572, 0.1862, 0.6156, 0.5310],
  8663. [0.6047, 0.4196, 0.8026, 0.3138, 0.4052, 0.2717, 0.6298, 0.5326],
  8664. [0.5253, 0.3631, 0.8580, 0.4306, 0.3677, 0.3510, 0.5075, 0.5822],
  8665. [0.5306, 0.3709, 0.7307, 0.2197, 0.3811, 0.2467, 0.5953, 0.5587],
  8666. [0.6008, 0.4069, 0.7982, 0.4377, 0.3647, 0.4292, 0.5564, 0.5311],
  8667. [0.5620, 0.3893, 0.8881, 0.4307, 0.3675, 0.4510, 0.6316, 0.5068],
  8668. [0.4929, 0.3426, 0.8852, 0.4981, 0.4812, 0.4883, 0.5724, 0.5553],
  8669. [0.5472, 0.3816, 0.8672, 0.4665, 0.3976, 0.4679, 0.5754, 0.5143]],
  8670. device='cuda:0', grad_fn=<AddmmBackward>)
  8671. landmarks are: tensor([[[0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  8672. [0.6201, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  8673. [0.6182, 0.4058, 0.8737, 0.4350, 0.3562, 0.3400, 0.5290, 0.5822],
  8674. [0.6114, 0.4018, 0.7212, 0.1967, 0.3762, 0.2700, 0.5875, 0.5533],
  8675. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  8676. [0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  8677. [0.6109, 0.4003, 0.8650, 0.4883, 0.4775, 0.4867, 0.5175, 0.5683],
  8678. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103]]],
  8679. device='cuda:0')
  8680. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  8681. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  8682. loss_train: 0.13450522517086938
  8683. step: 73
  8684. running loss: 0.0018425373311077998
  8685. Train Steps: 73/90 Loss: 0.0018 torch.Size([8, 600, 800])
  8686. torch.Size([8, 8])
  8687. tensor([[0.6346, 0.4165, 0.9138, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471],
  8688. [0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  8689. [0.6152, 0.4131, 0.6863, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  8690. [ nan, nan, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  8691. [0.6261, 0.4045, 0.8865, 0.5369, 0.3895, 0.4859, 0.6683, 0.5249],
  8692. [0.6274, 0.4270, 0.8938, 0.4967, 0.3550, 0.4283, 0.5700, 0.5733],
  8693. [0.6307, 0.4029, 0.8650, 0.5200, 0.3763, 0.4017, 0.7311, 0.5366],
  8694. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901]],
  8695. device='cuda:0', dtype=torch.float64)
  8696. predictions are: tensor([[0.5932, 0.4021, 0.9224, 0.4201, 0.4089, 0.4184, 0.6681, 0.5319],
  8697. [0.6426, 0.4420, 0.9131, 0.5361, 0.3849, 0.4679, 0.5603, 0.5626],
  8698. [0.5596, 0.3877, 0.7069, 0.2391, 0.3801, 0.3218, 0.5559, 0.5568],
  8699. [0.3145, 0.2176, 0.6766, 0.2093, 0.3874, 0.2276, 0.5255, 0.5453],
  8700. [0.6297, 0.4334, 0.8823, 0.5122, 0.3979, 0.5125, 0.6349, 0.5204],
  8701. [0.6396, 0.4406, 0.8905, 0.4983, 0.3858, 0.4697, 0.5679, 0.5379],
  8702. [0.6697, 0.4546, 0.8755, 0.4976, 0.4065, 0.3941, 0.6260, 0.5066],
  8703. [0.5101, 0.3365, 0.7944, 0.2642, 0.3986, 0.2635, 0.5119, 0.5375]],
  8704. device='cuda:0', grad_fn=<AddmmBackward>)
  8705. landmarks are: tensor([[[0.6346, 0.4165, 0.9137, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471],
  8706. [0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  8707. [0.6152, 0.4131, 0.6862, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  8708. [0.0000, 0.0000, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  8709. [0.6261, 0.4045, 0.8865, 0.5369, 0.3895, 0.4859, 0.6683, 0.5249],
  8710. [0.6274, 0.4270, 0.8938, 0.4967, 0.3550, 0.4283, 0.5700, 0.5733],
  8711. [0.6307, 0.4029, 0.8650, 0.5200, 0.3762, 0.4017, 0.7311, 0.5366],
  8712. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901]]],
  8713. device='cuda:0')
  8714. loss_train_step before backward: tensor(0.0034, device='cuda:0', grad_fn=<MseLossBackward>)
  8715. loss_train_step after backward: tensor(0.0034, device='cuda:0', grad_fn=<MseLossBackward>)
  8716. loss_train: 0.13787507318193093
  8717. step: 74
  8718. running loss: 0.0018631766646206882
  8719. Train Steps: 74/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8720. torch.Size([8, 8])
  8721. tensor([[0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  8722. [0.6161, 0.4024, 0.8838, 0.4583, 0.3688, 0.3733, 0.5311, 0.5344],
  8723. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  8724. [0.6275, 0.4048, 0.8488, 0.2883, 0.4463, 0.2033, 0.6321, 0.5155],
  8725. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  8726. [0.6072, 0.4029, 0.7037, 0.2150, 0.3912, 0.2267, 0.5516, 0.5507],
  8727. [0.6048, 0.3928, 0.8538, 0.5433, 0.3875, 0.5117, 0.5266, 0.4719],
  8728. [ nan, nan, 0.6900, 0.1917, 0.3937, 0.2367, 0.5240, 0.5246]],
  8729. device='cuda:0', dtype=torch.float64)
  8730. predictions are: tensor([[0.5797, 0.3975, 0.7268, 0.2070, 0.3868, 0.2786, 0.5860, 0.5435],
  8731. [0.6807, 0.4556, 0.8901, 0.5002, 0.3596, 0.4272, 0.5110, 0.5236],
  8732. [0.5820, 0.3928, 0.8670, 0.3012, 0.4777, 0.2253, 0.6494, 0.5298],
  8733. [0.6709, 0.4309, 0.8455, 0.2722, 0.4336, 0.2452, 0.6256, 0.5038],
  8734. [0.7241, 0.4788, 0.8761, 0.5397, 0.3933, 0.6080, 0.6220, 0.5062],
  8735. [0.5731, 0.3881, 0.6840, 0.2172, 0.3951, 0.2544, 0.5523, 0.5419],
  8736. [0.6635, 0.4368, 0.8471, 0.5739, 0.3895, 0.5456, 0.5621, 0.5044],
  8737. [0.0911, 0.0716, 0.7280, 0.2380, 0.4164, 0.2594, 0.5111, 0.5556]],
  8738. device='cuda:0', grad_fn=<AddmmBackward>)
  8739. landmarks are: tensor([[[0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  8740. [0.6161, 0.4024, 0.8838, 0.4583, 0.3688, 0.3733, 0.5311, 0.5344],
  8741. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  8742. [0.6275, 0.4048, 0.8487, 0.2883, 0.4462, 0.2033, 0.6321, 0.5155],
  8743. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  8744. [0.6072, 0.4029, 0.7038, 0.2150, 0.3913, 0.2267, 0.5516, 0.5507],
  8745. [0.6048, 0.3928, 0.8537, 0.5433, 0.3875, 0.5117, 0.5266, 0.4719],
  8746. [0.0000, 0.0000, 0.6900, 0.1917, 0.3938, 0.2367, 0.5240, 0.5246]]],
  8747. device='cuda:0')
  8748. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  8749. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  8750. loss_train: 0.13911375048337504
  8751. step: 75
  8752. running loss: 0.0018548500064450005
  8753. Train Steps: 75/90 Loss: 0.0019 torch.Size([8, 600, 800])
  8754. torch.Size([8, 8])
  8755. tensor([[0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  8756. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  8757. [0.6072, 0.4029, 0.7037, 0.2150, 0.3912, 0.2267, 0.5516, 0.5507],
  8758. [0.6346, 0.4144, 0.9088, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  8759. [0.6201, 0.4082, 0.8827, 0.3715, 0.3825, 0.2712, 0.5845, 0.5412],
  8760. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  8761. [0.6115, 0.4005, 0.8838, 0.3867, 0.3763, 0.4700, 0.5800, 0.5550],
  8762. [0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090]],
  8763. device='cuda:0', dtype=torch.float64)
  8764. predictions are: tensor([[0.6325, 0.4128, 0.8378, 0.4977, 0.4163, 0.5059, 0.5350, 0.5198],
  8765. [0.4977, 0.3286, 0.8924, 0.4457, 0.3964, 0.3302, 0.6415, 0.5278],
  8766. [0.5240, 0.3414, 0.6580, 0.1917, 0.3814, 0.2373, 0.5462, 0.5343],
  8767. [0.6375, 0.4162, 0.8675, 0.4607, 0.3843, 0.4125, 0.6786, 0.5469],
  8768. [0.6242, 0.3876, 0.8634, 0.3314, 0.3710, 0.2785, 0.5657, 0.5215],
  8769. [0.5205, 0.3214, 0.7670, 0.2383, 0.4139, 0.2655, 0.5800, 0.5364],
  8770. [0.6382, 0.4089, 0.8369, 0.3992, 0.3415, 0.5122, 0.5060, 0.5114],
  8771. [0.6169, 0.4138, 0.8641, 0.3638, 0.3694, 0.2957, 0.6184, 0.4895]],
  8772. device='cuda:0', grad_fn=<AddmmBackward>)
  8773. landmarks are: tensor([[[0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  8774. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  8775. [0.6072, 0.4029, 0.7038, 0.2150, 0.3913, 0.2267, 0.5516, 0.5507],
  8776. [0.6346, 0.4144, 0.9087, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  8777. [0.6201, 0.4082, 0.8827, 0.3715, 0.3825, 0.2712, 0.5845, 0.5412],
  8778. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  8779. [0.6115, 0.4005, 0.8838, 0.3867, 0.3762, 0.4700, 0.5800, 0.5550],
  8780. [0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090]]],
  8781. device='cuda:0')
  8782. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  8783. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  8784. loss_train: 0.14055398636264727
  8785. step: 76
  8786. running loss: 0.0018493945574032534
  8787.  
  8788. Train Steps: 76/90 Loss: 0.0018 torch.Size([8, 600, 800])
  8789. torch.Size([8, 8])
  8790. tensor([[0.6197, 0.4091, 0.8800, 0.4783, 0.3538, 0.4767, 0.5950, 0.5550],
  8791. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  8792. [ nan, nan, 0.7625, 0.2433, 0.3713, 0.2867, 0.5235, 0.5220],
  8793. [0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717],
  8794. [0.6126, 0.4067, 0.8638, 0.5383, 0.4188, 0.4850, 0.5016, 0.5392],
  8795. [0.6257, 0.4034, 0.8287, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  8796. [0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5413, 0.5433],
  8797. [0.6132, 0.4118, 0.8200, 0.3633, 0.3563, 0.5400, 0.5787, 0.5136]],
  8798. device='cuda:0', dtype=torch.float64)
  8799. predictions are: tensor([[0.7221, 0.4383, 0.8886, 0.4777, 0.3893, 0.4451, 0.5912, 0.5377],
  8800. [0.5946, 0.3824, 0.8502, 0.2653, 0.4305, 0.2288, 0.7005, 0.5474],
  8801. [0.0674, 0.0377, 0.7684, 0.2331, 0.3859, 0.2675, 0.5374, 0.5340],
  8802. [0.6066, 0.3915, 0.7468, 0.2663, 0.3789, 0.2621, 0.5982, 0.5673],
  8803. [0.7450, 0.4842, 0.8662, 0.5883, 0.4239, 0.4681, 0.5471, 0.5095],
  8804. [0.6306, 0.4095, 0.8095, 0.2387, 0.3810, 0.2444, 0.6396, 0.4975],
  8805. [0.6211, 0.4149, 0.8376, 0.3492, 0.3540, 0.3035, 0.5393, 0.5472],
  8806. [0.6774, 0.4275, 0.8233, 0.3700, 0.3682, 0.5086, 0.5857, 0.5150]],
  8807. device='cuda:0', grad_fn=<AddmmBackward>)
  8808. landmarks are: tensor([[[0.6197, 0.4091, 0.8800, 0.4783, 0.3537, 0.4767, 0.5950, 0.5550],
  8809. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  8810. [0.0000, 0.0000, 0.7625, 0.2433, 0.3713, 0.2867, 0.5235, 0.5220],
  8811. [0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717],
  8812. [0.6126, 0.4067, 0.8637, 0.5383, 0.4187, 0.4850, 0.5016, 0.5392],
  8813. [0.6257, 0.4034, 0.8288, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  8814. [0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5412, 0.5433],
  8815. [0.6132, 0.4118, 0.8200, 0.3633, 0.3562, 0.5400, 0.5787, 0.5136]]],
  8816. device='cuda:0')
  8817. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  8818. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  8819. loss_train: 0.14160173939308152
  8820. step: 77
  8821. running loss: 0.0018389836284815782
  8822. Train Steps: 77/90 Loss: 0.0018 torch.Size([8, 600, 800])
  8823. torch.Size([8, 8])
  8824. tensor([[0.6264, 0.4049, 0.8988, 0.4633, 0.3813, 0.4983, 0.6326, 0.4843],
  8825. [0.6200, 0.3993, 0.8519, 0.4923, 0.3962, 0.4717, 0.6013, 0.5433],
  8826. [0.6222, 0.4169, 0.8638, 0.5650, 0.4313, 0.4783, 0.5637, 0.5633],
  8827. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  8828. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  8829. [0.6152, 0.4131, 0.6863, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  8830. [0.6053, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  8831. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5513, 0.5750]],
  8832. device='cuda:0', dtype=torch.float64)
  8833. predictions are: tensor([[0.6685, 0.4316, 0.9437, 0.4782, 0.3725, 0.5001, 0.6333, 0.4880],
  8834. [0.6572, 0.4154, 0.9135, 0.4934, 0.3772, 0.4707, 0.6351, 0.5345],
  8835. [0.6967, 0.4506, 0.9175, 0.5615, 0.4209, 0.4681, 0.6014, 0.5842],
  8836. [0.5686, 0.3843, 0.8098, 0.2375, 0.4232, 0.1728, 0.5963, 0.5342],
  8837. [0.6842, 0.4281, 0.8738, 0.3398, 0.3418, 0.4645, 0.6383, 0.5833],
  8838. [0.5826, 0.3776, 0.7209, 0.2505, 0.3553, 0.3277, 0.5802, 0.5649],
  8839. [0.4752, 0.2821, 0.7004, 0.1778, 0.3955, 0.1971, 0.5591, 0.5399],
  8840. [0.4212, 0.2661, 0.7238, 0.2077, 0.4278, 0.1912, 0.5486, 0.5777]],
  8841. device='cuda:0', grad_fn=<AddmmBackward>)
  8842. landmarks are: tensor([[[0.6264, 0.4049, 0.8988, 0.4633, 0.3812, 0.4983, 0.6326, 0.4843],
  8843. [0.6200, 0.3993, 0.8519, 0.4923, 0.3963, 0.4717, 0.6012, 0.5433],
  8844. [0.6222, 0.4169, 0.8637, 0.5650, 0.4313, 0.4783, 0.5638, 0.5633],
  8845. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  8846. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  8847. [0.6152, 0.4131, 0.6862, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  8848. [0.6054, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  8849. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5512, 0.5750]]],
  8850. device='cuda:0')
  8851. loss_train_step before backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  8852. loss_train_step after backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  8853. loss_train: 0.14373794820858166
  8854. step: 78
  8855. running loss: 0.001842794207802329
  8856. Train Steps: 78/90 Loss: 0.0018 torch.Size([8, 600, 800])
  8857. torch.Size([8, 8])
  8858. tensor([[0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717],
  8859. [0.6300, 0.4133, 0.8538, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  8860. [0.6226, 0.4098, 0.8912, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  8861. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  8862. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  8863. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  8864. [0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  8865. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131]],
  8866. device='cuda:0', dtype=torch.float64)
  8867. predictions are: tensor([[0.6005, 0.3883, 0.7376, 0.2639, 0.3651, 0.2618, 0.5934, 0.5770],
  8868. [0.5534, 0.3800, 0.8344, 0.2513, 0.5063, 0.2552, 0.7289, 0.5471],
  8869. [0.6056, 0.3878, 0.8910, 0.4077, 0.4137, 0.2377, 0.5669, 0.5622],
  8870. [0.6107, 0.3752, 0.8548, 0.4358, 0.3673, 0.5117, 0.5738, 0.5285],
  8871. [0.5490, 0.3539, 0.8691, 0.4588, 0.4263, 0.5274, 0.5895, 0.5428],
  8872. [0.5000, 0.3380, 0.6937, 0.2551, 0.4053, 0.2099, 0.5916, 0.5737],
  8873. [0.5854, 0.3727, 0.8872, 0.4041, 0.3371, 0.4301, 0.6151, 0.5077],
  8874. [0.6438, 0.4108, 0.8618, 0.3766, 0.3392, 0.3494, 0.5652, 0.5354]],
  8875. device='cuda:0', grad_fn=<AddmmBackward>)
  8876. landmarks are: tensor([[[0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717],
  8877. [0.6300, 0.4133, 0.8537, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  8878. [0.6226, 0.4098, 0.8913, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  8879. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  8880. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  8881. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  8882. [0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  8883. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131]]],
  8884. device='cuda:0')
  8885. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  8886. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  8887. loss_train: 0.14473954314598814
  8888. step: 79
  8889. running loss: 0.0018321461157720018
  8890. Train Steps: 79/90 Loss: 0.0018 torch.Size([8, 600, 800])
  8891. torch.Size([8, 8])
  8892. tensor([[0.6100, 0.4016, 0.8600, 0.5067, 0.4612, 0.5233, 0.5086, 0.5519],
  8893. [0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6262, 0.5167],
  8894. [0.6132, 0.4066, 0.7259, 0.2402, 0.3588, 0.3300, 0.6000, 0.5600],
  8895. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  8896. [0.6200, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  8897. [0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6088, 0.5583],
  8898. [0.6168, 0.4029, 0.8523, 0.3417, 0.3588, 0.5000, 0.6125, 0.5400],
  8899. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5837, 0.5500]],
  8900. device='cuda:0', dtype=torch.float64)
  8901. predictions are: tensor([[0.5663, 0.3640, 0.8650, 0.4958, 0.4625, 0.4853, 0.5205, 0.5457],
  8902. [0.6266, 0.4113, 0.8763, 0.3917, 0.3851, 0.3266, 0.6356, 0.5400],
  8903. [0.6017, 0.4096, 0.7041, 0.2273, 0.3641, 0.3217, 0.6051, 0.5864],
  8904. [0.6583, 0.4444, 0.8952, 0.3599, 0.4401, 0.2048, 0.6525, 0.5195],
  8905. [0.5728, 0.3495, 0.8861, 0.4576, 0.3689, 0.3999, 0.5798, 0.5363],
  8906. [0.6590, 0.4320, 0.6928, 0.2192, 0.3806, 0.2505, 0.6148, 0.5537],
  8907. [0.6123, 0.3829, 0.8421, 0.3424, 0.3509, 0.4881, 0.6346, 0.5532],
  8908. [0.5733, 0.3608, 0.9131, 0.4369, 0.4142, 0.5019, 0.5784, 0.5521]],
  8909. device='cuda:0', grad_fn=<AddmmBackward>)
  8910. landmarks are: tensor([[[0.6100, 0.4016, 0.8600, 0.5067, 0.4613, 0.5233, 0.5086, 0.5519],
  8911. [0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6263, 0.5167],
  8912. [0.6132, 0.4066, 0.7259, 0.2402, 0.3587, 0.3300, 0.6000, 0.5600],
  8913. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  8914. [0.6201, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  8915. [0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6087, 0.5583],
  8916. [0.6168, 0.4029, 0.8523, 0.3417, 0.3587, 0.5000, 0.6125, 0.5400],
  8917. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5838, 0.5500]]],
  8918. device='cuda:0')
  8919. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  8920. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  8921. loss_train: 0.14520067651756108
  8922. step: 80
  8923. running loss: 0.0018150084564695136
  8924.  
  8925. Train Steps: 80/90 Loss: 0.0018 torch.Size([8, 600, 800])
  8926. torch.Size([8, 8])
  8927. tensor([[0.6272, 0.4071, 0.8738, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  8928. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  8929. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  8930. [0.6268, 0.4102, 0.8938, 0.3667, 0.4025, 0.2833, 0.6275, 0.5183],
  8931. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  8932. [0.6128, 0.4118, 0.8638, 0.5333, 0.4625, 0.5267, 0.5193, 0.5475],
  8933. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  8934. [0.6299, 0.4008, 0.8450, 0.5350, 0.4213, 0.5000, 0.6350, 0.5100]],
  8935. device='cuda:0', dtype=torch.float64)
  8936. predictions are: tensor([[0.7075, 0.4653, 0.8551, 0.4899, 0.3809, 0.3419, 0.6343, 0.4962],
  8937. [0.7156, 0.4710, 0.7829, 0.2770, 0.3385, 0.3715, 0.6403, 0.5497],
  8938. [0.6334, 0.4071, 0.8531, 0.3993, 0.4389, 0.4503, 0.5759, 0.5201],
  8939. [0.6304, 0.4207, 0.8709, 0.3307, 0.4067, 0.2475, 0.6356, 0.5532],
  8940. [0.5920, 0.3828, 0.8642, 0.4004, 0.4326, 0.5304, 0.6126, 0.5525],
  8941. [0.6249, 0.4180, 0.8541, 0.4851, 0.4458, 0.4441, 0.5536, 0.5530],
  8942. [0.4778, 0.3300, 0.8495, 0.3993, 0.4294, 0.5057, 0.6012, 0.5509],
  8943. [0.5924, 0.4014, 0.8258, 0.4693, 0.4127, 0.4612, 0.6556, 0.5125]],
  8944. device='cuda:0', grad_fn=<AddmmBackward>)
  8945. landmarks are: tensor([[[0.6272, 0.4071, 0.8737, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  8946. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  8947. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  8948. [0.6268, 0.4102, 0.8938, 0.3667, 0.4025, 0.2833, 0.6275, 0.5183],
  8949. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  8950. [0.6128, 0.4118, 0.8637, 0.5333, 0.4625, 0.5267, 0.5193, 0.5475],
  8951. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  8952. [0.6299, 0.4008, 0.8450, 0.5350, 0.4212, 0.5000, 0.6350, 0.5100]]],
  8953. device='cuda:0')
  8954. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  8955. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  8956. loss_train: 0.1471182822715491
  8957. step: 81
  8958. running loss: 0.0018162750897722112
  8959. Train Steps: 81/90 Loss: 0.0018 torch.Size([8, 600, 800])
  8960. torch.Size([8, 8])
  8961. tensor([[0.6117, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103],
  8962. [0.6161, 0.4024, 0.8838, 0.4583, 0.3688, 0.3733, 0.5311, 0.5344],
  8963. [ nan, nan, 0.7412, 0.2200, 0.4450, 0.1517, 0.5312, 0.4983],
  8964. [0.6053, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  8965. [0.6165, 0.4106, 0.7575, 0.1733, 0.3838, 0.2650, 0.5680, 0.5116],
  8966. [0.6203, 0.4096, 0.8862, 0.4267, 0.3538, 0.4117, 0.6025, 0.5650],
  8967. [0.6166, 0.4008, 0.8563, 0.5667, 0.4388, 0.4933, 0.5575, 0.5567],
  8968. [0.6179, 0.4118, 0.7278, 0.4237, 0.3588, 0.3400, 0.5675, 0.5917]],
  8969. device='cuda:0', dtype=torch.float64)
  8970. predictions are: tensor([[0.6829, 0.4525, 0.6718, 0.2448, 0.4178, 0.2707, 0.5433, 0.5212],
  8971. [0.6605, 0.4421, 0.9349, 0.4715, 0.4095, 0.4238, 0.5632, 0.5242],
  8972. [0.1362, 0.0846, 0.7230, 0.2268, 0.4773, 0.2087, 0.5771, 0.5049],
  8973. [0.7012, 0.4477, 0.6752, 0.2197, 0.4411, 0.2342, 0.5737, 0.5244],
  8974. [0.7194, 0.4713, 0.7563, 0.2293, 0.4259, 0.2597, 0.6057, 0.4829],
  8975. [0.5869, 0.3943, 0.9245, 0.4066, 0.3863, 0.4469, 0.6174, 0.5767],
  8976. [0.6784, 0.4460, 0.8884, 0.5537, 0.4868, 0.5206, 0.5965, 0.5378],
  8977. [0.6576, 0.4474, 0.7983, 0.4115, 0.3844, 0.3781, 0.5655, 0.5796]],
  8978. device='cuda:0', grad_fn=<AddmmBackward>)
  8979. landmarks are: tensor([[[0.6116, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103],
  8980. [0.6161, 0.4024, 0.8838, 0.4583, 0.3688, 0.3733, 0.5311, 0.5344],
  8981. [0.0000, 0.0000, 0.7412, 0.2200, 0.4450, 0.1517, 0.5312, 0.4983],
  8982. [0.6054, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  8983. [0.6165, 0.4106, 0.7575, 0.1733, 0.3837, 0.2650, 0.5680, 0.5116],
  8984. [0.6203, 0.4096, 0.8863, 0.4267, 0.3537, 0.4117, 0.6025, 0.5650],
  8985. [0.6166, 0.4008, 0.8562, 0.5667, 0.4387, 0.4933, 0.5575, 0.5567],
  8986. [0.6179, 0.4118, 0.7278, 0.4237, 0.3587, 0.3400, 0.5675, 0.5917]]],
  8987. device='cuda:0')
  8988. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  8989. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  8990. loss_train: 0.1490206648595631
  8991. step: 82
  8992. running loss: 0.0018173251812141843
  8993. Train Steps: 82/90 Loss: 0.0018 torch.Size([8, 600, 800])
  8994. torch.Size([8, 8])
  8995. tensor([[0.6222, 0.4169, 0.8638, 0.5650, 0.4313, 0.4783, 0.5637, 0.5633],
  8996. [0.6198, 0.4115, 0.7762, 0.2717, 0.3713, 0.3200, 0.5837, 0.5683],
  8997. [0.6198, 0.4101, 0.8838, 0.5283, 0.3763, 0.5267, 0.5913, 0.5567],
  8998. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  8999. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  9000. [0.6339, 0.4102, 0.9088, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  9001. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5413, 0.5717],
  9002. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250]],
  9003. device='cuda:0', dtype=torch.float64)
  9004. predictions are: tensor([[0.6113, 0.4148, 0.8368, 0.5495, 0.4525, 0.4887, 0.5821, 0.5630],
  9005. [0.6829, 0.4633, 0.7749, 0.2959, 0.3903, 0.3291, 0.5708, 0.5146],
  9006. [0.6550, 0.4410, 0.8355, 0.5153, 0.4186, 0.5387, 0.5706, 0.5386],
  9007. [0.6351, 0.4176, 0.8367, 0.4638, 0.4074, 0.4470, 0.5879, 0.5185],
  9008. [0.5865, 0.3998, 0.8404, 0.4334, 0.4042, 0.4874, 0.5415, 0.4974],
  9009. [0.6618, 0.4470, 0.8834, 0.4661, 0.4353, 0.5668, 0.7026, 0.5230],
  9010. [0.6311, 0.4185, 0.8328, 0.4574, 0.4471, 0.4798, 0.5398, 0.5308],
  9011. [0.6300, 0.4266, 0.8661, 0.3974, 0.4017, 0.3521, 0.6201, 0.4980]],
  9012. device='cuda:0', grad_fn=<AddmmBackward>)
  9013. landmarks are: tensor([[[0.6222, 0.4169, 0.8637, 0.5650, 0.4313, 0.4783, 0.5638, 0.5633],
  9014. [0.6198, 0.4115, 0.7763, 0.2717, 0.3713, 0.3200, 0.5838, 0.5683],
  9015. [0.6198, 0.4101, 0.8838, 0.5283, 0.3762, 0.5267, 0.5913, 0.5567],
  9016. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  9017. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  9018. [0.6339, 0.4102, 0.9087, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  9019. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5412, 0.5717],
  9020. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250]]],
  9021. device='cuda:0')
  9022. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  9023. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  9024. loss_train: 0.14975211105775088
  9025. step: 83
  9026. running loss: 0.001804242301900613
  9027. Train Steps: 83/90 Loss: 0.0018 torch.Size([8, 600, 800])
  9028. torch.Size([8, 8])
  9029. tensor([[0.6143, 0.4040, 0.8237, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  9030. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  9031. [0.6114, 0.4018, 0.7213, 0.1967, 0.3763, 0.2700, 0.5875, 0.5533],
  9032. [0.6198, 0.4164, 0.8700, 0.5067, 0.4625, 0.5650, 0.5464, 0.5197],
  9033. [0.6075, 0.4000, 0.8513, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  9034. [0.6127, 0.4084, 0.8700, 0.4467, 0.3987, 0.4317, 0.5013, 0.5471],
  9035. [0.6026, 0.3979, 0.8550, 0.4233, 0.3613, 0.5233, 0.5582, 0.4967],
  9036. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667]],
  9037. device='cuda:0', dtype=torch.float64)
  9038. predictions are: tensor([[0.6903, 0.4541, 0.8041, 0.3519, 0.4225, 0.2267, 0.5705, 0.4859],
  9039. [0.5929, 0.3934, 0.8567, 0.5297, 0.4653, 0.5290, 0.6259, 0.4991],
  9040. [0.6601, 0.4450, 0.6932, 0.2327, 0.3852, 0.2471, 0.6020, 0.5289],
  9041. [0.5416, 0.3797, 0.8674, 0.5292, 0.4690, 0.5223, 0.5841, 0.5070],
  9042. [0.5711, 0.3844, 0.8338, 0.5346, 0.4663, 0.5404, 0.5545, 0.5090],
  9043. [0.6060, 0.4009, 0.8575, 0.4630, 0.3739, 0.4338, 0.4998, 0.5272],
  9044. [0.6225, 0.3925, 0.8570, 0.4345, 0.3532, 0.4917, 0.6009, 0.5021],
  9045. [0.6001, 0.4009, 0.8238, 0.3495, 0.3516, 0.3660, 0.5789, 0.5336]],
  9046. device='cuda:0', grad_fn=<AddmmBackward>)
  9047. landmarks are: tensor([[[0.6143, 0.4040, 0.8238, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  9048. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  9049. [0.6114, 0.4018, 0.7212, 0.1967, 0.3762, 0.2700, 0.5875, 0.5533],
  9050. [0.6198, 0.4164, 0.8700, 0.5067, 0.4625, 0.5650, 0.5464, 0.5197],
  9051. [0.6075, 0.4000, 0.8512, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  9052. [0.6127, 0.4084, 0.8700, 0.4467, 0.3988, 0.4317, 0.5013, 0.5471],
  9053. [0.6026, 0.3979, 0.8550, 0.4233, 0.3613, 0.5233, 0.5582, 0.4967],
  9054. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667]]],
  9055. device='cuda:0')
  9056. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  9057. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  9058. loss_train: 0.15045929490588605
  9059. step: 84
  9060. running loss: 0.0017911820822129293
  9061.  
  9062. Train Steps: 84/90 Loss: 0.0018 torch.Size([8, 600, 800])
  9063. torch.Size([8, 8])
  9064. tensor([[0.6197, 0.4091, 0.8800, 0.4783, 0.3538, 0.4767, 0.5950, 0.5550],
  9065. [0.6072, 0.4029, 0.7037, 0.2150, 0.3912, 0.2267, 0.5516, 0.5507],
  9066. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5463, 0.5800],
  9067. [0.6236, 0.4084, 0.7738, 0.2133, 0.3663, 0.3233, 0.5813, 0.5567],
  9068. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  9069. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896],
  9070. [0.6234, 0.4179, 0.7825, 0.3450, 0.3813, 0.2867, 0.5675, 0.5617],
  9071. [0.6212, 0.4159, 0.8675, 0.5783, 0.4088, 0.4317, 0.5613, 0.5917]],
  9072. device='cuda:0', dtype=torch.float64)
  9073. predictions are: tensor([[0.5609, 0.3478, 0.9110, 0.5086, 0.3812, 0.5075, 0.5569, 0.5241],
  9074. [0.6739, 0.4449, 0.7094, 0.2639, 0.4078, 0.2612, 0.5407, 0.5280],
  9075. [0.5018, 0.3333, 0.7566, 0.3187, 0.3691, 0.3055, 0.4783, 0.5424],
  9076. [0.6060, 0.4071, 0.7797, 0.2923, 0.3817, 0.3736, 0.5510, 0.5038],
  9077. [0.6098, 0.4134, 0.8727, 0.2888, 0.5322, 0.2752, 0.7309, 0.5040],
  9078. [0.5934, 0.3906, 0.8474, 0.4316, 0.3445, 0.3416, 0.5021, 0.5575],
  9079. [0.5262, 0.3429, 0.8057, 0.3826, 0.3886, 0.3141, 0.5591, 0.5266],
  9080. [0.5721, 0.3836, 0.8637, 0.6086, 0.4031, 0.4786, 0.5398, 0.5554]],
  9081. device='cuda:0', grad_fn=<AddmmBackward>)
  9082. landmarks are: tensor([[[0.6197, 0.4091, 0.8800, 0.4783, 0.3537, 0.4767, 0.5950, 0.5550],
  9083. [0.6072, 0.4029, 0.7038, 0.2150, 0.3913, 0.2267, 0.5516, 0.5507],
  9084. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5462, 0.5800],
  9085. [0.6236, 0.4084, 0.7738, 0.2133, 0.3663, 0.3233, 0.5813, 0.5567],
  9086. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  9087. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896],
  9088. [0.6234, 0.4179, 0.7825, 0.3450, 0.3812, 0.2867, 0.5675, 0.5617],
  9089. [0.6212, 0.4159, 0.8675, 0.5783, 0.4087, 0.4317, 0.5612, 0.5917]]],
  9090. device='cuda:0')
  9091. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  9092. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  9093. loss_train: 0.15215556626208127
  9094. step: 85
  9095. running loss: 0.0017900654854362502
  9096. Train Steps: 85/90 Loss: 0.0018 torch.Size([8, 600, 800])
  9097. torch.Size([8, 8])
  9098. tensor([[0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  9099. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  9100. [0.6230, 0.4152, 0.7588, 0.2283, 0.4012, 0.2883, 0.6200, 0.5767],
  9101. [0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6262, 0.5167],
  9102. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6188, 0.5283],
  9103. [ nan, nan, 0.7512, 0.2117, 0.4288, 0.2000, 0.5600, 0.5617],
  9104. [ nan, nan, 0.7192, 0.2346, 0.4037, 0.2050, 0.5138, 0.5650],
  9105. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892]],
  9106. device='cuda:0', dtype=torch.float64)
  9107. predictions are: tensor([[0.6643, 0.4457, 0.8618, 0.2849, 0.5180, 0.2876, 0.6996, 0.5425],
  9108. [0.7646, 0.5065, 0.7449, 0.2939, 0.4197, 0.1968, 0.5505, 0.5345],
  9109. [0.6854, 0.4418, 0.7921, 0.3072, 0.3834, 0.3308, 0.5712, 0.5596],
  9110. [0.6527, 0.4247, 0.8917, 0.4491, 0.3591, 0.4004, 0.5751, 0.5346],
  9111. [0.6775, 0.4342, 0.9259, 0.4055, 0.3780, 0.2888, 0.5830, 0.5323],
  9112. [0.2793, 0.1801, 0.7538, 0.2756, 0.4184, 0.2544, 0.5126, 0.5533],
  9113. [0.0762, 0.0398, 0.7043, 0.2500, 0.3781, 0.2918, 0.4472, 0.5579],
  9114. [0.6697, 0.4160, 0.8804, 0.4967, 0.3482, 0.4281, 0.4500, 0.5749]],
  9115. device='cuda:0', grad_fn=<AddmmBackward>)
  9116. landmarks are: tensor([[[0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  9117. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  9118. [0.6230, 0.4152, 0.7588, 0.2283, 0.4013, 0.2883, 0.6200, 0.5767],
  9119. [0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6263, 0.5167],
  9120. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6187, 0.5283],
  9121. [0.0000, 0.0000, 0.7513, 0.2117, 0.4288, 0.2000, 0.5600, 0.5617],
  9122. [0.0000, 0.0000, 0.7192, 0.2346, 0.4038, 0.2050, 0.5138, 0.5650],
  9123. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892]]],
  9124. device='cuda:0')
  9125. loss_train_step before backward: tensor(0.0037, device='cuda:0', grad_fn=<MseLossBackward>)
  9126. loss_train_step after backward: tensor(0.0037, device='cuda:0', grad_fn=<MseLossBackward>)
  9127. loss_train: 0.15586218307726085
  9128. step: 86
  9129. running loss: 0.0018123509660146612
  9130. Train Steps: 86/90 Loss: 0.0018 torch.Size([8, 600, 800])
  9131. torch.Size([8, 8])
  9132. tensor([[0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  9133. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  9134. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  9135. [0.6151, 0.4125, 0.8738, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  9136. [0.6120, 0.4014, 0.6863, 0.2817, 0.3700, 0.2783, 0.5513, 0.5667],
  9137. [0.6185, 0.4098, 0.8838, 0.4900, 0.4537, 0.5800, 0.6288, 0.5400],
  9138. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6138, 0.5333],
  9139. [0.6083, 0.3957, 0.8638, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980]],
  9140. device='cuda:0', dtype=torch.float64)
  9141. predictions are: tensor([[0.5563, 0.3623, 0.9090, 0.4497, 0.4262, 0.5491, 0.6082, 0.5700],
  9142. [0.6014, 0.4059, 0.8575, 0.3241, 0.3529, 0.2868, 0.5669, 0.5697],
  9143. [0.6375, 0.4184, 0.8368, 0.5489, 0.3802, 0.4560, 0.5802, 0.6651],
  9144. [0.6058, 0.3839, 0.8531, 0.4630, 0.3414, 0.3593, 0.4970, 0.5782],
  9145. [0.5963, 0.3940, 0.7217, 0.2645, 0.3463, 0.2958, 0.5310, 0.5844],
  9146. [0.6387, 0.3952, 0.9072, 0.4854, 0.4515, 0.5408, 0.5929, 0.5492],
  9147. [0.6590, 0.4444, 0.8863, 0.4673, 0.3424, 0.4635, 0.6213, 0.5645],
  9148. [0.5478, 0.3320, 0.8871, 0.4691, 0.4188, 0.4898, 0.5487, 0.5225]],
  9149. device='cuda:0', grad_fn=<AddmmBackward>)
  9150. landmarks are: tensor([[[0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  9151. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  9152. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  9153. [0.6151, 0.4125, 0.8737, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  9154. [0.6120, 0.4013, 0.6862, 0.2817, 0.3700, 0.2783, 0.5512, 0.5667],
  9155. [0.6185, 0.4098, 0.8838, 0.4900, 0.4538, 0.5800, 0.6288, 0.5400],
  9156. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6137, 0.5333],
  9157. [0.6083, 0.3957, 0.8637, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980]]],
  9158. device='cuda:0')
  9159. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  9160. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  9161. loss_train: 0.15654120134422556
  9162. step: 87
  9163. running loss: 0.001799324153381903
  9164. Train Steps: 87/90 Loss: 0.0018 torch.Size([8, 600, 800])
  9165. torch.Size([8, 8])
  9166. tensor([[0.6357, 0.4118, 0.8400, 0.2500, 0.5413, 0.1633, 0.6725, 0.5586],
  9167. [0.6236, 0.3967, 0.8675, 0.5400, 0.3862, 0.4517, 0.5825, 0.5200],
  9168. [0.6113, 0.4006, 0.8700, 0.5350, 0.3638, 0.3767, 0.5097, 0.4882],
  9169. [0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  9170. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  9171. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  9172. [0.6199, 0.4102, 0.8950, 0.4417, 0.4012, 0.5367, 0.6112, 0.5967],
  9173. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050]],
  9174. device='cuda:0', dtype=torch.float64)
  9175. predictions are: tensor([[0.5790, 0.3507, 0.8852, 0.2404, 0.5145, 0.1579, 0.6487, 0.5750],
  9176. [0.5448, 0.3536, 0.8686, 0.5279, 0.3656, 0.4509, 0.5883, 0.5766],
  9177. [0.6526, 0.4043, 0.8698, 0.4957, 0.3553, 0.3628, 0.5310, 0.5696],
  9178. [0.5740, 0.3852, 0.8894, 0.4132, 0.3652, 0.5292, 0.5823, 0.5577],
  9179. [0.5861, 0.3909, 0.7450, 0.2332, 0.3718, 0.2609, 0.5354, 0.5544],
  9180. [0.5748, 0.3649, 0.8456, 0.4180, 0.3466, 0.3991, 0.5565, 0.6004],
  9181. [0.6497, 0.3969, 0.8797, 0.4306, 0.3741, 0.5402, 0.5959, 0.6054],
  9182. [0.6204, 0.3975, 0.8895, 0.4839, 0.4497, 0.4924, 0.5877, 0.5582]],
  9183. device='cuda:0', grad_fn=<AddmmBackward>)
  9184. landmarks are: tensor([[[0.6357, 0.4118, 0.8400, 0.2500, 0.5412, 0.1633, 0.6725, 0.5586],
  9185. [0.6236, 0.3967, 0.8675, 0.5400, 0.3862, 0.4517, 0.5825, 0.5200],
  9186. [0.6113, 0.4006, 0.8700, 0.5350, 0.3638, 0.3767, 0.5097, 0.4882],
  9187. [0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  9188. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  9189. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  9190. [0.6199, 0.4102, 0.8950, 0.4417, 0.4013, 0.5367, 0.6112, 0.5967],
  9191. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050]]],
  9192. device='cuda:0')
  9193. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  9194. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  9195. loss_train: 0.1574872225173749
  9196. step: 88
  9197. running loss: 0.001789627528606533
  9198.  
  9199. Train Steps: 88/90 Loss: 0.0018 torch.Size([8, 600, 800])
  9200. torch.Size([8, 8])
  9201. tensor([[0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683],
  9202. [0.6201, 0.3970, 0.8413, 0.4950, 0.4413, 0.5183, 0.6088, 0.5400],
  9203. [0.6163, 0.4006, 0.8788, 0.4683, 0.3663, 0.4883, 0.5887, 0.5017],
  9204. [0.6257, 0.4024, 0.8672, 0.5422, 0.4196, 0.5198, 0.6694, 0.5204],
  9205. [0.6140, 0.4070, 0.8700, 0.5000, 0.4612, 0.4900, 0.5260, 0.5852],
  9206. [0.6218, 0.4098, 0.7238, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  9207. [0.6267, 0.4080, 0.8438, 0.2633, 0.4763, 0.1800, 0.6259, 0.5240],
  9208. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895]],
  9209. device='cuda:0', dtype=torch.float64)
  9210. predictions are: tensor([[0.6563, 0.4143, 0.8711, 0.5141, 0.3567, 0.3754, 0.5637, 0.5939],
  9211. [0.4888, 0.3187, 0.8533, 0.4704, 0.4120, 0.5141, 0.5794, 0.5805],
  9212. [0.5305, 0.3432, 0.8643, 0.4474, 0.3680, 0.4902, 0.5796, 0.5600],
  9213. [0.5621, 0.3650, 0.8665, 0.5146, 0.3924, 0.5260, 0.6470, 0.5585],
  9214. [0.5229, 0.3579, 0.8662, 0.4833, 0.4403, 0.4909, 0.5109, 0.5972],
  9215. [0.6561, 0.4365, 0.7485, 0.2128, 0.4266, 0.2200, 0.5967, 0.5774],
  9216. [0.6677, 0.4292, 0.8820, 0.2452, 0.4535, 0.1810, 0.6232, 0.5437],
  9217. [0.5569, 0.3679, 0.7763, 0.1875, 0.4438, 0.1674, 0.5821, 0.5393]],
  9218. device='cuda:0', grad_fn=<AddmmBackward>)
  9219. landmarks are: tensor([[[0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683],
  9220. [0.6201, 0.3970, 0.8413, 0.4950, 0.4412, 0.5183, 0.6087, 0.5400],
  9221. [0.6163, 0.4006, 0.8788, 0.4683, 0.3663, 0.4883, 0.5888, 0.5017],
  9222. [0.6257, 0.4024, 0.8672, 0.5422, 0.4196, 0.5198, 0.6694, 0.5204],
  9223. [0.6140, 0.4070, 0.8700, 0.5000, 0.4613, 0.4900, 0.5260, 0.5852],
  9224. [0.6218, 0.4098, 0.7237, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  9225. [0.6267, 0.4080, 0.8438, 0.2633, 0.4762, 0.1800, 0.6259, 0.5240],
  9226. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895]]],
  9227. device='cuda:0')
  9228. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  9229. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  9230. loss_train: 0.15884605079190806
  9231. step: 89
  9232. running loss: 0.0017847870875495287
  9233. Train Steps: 89/90 Loss: 0.0018 torch.Size([8, 600, 800])
  9234. torch.Size([8, 8])
  9235. tensor([[0.6248, 0.4032, 0.7738, 0.1900, 0.4813, 0.1400, 0.5941, 0.4904],
  9236. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  9237. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683],
  9238. [0.6171, 0.4127, 0.8900, 0.4800, 0.4325, 0.5783, 0.5769, 0.5090],
  9239. [0.6284, 0.4093, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  9240. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  9241. [ nan, nan, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  9242. [0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364]],
  9243. device='cuda:0', dtype=torch.float64)
  9244. predictions are: tensor([[0.5701, 0.3899, 0.7749, 0.2235, 0.4891, 0.1285, 0.6125, 0.5358],
  9245. [0.6624, 0.4344, 0.8536, 0.4378, 0.3787, 0.3752, 0.5085, 0.5879],
  9246. [0.6781, 0.4539, 0.8667, 0.5125, 0.3853, 0.3703, 0.5882, 0.5832],
  9247. [0.6412, 0.4408, 0.9126, 0.4809, 0.4415, 0.5653, 0.5623, 0.5423],
  9248. [0.6393, 0.4321, 0.8859, 0.4558, 0.3700, 0.3787, 0.6376, 0.5289],
  9249. [0.5884, 0.4076, 0.9018, 0.4764, 0.4055, 0.5284, 0.6674, 0.5022],
  9250. [0.1344, 0.0866, 0.6973, 0.2091, 0.4333, 0.1891, 0.5127, 0.5675],
  9251. [0.6251, 0.4193, 0.7968, 0.2227, 0.4558, 0.2134, 0.6566, 0.5755]],
  9252. device='cuda:0', grad_fn=<AddmmBackward>)
  9253. landmarks are: tensor([[[0.6248, 0.4032, 0.7738, 0.1900, 0.4812, 0.1400, 0.5941, 0.4904],
  9254. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  9255. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683],
  9256. [0.6171, 0.4127, 0.8900, 0.4800, 0.4325, 0.5783, 0.5769, 0.5090],
  9257. [0.6284, 0.4092, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  9258. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  9259. [0.0000, 0.0000, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  9260. [0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364]]],
  9261. device='cuda:0')
  9262. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  9263. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  9264. loss_train: 0.1597334560356103
  9265. step: 90
  9266. running loss: 0.0017748161781734476
  9267. Valid Steps: 10/10 Loss: nan 1.0033
  9268. --------------------------------------------------
  9269. Epoch: 3 Train Loss: 0.0018 Valid Loss: nan
  9270. --------------------------------------------------
  9271. size of train loader is: 90
  9272. torch.Size([8, 600, 800])
  9273. torch.Size([8, 8])
  9274. tensor([[0.6069, 0.3975, 0.8625, 0.5083, 0.4388, 0.5483, 0.5650, 0.4967],
  9275. [0.6164, 0.4119, 0.7913, 0.2650, 0.3538, 0.3500, 0.5614, 0.5038],
  9276. [0.6277, 0.4057, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  9277. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  9278. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6063, 0.6083],
  9279. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  9280. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  9281. [0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726]],
  9282. device='cuda:0', dtype=torch.float64)
  9283. predictions are: tensor([[0.5968, 0.4003, 0.8611, 0.5263, 0.4863, 0.5078, 0.5804, 0.5099],
  9284. [0.5379, 0.3639, 0.7909, 0.2545, 0.3858, 0.3378, 0.6020, 0.4907],
  9285. [0.5206, 0.3413, 0.8314, 0.2572, 0.4827, 0.1848, 0.6728, 0.4873],
  9286. [0.6144, 0.4080, 0.8864, 0.5306, 0.4392, 0.4873, 0.6470, 0.5047],
  9287. [0.6072, 0.4264, 0.8403, 0.3939, 0.3851, 0.4020, 0.6346, 0.5666],
  9288. [0.5697, 0.3890, 0.7807, 0.2813, 0.4378, 0.2358, 0.5794, 0.5491],
  9289. [0.6462, 0.4159, 0.8569, 0.4667, 0.4047, 0.3509, 0.5302, 0.5521],
  9290. [0.5676, 0.3707, 0.7575, 0.2766, 0.4180, 0.2593, 0.5911, 0.4781]],
  9291. device='cuda:0', grad_fn=<AddmmBackward>)
  9292. landmarks are: tensor([[[0.6069, 0.3975, 0.8625, 0.5083, 0.4387, 0.5483, 0.5650, 0.4967],
  9293. [0.6164, 0.4119, 0.7912, 0.2650, 0.3537, 0.3500, 0.5614, 0.5038],
  9294. [0.6277, 0.4056, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  9295. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  9296. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6062, 0.6083],
  9297. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  9298. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  9299. [0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726]]],
  9300. device='cuda:0')
  9301. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  9302. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  9303. loss_train: 0.0011329366825520992
  9304. step: 1
  9305. running loss: 0.0011329366825520992
  9306. Train Steps: 1/90 Loss: 0.0011 torch.Size([8, 600, 800])
  9307. torch.Size([8, 8])
  9308. tensor([[0.6203, 0.4021, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  9309. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  9310. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  9311. [0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500],
  9312. [0.6214, 0.4112, 0.7838, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  9313. [0.6200, 0.4098, 0.8237, 0.2917, 0.4012, 0.2967, 0.6000, 0.5683],
  9314. [0.6289, 0.4081, 0.8720, 0.3487, 0.3900, 0.3183, 0.6703, 0.5376],
  9315. [0.6165, 0.4106, 0.7575, 0.1733, 0.3838, 0.2650, 0.5680, 0.5116]],
  9316. device='cuda:0', dtype=torch.float64)
  9317. predictions are: tensor([[0.6306, 0.3991, 0.8891, 0.5327, 0.4077, 0.3841, 0.5966, 0.4759],
  9318. [0.5868, 0.3788, 0.8220, 0.5059, 0.4200, 0.4516, 0.5521, 0.5059],
  9319. [0.5743, 0.3819, 0.8683, 0.4992, 0.4607, 0.5130, 0.6418, 0.4970],
  9320. [0.5523, 0.3709, 0.7587, 0.2767, 0.3894, 0.3869, 0.5942, 0.5189],
  9321. [0.5849, 0.3792, 0.7901, 0.2298, 0.3933, 0.2529, 0.5977, 0.4827],
  9322. [0.6309, 0.4188, 0.8185, 0.3005, 0.4504, 0.2756, 0.6154, 0.5288],
  9323. [0.6407, 0.4076, 0.8784, 0.3673, 0.4201, 0.2635, 0.6971, 0.5146],
  9324. [0.5239, 0.3321, 0.7461, 0.2019, 0.4320, 0.2133, 0.5811, 0.4603]],
  9325. device='cuda:0', grad_fn=<AddmmBackward>)
  9326. landmarks are: tensor([[[0.6203, 0.4020, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  9327. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  9328. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  9329. [0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500],
  9330. [0.6214, 0.4112, 0.7837, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  9331. [0.6200, 0.4098, 0.8238, 0.2917, 0.4013, 0.2967, 0.6000, 0.5683],
  9332. [0.6289, 0.4081, 0.8720, 0.3487, 0.3900, 0.3183, 0.6703, 0.5376],
  9333. [0.6165, 0.4106, 0.7575, 0.1733, 0.3837, 0.2650, 0.5680, 0.5116]]],
  9334. device='cuda:0')
  9335. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  9336. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  9337. loss_train: 0.0023326061200350523
  9338. step: 2
  9339. running loss: 0.0011663030600175261
  9340.  
  9341. Train Steps: 2/90 Loss: 0.0012 torch.Size([8, 600, 800])
  9342. torch.Size([8, 8])
  9343. tensor([[0.6207, 0.4081, 0.7662, 0.2067, 0.3962, 0.3200, 0.6312, 0.5300],
  9344. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  9345. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864],
  9346. [0.6332, 0.4118, 0.9238, 0.4267, 0.4012, 0.4733, 0.7525, 0.5436],
  9347. [0.6250, 0.4103, 0.8950, 0.4400, 0.3912, 0.5650, 0.6050, 0.5133],
  9348. [ nan, nan, 0.6900, 0.1917, 0.3937, 0.2367, 0.5240, 0.5246],
  9349. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  9350. [0.6300, 0.4133, 0.8538, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413]],
  9351. device='cuda:0', dtype=torch.float64)
  9352. predictions are: tensor([[0.5860, 0.3922, 0.7268, 0.2220, 0.3817, 0.2936, 0.5812, 0.4967],
  9353. [0.4988, 0.3249, 0.8917, 0.3148, 0.4922, 0.2506, 0.6934, 0.4801],
  9354. [0.7497, 0.4933, 0.8419, 0.5241, 0.3644, 0.4834, 0.6657, 0.5144],
  9355. [0.7188, 0.4813, 0.8878, 0.4690, 0.3864, 0.4804, 0.6865, 0.5008],
  9356. [0.7292, 0.4875, 0.8690, 0.4671, 0.3839, 0.5779, 0.5837, 0.4822],
  9357. [0.0061, 0.0024, 0.7012, 0.2367, 0.3962, 0.2293, 0.4864, 0.4885],
  9358. [0.7251, 0.4817, 0.8466, 0.4170, 0.3446, 0.3731, 0.5056, 0.4686],
  9359. [0.6953, 0.4570, 0.8304, 0.2385, 0.5326, 0.2588, 0.6773, 0.4800]],
  9360. device='cuda:0', grad_fn=<AddmmBackward>)
  9361. landmarks are: tensor([[[0.6207, 0.4081, 0.7663, 0.2067, 0.3963, 0.3200, 0.6313, 0.5300],
  9362. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  9363. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864],
  9364. [0.6332, 0.4118, 0.9237, 0.4267, 0.4013, 0.4733, 0.7525, 0.5436],
  9365. [0.6250, 0.4103, 0.8950, 0.4400, 0.3913, 0.5650, 0.6050, 0.5133],
  9366. [0.0000, 0.0000, 0.6900, 0.1917, 0.3938, 0.2367, 0.5240, 0.5246],
  9367. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  9368. [0.6300, 0.4133, 0.8537, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413]]],
  9369. device='cuda:0')
  9370. loss_train_step before backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  9371. loss_train_step after backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  9372. loss_train: 0.004649204900488257
  9373. step: 3
  9374. running loss: 0.0015497349668294191
  9375. Train Steps: 3/90 Loss: 0.0015 torch.Size([8, 600, 800])
  9376. torch.Size([8, 8])
  9377. tensor([[0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  9378. [0.6300, 0.4133, 0.8538, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  9379. [0.6151, 0.4125, 0.8738, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  9380. [0.6099, 0.4030, 0.8638, 0.5117, 0.4983, 0.4965, 0.5086, 0.5388],
  9381. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  9382. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  9383. [0.6142, 0.3982, 0.8650, 0.4883, 0.3912, 0.4317, 0.5315, 0.5350],
  9384. [0.6267, 0.4065, 0.8313, 0.2467, 0.4788, 0.1733, 0.6312, 0.5133]],
  9385. device='cuda:0', dtype=torch.float64)
  9386. predictions are: tensor([[0.6228, 0.4140, 0.8659, 0.4301, 0.3465, 0.3910, 0.6165, 0.5447],
  9387. [0.6294, 0.4215, 0.8306, 0.2009, 0.5087, 0.2778, 0.7287, 0.4982],
  9388. [0.5685, 0.3807, 0.8171, 0.4263, 0.3248, 0.3840, 0.5332, 0.5097],
  9389. [0.6234, 0.4169, 0.8464, 0.4805, 0.4680, 0.4954, 0.5454, 0.4970],
  9390. [0.6373, 0.4156, 0.8276, 0.4305, 0.3745, 0.5263, 0.6380, 0.4825],
  9391. [0.5885, 0.3843, 0.8212, 0.4443, 0.3774, 0.5082, 0.5607, 0.4995],
  9392. [0.5832, 0.3737, 0.8340, 0.4706, 0.3457, 0.4484, 0.5524, 0.4831],
  9393. [0.6135, 0.4110, 0.8132, 0.1869, 0.4392, 0.1840, 0.6581, 0.4998]],
  9394. device='cuda:0', grad_fn=<AddmmBackward>)
  9395. landmarks are: tensor([[[0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  9396. [0.6300, 0.4133, 0.8537, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  9397. [0.6151, 0.4125, 0.8737, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  9398. [0.6098, 0.4030, 0.8637, 0.5117, 0.4983, 0.4965, 0.5086, 0.5388],
  9399. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  9400. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  9401. [0.6143, 0.3982, 0.8650, 0.4883, 0.3913, 0.4317, 0.5315, 0.5350],
  9402. [0.6266, 0.4065, 0.8313, 0.2467, 0.4787, 0.1733, 0.6313, 0.5133]]],
  9403. device='cuda:0')
  9404. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  9405. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  9406. loss_train: 0.00562284019542858
  9407. step: 4
  9408. running loss: 0.001405710048857145
  9409. Train Steps: 4/90 Loss: 0.0014 torch.Size([8, 600, 800])
  9410. torch.Size([8, 8])
  9411. tensor([[0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  9412. [ nan, nan, 0.8488, 0.2300, 0.5563, 0.2100, 0.7390, 0.5679],
  9413. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092],
  9414. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  9415. [0.6185, 0.4080, 0.8625, 0.3483, 0.3788, 0.2650, 0.5320, 0.5272],
  9416. [ nan, nan, 0.6688, 0.2513, 0.4113, 0.2117, 0.5193, 0.5933],
  9417. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5988, 0.5667],
  9418. [0.6109, 0.4041, 0.6975, 0.3167, 0.3513, 0.3383, 0.5153, 0.5319]],
  9419. device='cuda:0', dtype=torch.float64)
  9420. predictions are: tensor([[0.6993, 0.4755, 0.8851, 0.5005, 0.3740, 0.4976, 0.6179, 0.5072],
  9421. [0.4512, 0.3040, 0.8451, 0.2089, 0.4832, 0.2711, 0.7245, 0.5286],
  9422. [0.6676, 0.4357, 0.8666, 0.4902, 0.3854, 0.5505, 0.5973, 0.4928],
  9423. [0.6404, 0.4130, 0.8976, 0.3328, 0.4247, 0.2651, 0.6338, 0.5441],
  9424. [0.6596, 0.4259, 0.8557, 0.3324, 0.3837, 0.2897, 0.5354, 0.5428],
  9425. [0.2002, 0.1357, 0.6766, 0.1868, 0.3779, 0.2244, 0.5412, 0.5682],
  9426. [0.6832, 0.4638, 0.8442, 0.3196, 0.3664, 0.3608, 0.6115, 0.5717],
  9427. [0.5639, 0.3835, 0.7373, 0.2500, 0.3418, 0.3510, 0.5388, 0.5312]],
  9428. device='cuda:0', grad_fn=<AddmmBackward>)
  9429. landmarks are: tensor([[[0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  9430. [0.0000, 0.0000, 0.8487, 0.2300, 0.5562, 0.2100, 0.7390, 0.5679],
  9431. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092],
  9432. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  9433. [0.6186, 0.4080, 0.8625, 0.3483, 0.3787, 0.2650, 0.5320, 0.5272],
  9434. [0.0000, 0.0000, 0.6688, 0.2513, 0.4112, 0.2117, 0.5193, 0.5933],
  9435. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5987, 0.5667],
  9436. [0.6109, 0.4041, 0.6975, 0.3167, 0.3512, 0.3383, 0.5153, 0.5319]]],
  9437. device='cuda:0')
  9438. loss_train_step before backward: tensor(0.0065, device='cuda:0', grad_fn=<MseLossBackward>)
  9439. loss_train_step after backward: tensor(0.0065, device='cuda:0', grad_fn=<MseLossBackward>)
  9440. loss_train: 0.01212452695472166
  9441. step: 5
  9442. running loss: 0.002424905390944332
  9443. Train Steps: 5/90 Loss: 0.0024 torch.Size([8, 600, 800])
  9444. torch.Size([8, 8])
  9445. tensor([[0.6236, 0.3966, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  9446. [0.6273, 0.4110, 0.8900, 0.3817, 0.4188, 0.2167, 0.5858, 0.4835],
  9447. [0.6188, 0.4099, 0.7400, 0.2433, 0.3962, 0.2750, 0.6162, 0.5467],
  9448. [0.6353, 0.4128, 0.9138, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  9449. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  9450. [0.6087, 0.3976, 0.8337, 0.3867, 0.3713, 0.3117, 0.5938, 0.5300],
  9451. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  9452. [0.6075, 0.4007, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280]],
  9453. device='cuda:0', dtype=torch.float64)
  9454. predictions are: tensor([[0.5911, 0.3556, 0.8988, 0.4392, 0.3467, 0.4394, 0.5725, 0.5445],
  9455. [0.6665, 0.4088, 0.9202, 0.3296, 0.4442, 0.2362, 0.6043, 0.5236],
  9456. [0.5699, 0.3660, 0.7474, 0.1740, 0.3935, 0.3022, 0.5830, 0.5746],
  9457. [0.5511, 0.3537, 0.9361, 0.3323, 0.4631, 0.3199, 0.7435, 0.5976],
  9458. [0.3874, 0.2319, 0.7176, 0.1733, 0.4439, 0.1785, 0.5358, 0.5427],
  9459. [0.6752, 0.4473, 0.8615, 0.3541, 0.3713, 0.3553, 0.5935, 0.5718],
  9460. [0.5413, 0.3587, 0.8534, 0.5498, 0.3760, 0.4432, 0.5977, 0.5819],
  9461. [0.6556, 0.4265, 0.8246, 0.4529, 0.4112, 0.5356, 0.5225, 0.5649]],
  9462. device='cuda:0', grad_fn=<AddmmBackward>)
  9463. landmarks are: tensor([[[0.6236, 0.3965, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  9464. [0.6273, 0.4110, 0.8900, 0.3817, 0.4187, 0.2167, 0.5858, 0.4835],
  9465. [0.6188, 0.4099, 0.7400, 0.2433, 0.3963, 0.2750, 0.6162, 0.5467],
  9466. [0.6353, 0.4128, 0.9137, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  9467. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  9468. [0.6087, 0.3976, 0.8338, 0.3867, 0.3713, 0.3117, 0.5938, 0.5300],
  9469. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  9470. [0.6075, 0.4006, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280]]],
  9471. device='cuda:0')
  9472. loss_train_step before backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  9473. loss_train_step after backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  9474. loss_train: 0.014573542110156268
  9475. step: 6
  9476. running loss: 0.0024289236850260445
  9477.  
  9478. Train Steps: 6/90 Loss: 0.0024 torch.Size([8, 600, 800])
  9479. torch.Size([8, 8])
  9480. tensor([[0.6277, 0.4118, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  9481. [ nan, nan, 0.7648, 0.2722, 0.3962, 0.2183, 0.5060, 0.5422],
  9482. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  9483. [0.6252, 0.4158, 0.8988, 0.4083, 0.3788, 0.4783, 0.6225, 0.5633],
  9484. [0.6230, 0.4113, 0.7213, 0.1983, 0.4325, 0.2367, 0.6262, 0.5400],
  9485. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  9486. [0.6229, 0.4198, 0.7662, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783],
  9487. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250]],
  9488. device='cuda:0', dtype=torch.float64)
  9489. predictions are: tensor([[ 0.6546, 0.4125, 0.9096, 0.4107, 0.3861, 0.2479, 0.6083, 0.5739],
  9490. [ 0.0055, -0.0085, 0.7559, 0.2674, 0.3948, 0.2246, 0.4803, 0.5537],
  9491. [ 0.5819, 0.3834, 0.7958, 0.2445, 0.4335, 0.2698, 0.6186, 0.5927],
  9492. [ 0.7106, 0.4729, 0.9110, 0.4499, 0.3817, 0.4728, 0.6339, 0.5947],
  9493. [ 0.5566, 0.3425, 0.7486, 0.2117, 0.4368, 0.2164, 0.6001, 0.5948],
  9494. [ 0.5981, 0.3679, 0.8987, 0.3423, 0.4487, 0.3167, 0.6891, 0.5739],
  9495. [ 0.6130, 0.3963, 0.7683, 0.2571, 0.4715, 0.2283, 0.5340, 0.6134],
  9496. [ 0.6651, 0.4216, 0.9075, 0.5245, 0.4099, 0.5539, 0.5701, 0.5658]],
  9497. device='cuda:0', grad_fn=<AddmmBackward>)
  9498. landmarks are: tensor([[[0.6277, 0.4117, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  9499. [0.0000, 0.0000, 0.7648, 0.2722, 0.3963, 0.2183, 0.5060, 0.5422],
  9500. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  9501. [0.6252, 0.4158, 0.8988, 0.4083, 0.3787, 0.4783, 0.6225, 0.5633],
  9502. [0.6230, 0.4113, 0.7212, 0.1983, 0.4325, 0.2367, 0.6263, 0.5400],
  9503. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  9504. [0.6229, 0.4198, 0.7663, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783],
  9505. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250]]],
  9506. device='cuda:0')
  9507. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  9508. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  9509. loss_train: 0.015636640891898423
  9510. step: 7
  9511. running loss: 0.0022338058416997747
  9512. Train Steps: 7/90 Loss: 0.0022 torch.Size([8, 600, 800])
  9513. torch.Size([8, 8])
  9514. tensor([[0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  9515. [0.6053, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  9516. [0.6202, 0.4079, 0.8025, 0.2500, 0.3763, 0.3217, 0.6125, 0.5533],
  9517. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5787, 0.5550],
  9518. [0.6230, 0.4152, 0.7588, 0.2283, 0.4012, 0.2883, 0.6200, 0.5767],
  9519. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167],
  9520. [0.6346, 0.4144, 0.9088, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  9521. [0.6239, 0.4206, 0.8750, 0.5400, 0.3688, 0.4850, 0.5737, 0.5700]],
  9522. device='cuda:0', dtype=torch.float64)
  9523. predictions are: tensor([[0.5450, 0.3389, 0.9121, 0.4218, 0.4478, 0.3192, 0.7040, 0.5919],
  9524. [0.4682, 0.2779, 0.7069, 0.2023, 0.4132, 0.1661, 0.5391, 0.5515],
  9525. [0.5974, 0.3781, 0.8077, 0.2790, 0.4024, 0.2788, 0.5935, 0.5618],
  9526. [0.5002, 0.3248, 0.8132, 0.3105, 0.3683, 0.3401, 0.5391, 0.5600],
  9527. [0.5790, 0.3737, 0.7905, 0.2610, 0.4246, 0.2579, 0.5991, 0.5909],
  9528. [0.5536, 0.3923, 0.8460, 0.3414, 0.3876, 0.2783, 0.5599, 0.5604],
  9529. [0.5235, 0.3519, 0.9101, 0.4799, 0.4183, 0.3825, 0.6948, 0.6061],
  9530. [0.6200, 0.4087, 0.8988, 0.5803, 0.4056, 0.4541, 0.5931, 0.5973]],
  9531. device='cuda:0', grad_fn=<AddmmBackward>)
  9532. landmarks are: tensor([[[0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  9533. [0.6054, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  9534. [0.6202, 0.4079, 0.8025, 0.2500, 0.3762, 0.3217, 0.6125, 0.5533],
  9535. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5788, 0.5550],
  9536. [0.6230, 0.4152, 0.7588, 0.2283, 0.4013, 0.2883, 0.6200, 0.5767],
  9537. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167],
  9538. [0.6346, 0.4144, 0.9087, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  9539. [0.6239, 0.4206, 0.8750, 0.5400, 0.3688, 0.4850, 0.5738, 0.5700]]],
  9540. device='cuda:0')
  9541. loss_train_step before backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  9542. loss_train_step after backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  9543. loss_train: 0.017651385918725282
  9544. step: 8
  9545. running loss: 0.0022064232398406602
  9546. Train Steps: 8/90 Loss: 0.0022 torch.Size([8, 600, 800])
  9547. torch.Size([8, 8])
  9548. tensor([[ nan, nan, 0.6900, 0.1917, 0.3937, 0.2367, 0.5240, 0.5246],
  9549. [0.6198, 0.4164, 0.8700, 0.5067, 0.4625, 0.5650, 0.5464, 0.5197],
  9550. [0.6202, 0.4066, 0.8398, 0.2648, 0.3925, 0.2627, 0.5845, 0.5124],
  9551. [0.6299, 0.4303, 0.7963, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  9552. [0.6198, 0.4101, 0.8838, 0.5283, 0.3763, 0.5267, 0.5913, 0.5567],
  9553. [0.6162, 0.4014, 0.8800, 0.5333, 0.3750, 0.4817, 0.5988, 0.5283],
  9554. [0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  9555. [ nan, nan, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729]],
  9556. device='cuda:0', dtype=torch.float64)
  9557. predictions are: tensor([[0.0428, 0.0314, 0.7288, 0.2345, 0.4140, 0.2244, 0.5735, 0.5468],
  9558. [0.7112, 0.4756, 0.9116, 0.5149, 0.4737, 0.5195, 0.6098, 0.5295],
  9559. [0.6298, 0.4101, 0.8339, 0.2671, 0.4087, 0.2395, 0.6392, 0.5416],
  9560. [0.6986, 0.4763, 0.8342, 0.3661, 0.4985, 0.2606, 0.5705, 0.6165],
  9561. [0.7682, 0.4894, 0.9038, 0.5506, 0.3864, 0.5193, 0.6369, 0.5682],
  9562. [0.7155, 0.4652, 0.8983, 0.5525, 0.3863, 0.4746, 0.6439, 0.5403],
  9563. [0.6947, 0.4640, 0.9098, 0.4999, 0.3746, 0.4000, 0.5840, 0.5835],
  9564. [0.1006, 0.0617, 0.7249, 0.2260, 0.4266, 0.1740, 0.5654, 0.5623]],
  9565. device='cuda:0', grad_fn=<AddmmBackward>)
  9566. landmarks are: tensor([[[0.0000, 0.0000, 0.6900, 0.1917, 0.3938, 0.2367, 0.5240, 0.5246],
  9567. [0.6198, 0.4164, 0.8700, 0.5067, 0.4625, 0.5650, 0.5464, 0.5197],
  9568. [0.6202, 0.4066, 0.8398, 0.2648, 0.3925, 0.2627, 0.5845, 0.5124],
  9569. [0.6299, 0.4303, 0.7962, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  9570. [0.6198, 0.4101, 0.8838, 0.5283, 0.3762, 0.5267, 0.5913, 0.5567],
  9571. [0.6162, 0.4014, 0.8800, 0.5333, 0.3750, 0.4817, 0.5987, 0.5283],
  9572. [0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  9573. [0.0000, 0.0000, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729]]],
  9574. device='cuda:0')
  9575. loss_train_step before backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  9576. loss_train_step after backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  9577. loss_train: 0.019613695156294852
  9578. step: 9
  9579. running loss: 0.002179299461810539
  9580. Train Steps: 9/90 Loss: 0.0022 torch.Size([8, 600, 800])
  9581. torch.Size([8, 8])
  9582. tensor([[0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6138, 0.5400],
  9583. [0.6286, 0.4055, 0.9000, 0.4717, 0.3763, 0.4683, 0.7018, 0.5494],
  9584. [0.6160, 0.4093, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  9585. [0.6224, 0.4061, 0.8988, 0.4300, 0.3838, 0.4750, 0.6112, 0.5483],
  9586. [0.6321, 0.4048, 0.8738, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  9587. [0.6064, 0.4019, 0.8650, 0.4517, 0.4037, 0.5367, 0.5703, 0.5609],
  9588. [0.6188, 0.4099, 0.7400, 0.2433, 0.3962, 0.2750, 0.6162, 0.5467],
  9589. [0.6072, 0.4029, 0.7037, 0.2150, 0.3912, 0.2267, 0.5516, 0.5507]],
  9590. device='cuda:0', dtype=torch.float64)
  9591. predictions are: tensor([[0.6030, 0.4097, 0.8930, 0.4494, 0.3813, 0.4447, 0.6150, 0.5515],
  9592. [0.6140, 0.4132, 0.9168, 0.5096, 0.3778, 0.4366, 0.7009, 0.5404],
  9593. [0.5959, 0.4015, 0.8532, 0.4779, 0.3983, 0.4302, 0.5673, 0.5675],
  9594. [0.6657, 0.4229, 0.9223, 0.4793, 0.3905, 0.4876, 0.6136, 0.5228],
  9595. [0.6222, 0.3936, 0.8991, 0.5774, 0.3917, 0.4128, 0.6241, 0.4926],
  9596. [0.5483, 0.3770, 0.8813, 0.4745, 0.4251, 0.5456, 0.6047, 0.5475],
  9597. [0.5623, 0.3769, 0.7501, 0.2297, 0.4064, 0.2503, 0.5999, 0.5592],
  9598. [0.5206, 0.3534, 0.7160, 0.2273, 0.4247, 0.2127, 0.5603, 0.5618]],
  9599. device='cuda:0', grad_fn=<AddmmBackward>)
  9600. landmarks are: tensor([[[0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6137, 0.5400],
  9601. [0.6286, 0.4055, 0.9000, 0.4717, 0.3762, 0.4683, 0.7018, 0.5494],
  9602. [0.6160, 0.4092, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  9603. [0.6224, 0.4061, 0.8988, 0.4300, 0.3837, 0.4750, 0.6112, 0.5483],
  9604. [0.6321, 0.4048, 0.8737, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  9605. [0.6064, 0.4019, 0.8650, 0.4517, 0.4038, 0.5367, 0.5703, 0.5609],
  9606. [0.6188, 0.4099, 0.7400, 0.2433, 0.3963, 0.2750, 0.6162, 0.5467],
  9607. [0.6072, 0.4029, 0.7038, 0.2150, 0.3913, 0.2267, 0.5516, 0.5507]]],
  9608. device='cuda:0')
  9609. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  9610. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  9611. loss_train: 0.020335583365522325
  9612. step: 10
  9613. running loss: 0.0020335583365522327
  9614.  
  9615. Train Steps: 10/90 Loss: 0.0020 torch.Size([8, 600, 800])
  9616. torch.Size([8, 8])
  9617. tensor([[ nan, nan, 0.8850, 0.2817, 0.5112, 0.2183, 0.7184, 0.5436],
  9618. [0.6199, 0.4060, 0.8888, 0.4667, 0.3800, 0.5050, 0.6188, 0.5433],
  9619. [0.6127, 0.4084, 0.8700, 0.4467, 0.3987, 0.4317, 0.5013, 0.5471],
  9620. [0.6193, 0.4079, 0.7288, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  9621. [0.6213, 0.4131, 0.8438, 0.3550, 0.3513, 0.4400, 0.5716, 0.5123],
  9622. [0.6222, 0.4169, 0.8638, 0.5650, 0.4313, 0.4783, 0.5637, 0.5633],
  9623. [0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  9624. [0.6189, 0.4049, 0.8888, 0.4417, 0.4213, 0.5200, 0.5988, 0.5633]],
  9625. device='cuda:0', dtype=torch.float64)
  9626. predictions are: tensor([[0.1967, 0.1461, 0.8668, 0.2789, 0.5008, 0.2300, 0.7540, 0.5135],
  9627. [0.6227, 0.4107, 0.8813, 0.4764, 0.3725, 0.5147, 0.6305, 0.5193],
  9628. [0.6610, 0.4424, 0.8588, 0.4559, 0.3658, 0.4232, 0.5012, 0.5153],
  9629. [0.5926, 0.3841, 0.7065, 0.2488, 0.4126, 0.2771, 0.6147, 0.5956],
  9630. [0.6309, 0.4225, 0.8329, 0.3700, 0.3376, 0.4131, 0.5710, 0.5053],
  9631. [0.5498, 0.3652, 0.8582, 0.5761, 0.4092, 0.4807, 0.5568, 0.5532],
  9632. [0.6210, 0.4160, 0.8617, 0.4565, 0.3626, 0.3940, 0.5778, 0.4699],
  9633. [0.6542, 0.4385, 0.8640, 0.4484, 0.4057, 0.5174, 0.6278, 0.5312]],
  9634. device='cuda:0', grad_fn=<AddmmBackward>)
  9635. landmarks are: tensor([[[0.0000, 0.0000, 0.8850, 0.2817, 0.5113, 0.2183, 0.7184, 0.5436],
  9636. [0.6199, 0.4060, 0.8888, 0.4667, 0.3800, 0.5050, 0.6187, 0.5433],
  9637. [0.6127, 0.4084, 0.8700, 0.4467, 0.3988, 0.4317, 0.5013, 0.5471],
  9638. [0.6193, 0.4078, 0.7287, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  9639. [0.6213, 0.4131, 0.8438, 0.3550, 0.3512, 0.4400, 0.5716, 0.5123],
  9640. [0.6222, 0.4169, 0.8637, 0.5650, 0.4313, 0.4783, 0.5638, 0.5633],
  9641. [0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  9642. [0.6189, 0.4049, 0.8888, 0.4417, 0.4212, 0.5200, 0.5987, 0.5633]]],
  9643. device='cuda:0')
  9644. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  9645. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  9646. loss_train: 0.02176640962716192
  9647. step: 11
  9648. running loss: 0.0019787645115601745
  9649. Train Steps: 11/90 Loss: 0.0020 torch.Size([8, 600, 800])
  9650. torch.Size([8, 8])
  9651. tensor([[0.6254, 0.3993, 0.8988, 0.4767, 0.3987, 0.5517, 0.6955, 0.5285],
  9652. [0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050],
  9653. [0.6199, 0.4065, 0.7598, 0.2385, 0.4317, 0.1981, 0.5933, 0.5221],
  9654. [0.6277, 0.4103, 0.8087, 0.5717, 0.4188, 0.4750, 0.5663, 0.6083],
  9655. [0.6250, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6088, 0.5183],
  9656. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  9657. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  9658. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5363, 0.5550]],
  9659. device='cuda:0', dtype=torch.float64)
  9660. predictions are: tensor([[0.5811, 0.3768, 0.9099, 0.4876, 0.3926, 0.5734, 0.6679, 0.5131],
  9661. [0.5699, 0.4015, 0.8979, 0.4165, 0.3709, 0.4824, 0.6132, 0.5498],
  9662. [0.5431, 0.3632, 0.7523, 0.2119, 0.4363, 0.2313, 0.5866, 0.4968],
  9663. [0.5687, 0.3855, 0.8256, 0.5647, 0.3947, 0.4864, 0.5870, 0.5849],
  9664. [0.6221, 0.4016, 0.9093, 0.4778, 0.4513, 0.5638, 0.6455, 0.5084],
  9665. [0.4819, 0.3230, 0.6971, 0.3072, 0.3550, 0.3079, 0.5256, 0.5407],
  9666. [0.5666, 0.3805, 0.8646, 0.4635, 0.3537, 0.3870, 0.5214, 0.5420],
  9667. [0.5962, 0.4013, 0.7175, 0.2607, 0.3857, 0.2391, 0.5381, 0.5099]],
  9668. device='cuda:0', grad_fn=<AddmmBackward>)
  9669. landmarks are: tensor([[[0.6254, 0.3993, 0.8988, 0.4767, 0.3988, 0.5517, 0.6955, 0.5285],
  9670. [0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050],
  9671. [0.6199, 0.4065, 0.7598, 0.2385, 0.4317, 0.1981, 0.5933, 0.5221],
  9672. [0.6277, 0.4103, 0.8087, 0.5717, 0.4187, 0.4750, 0.5663, 0.6083],
  9673. [0.6251, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6087, 0.5183],
  9674. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  9675. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  9676. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5362, 0.5550]]],
  9677. device='cuda:0')
  9678. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  9679. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  9680. loss_train: 0.02287836465984583
  9681. step: 12
  9682. running loss: 0.0019065303883204858
  9683. Train Steps: 12/90 Loss: 0.0019 torch.Size([8, 600, 800])
  9684. torch.Size([8, 8])
  9685. tensor([[0.6214, 0.4112, 0.7838, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  9686. [0.6162, 0.4014, 0.8800, 0.5333, 0.3750, 0.4817, 0.5988, 0.5283],
  9687. [0.6200, 0.4059, 0.8700, 0.4900, 0.4163, 0.5000, 0.6162, 0.5467],
  9688. [0.6182, 0.4099, 0.7812, 0.3000, 0.3937, 0.2367, 0.5325, 0.5750],
  9689. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  9690. [0.6317, 0.4038, 0.8287, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  9691. [0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  9692. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533]],
  9693. device='cuda:0', dtype=torch.float64)
  9694. predictions are: tensor([[0.5935, 0.3896, 0.7540, 0.1949, 0.3524, 0.3019, 0.5652, 0.5206],
  9695. [0.5673, 0.3691, 0.8442, 0.5159, 0.3690, 0.4971, 0.5622, 0.5308],
  9696. [0.5962, 0.3964, 0.8590, 0.4969, 0.3963, 0.5442, 0.5967, 0.5409],
  9697. [0.4966, 0.3253, 0.7469, 0.2914, 0.4054, 0.2778, 0.5003, 0.5784],
  9698. [0.5631, 0.3603, 0.8656, 0.3860, 0.4165, 0.3810, 0.6919, 0.5644],
  9699. [0.5929, 0.3764, 0.8253, 0.5604, 0.3623, 0.4864, 0.5951, 0.5051],
  9700. [0.5627, 0.3630, 0.8148, 0.3790, 0.3547, 0.4484, 0.5368, 0.5210],
  9701. [0.5737, 0.4015, 0.8589, 0.4194, 0.4251, 0.5275, 0.5585, 0.5323]],
  9702. device='cuda:0', grad_fn=<AddmmBackward>)
  9703. landmarks are: tensor([[[0.6214, 0.4112, 0.7837, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  9704. [0.6162, 0.4014, 0.8800, 0.5333, 0.3750, 0.4817, 0.5987, 0.5283],
  9705. [0.6199, 0.4059, 0.8700, 0.4900, 0.4162, 0.5000, 0.6162, 0.5467],
  9706. [0.6182, 0.4099, 0.7812, 0.3000, 0.3938, 0.2367, 0.5325, 0.5750],
  9707. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  9708. [0.6317, 0.4038, 0.8288, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  9709. [0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  9710. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533]]],
  9711. device='cuda:0')
  9712. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  9713. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  9714. loss_train: 0.02397310920059681
  9715. step: 13
  9716. running loss: 0.0018440853231228315
  9717. Train Steps: 13/90 Loss: 0.0018 torch.Size([8, 600, 800])
  9718. torch.Size([8, 8])
  9719. tensor([[0.6277, 0.4083, 0.8350, 0.2717, 0.4562, 0.1800, 0.5918, 0.4878],
  9720. [0.6175, 0.4091, 0.7863, 0.2800, 0.3638, 0.3583, 0.6188, 0.5433],
  9721. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5787, 0.5550],
  9722. [0.6059, 0.4002, 0.7562, 0.2767, 0.3538, 0.3033, 0.5529, 0.5455],
  9723. [0.6307, 0.4029, 0.8650, 0.5200, 0.3763, 0.4017, 0.7311, 0.5366],
  9724. [0.6296, 0.4045, 0.9138, 0.4100, 0.4232, 0.4242, 0.7422, 0.5297],
  9725. [0.6275, 0.4111, 0.8463, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  9726. [0.6200, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183]],
  9727. device='cuda:0', dtype=torch.float64)
  9728. predictions are: tensor([[0.5830, 0.3639, 0.7772, 0.2662, 0.4585, 0.2378, 0.5365, 0.5227],
  9729. [0.5380, 0.3593, 0.7519, 0.2464, 0.3452, 0.3739, 0.5676, 0.5757],
  9730. [0.5398, 0.3570, 0.7709, 0.2866, 0.3395, 0.4024, 0.5126, 0.5589],
  9731. [0.5468, 0.3776, 0.7313, 0.2644, 0.3622, 0.3308, 0.5612, 0.5544],
  9732. [0.5890, 0.3941, 0.8587, 0.5155, 0.3783, 0.4066, 0.6172, 0.5367],
  9733. [0.6245, 0.4113, 0.8901, 0.4214, 0.3851, 0.4393, 0.6568, 0.5655],
  9734. [0.5801, 0.3806, 0.8156, 0.2517, 0.4710, 0.2306, 0.5710, 0.5322],
  9735. [0.5394, 0.3569, 0.7571, 0.2928, 0.4019, 0.3020, 0.5815, 0.5393]],
  9736. device='cuda:0', grad_fn=<AddmmBackward>)
  9737. landmarks are: tensor([[[0.6277, 0.4083, 0.8350, 0.2717, 0.4563, 0.1800, 0.5918, 0.4878],
  9738. [0.6175, 0.4091, 0.7862, 0.2800, 0.3638, 0.3583, 0.6187, 0.5433],
  9739. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5788, 0.5550],
  9740. [0.6059, 0.4002, 0.7563, 0.2767, 0.3537, 0.3033, 0.5529, 0.5455],
  9741. [0.6307, 0.4029, 0.8650, 0.5200, 0.3762, 0.4017, 0.7311, 0.5366],
  9742. [0.6296, 0.4045, 0.9137, 0.4100, 0.4232, 0.4242, 0.7422, 0.5297],
  9743. [0.6275, 0.4111, 0.8462, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  9744. [0.6201, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183]]],
  9745. device='cuda:0')
  9746. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  9747. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  9748. loss_train: 0.025551110040396452
  9749. step: 14
  9750. running loss: 0.0018250792885997466
  9751.  
  9752. Train Steps: 14/90 Loss: 0.0018 torch.Size([8, 600, 800])
  9753. torch.Size([8, 8])
  9754. tensor([[0.6250, 0.4013, 0.8525, 0.5417, 0.4037, 0.5117, 0.6325, 0.5017],
  9755. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  9756. [0.6246, 0.4008, 0.8757, 0.5088, 0.4101, 0.5392, 0.6644, 0.5133],
  9757. [0.6262, 0.4052, 0.8888, 0.4700, 0.3675, 0.5117, 0.6350, 0.5233],
  9758. [0.6267, 0.4080, 0.8438, 0.2633, 0.4763, 0.1800, 0.6259, 0.5240],
  9759. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309],
  9760. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  9761. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795]],
  9762. device='cuda:0', dtype=torch.float64)
  9763. predictions are: tensor([[0.5966, 0.3827, 0.8279, 0.5126, 0.3845, 0.5103, 0.5687, 0.5114],
  9764. [0.6408, 0.4424, 0.7315, 0.2028, 0.3657, 0.2999, 0.5820, 0.5476],
  9765. [0.6043, 0.3967, 0.8429, 0.4752, 0.3932, 0.5304, 0.6078, 0.5437],
  9766. [0.5796, 0.3996, 0.8783, 0.4260, 0.3826, 0.5202, 0.6104, 0.5483],
  9767. [0.6333, 0.4238, 0.8142, 0.2279, 0.4601, 0.1983, 0.5953, 0.5282],
  9768. [0.6257, 0.4307, 0.8559, 0.4417, 0.3964, 0.3713, 0.5981, 0.5385],
  9769. [0.6219, 0.4237, 0.8560, 0.4450, 0.3656, 0.4629, 0.5451, 0.5470],
  9770. [0.5676, 0.3878, 0.8360, 0.4887, 0.3789, 0.5326, 0.6601, 0.5957]],
  9771. device='cuda:0', grad_fn=<AddmmBackward>)
  9772. landmarks are: tensor([[[0.6250, 0.4013, 0.8525, 0.5417, 0.4038, 0.5117, 0.6325, 0.5017],
  9773. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  9774. [0.6246, 0.4008, 0.8757, 0.5088, 0.4101, 0.5392, 0.6644, 0.5133],
  9775. [0.6262, 0.4052, 0.8888, 0.4700, 0.3675, 0.5117, 0.6350, 0.5233],
  9776. [0.6267, 0.4080, 0.8438, 0.2633, 0.4762, 0.1800, 0.6259, 0.5240],
  9777. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309],
  9778. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  9779. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795]]],
  9780. device='cuda:0')
  9781. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  9782. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  9783. loss_train: 0.026409554469864815
  9784. step: 15
  9785. running loss: 0.0017606369646576544
  9786. Train Steps: 15/90 Loss: 0.0018 torch.Size([8, 600, 800])
  9787. torch.Size([8, 8])
  9788. tensor([[0.6210, 0.4164, 0.7202, 0.2930, 0.4025, 0.2483, 0.5687, 0.5567],
  9789. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  9790. [0.6126, 0.4073, 0.8750, 0.5133, 0.3800, 0.4333, 0.4986, 0.5378],
  9791. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  9792. [0.6179, 0.4040, 0.7412, 0.1850, 0.3825, 0.2783, 0.5837, 0.5600],
  9793. [0.6135, 0.3994, 0.7913, 0.3050, 0.3625, 0.3050, 0.5837, 0.5050],
  9794. [0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500],
  9795. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483]],
  9796. device='cuda:0', dtype=torch.float64)
  9797. predictions are: tensor([[0.5237, 0.3421, 0.7651, 0.2634, 0.4441, 0.2637, 0.6147, 0.5645],
  9798. [0.6785, 0.4316, 0.8797, 0.5774, 0.3911, 0.4217, 0.5852, 0.5596],
  9799. [0.6267, 0.4133, 0.8920, 0.5273, 0.3849, 0.4389, 0.5403, 0.5284],
  9800. [0.6243, 0.3974, 0.8944, 0.5076, 0.4023, 0.5420, 0.6750, 0.5255],
  9801. [0.7272, 0.4969, 0.7238, 0.2236, 0.3934, 0.2854, 0.6170, 0.5553],
  9802. [0.6593, 0.4337, 0.7901, 0.2948, 0.3699, 0.3122, 0.6406, 0.5055],
  9803. [0.6604, 0.4426, 0.7870, 0.2455, 0.3581, 0.3741, 0.6380, 0.5378],
  9804. [0.5822, 0.3795, 0.8651, 0.4301, 0.3749, 0.4536, 0.6211, 0.5225]],
  9805. device='cuda:0', grad_fn=<AddmmBackward>)
  9806. landmarks are: tensor([[[0.6210, 0.4164, 0.7202, 0.2930, 0.4025, 0.2483, 0.5688, 0.5567],
  9807. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  9808. [0.6126, 0.4073, 0.8750, 0.5133, 0.3800, 0.4333, 0.4986, 0.5378],
  9809. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  9810. [0.6179, 0.4040, 0.7412, 0.1850, 0.3825, 0.2783, 0.5838, 0.5600],
  9811. [0.6135, 0.3994, 0.7912, 0.3050, 0.3625, 0.3050, 0.5838, 0.5050],
  9812. [0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500],
  9813. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483]]],
  9814. device='cuda:0')
  9815. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  9816. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  9817. loss_train: 0.02758348052157089
  9818. step: 16
  9819. running loss: 0.0017239675325981807
  9820. Train Steps: 16/90 Loss: 0.0017 torch.Size([8, 600, 800])
  9821. torch.Size([8, 8])
  9822. tensor([[0.6304, 0.4024, 0.8925, 0.4800, 0.3937, 0.4817, 0.7485, 0.5297],
  9823. [0.6187, 0.4104, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  9824. [0.6201, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  9825. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290],
  9826. [0.6206, 0.4001, 0.8900, 0.3933, 0.3588, 0.3567, 0.5837, 0.5083],
  9827. [0.6122, 0.3993, 0.8738, 0.4667, 0.4517, 0.4879, 0.5155, 0.4927],
  9828. [ nan, nan, 0.7515, 0.2708, 0.3987, 0.2267, 0.5162, 0.5567],
  9829. [0.6112, 0.4029, 0.8638, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567]],
  9830. device='cuda:0', dtype=torch.float64)
  9831. predictions are: tensor([[0.7156, 0.4425, 0.8823, 0.4837, 0.3652, 0.4849, 0.7299, 0.5256],
  9832. [0.7126, 0.4818, 0.7093, 0.2302, 0.3714, 0.2479, 0.5777, 0.5423],
  9833. [0.7234, 0.4645, 0.7586, 0.2182, 0.4175, 0.2038, 0.6208, 0.5105],
  9834. [0.6508, 0.4159, 0.7700, 0.2354, 0.4291, 0.2011, 0.6831, 0.5272],
  9835. [0.7000, 0.4389, 0.9155, 0.4157, 0.3596, 0.3509, 0.5941, 0.5133],
  9836. [0.7093, 0.4505, 0.9017, 0.4934, 0.4225, 0.5042, 0.5203, 0.5092],
  9837. [0.1840, 0.1324, 0.7520, 0.2711, 0.3754, 0.2328, 0.5478, 0.5738],
  9838. [0.6669, 0.4338, 0.9112, 0.4935, 0.4624, 0.4776, 0.6183, 0.5521]],
  9839. device='cuda:0', grad_fn=<AddmmBackward>)
  9840. landmarks are: tensor([[[0.6304, 0.4024, 0.8925, 0.4800, 0.3938, 0.4817, 0.7485, 0.5297],
  9841. [0.6187, 0.4103, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  9842. [0.6202, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  9843. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290],
  9844. [0.6206, 0.4001, 0.8900, 0.3933, 0.3587, 0.3567, 0.5838, 0.5083],
  9845. [0.6122, 0.3993, 0.8737, 0.4667, 0.4517, 0.4879, 0.5155, 0.4927],
  9846. [0.0000, 0.0000, 0.7515, 0.2708, 0.3988, 0.2267, 0.5163, 0.5567],
  9847. [0.6112, 0.4029, 0.8637, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567]]],
  9848. device='cuda:0')
  9849. loss_train_step before backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  9850. loss_train_step after backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  9851. loss_train: 0.029665038164239377
  9852. step: 17
  9853. running loss: 0.0017450022449552575
  9854. Train Steps: 17/90 Loss: 0.0017 torch.Size([8, 600, 800])
  9855. torch.Size([8, 8])
  9856. tensor([[0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  9857. [0.6289, 0.4032, 0.8419, 0.5446, 0.4075, 0.5017, 0.6312, 0.5117],
  9858. [0.6118, 0.4052, 0.8463, 0.3917, 0.3538, 0.3450, 0.5053, 0.5593],
  9859. [0.6212, 0.4159, 0.8675, 0.5783, 0.4088, 0.4317, 0.5613, 0.5917],
  9860. [0.6226, 0.4098, 0.8912, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  9861. [0.6182, 0.3972, 0.8552, 0.5914, 0.3683, 0.4181, 0.5688, 0.5378],
  9862. [0.6102, 0.4001, 0.7738, 0.3583, 0.3463, 0.3800, 0.5524, 0.5689],
  9863. [0.6276, 0.4235, 0.8888, 0.5333, 0.3800, 0.3117, 0.5427, 0.6164]],
  9864. device='cuda:0', dtype=torch.float64)
  9865. predictions are: tensor([[0.6093, 0.3796, 0.8609, 0.5016, 0.3703, 0.3870, 0.6474, 0.5185],
  9866. [0.7249, 0.4409, 0.8565, 0.4559, 0.3879, 0.4894, 0.6929, 0.4906],
  9867. [0.6259, 0.4103, 0.8489, 0.3080, 0.3685, 0.3349, 0.5577, 0.5404],
  9868. [0.6718, 0.4446, 0.8509, 0.4925, 0.4141, 0.4078, 0.6081, 0.5764],
  9869. [0.6810, 0.4354, 0.8846, 0.3431, 0.4289, 0.2577, 0.6137, 0.5327],
  9870. [0.6487, 0.4090, 0.8828, 0.5071, 0.3807, 0.4111, 0.6443, 0.5090],
  9871. [0.6511, 0.4149, 0.7971, 0.2856, 0.3502, 0.3732, 0.6101, 0.5407],
  9872. [0.6062, 0.3869, 0.8544, 0.4510, 0.4083, 0.3646, 0.6009, 0.5720]],
  9873. device='cuda:0', grad_fn=<AddmmBackward>)
  9874. landmarks are: tensor([[[0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  9875. [0.6289, 0.4031, 0.8419, 0.5446, 0.4075, 0.5017, 0.6313, 0.5117],
  9876. [0.6118, 0.4052, 0.8462, 0.3917, 0.3537, 0.3450, 0.5053, 0.5593],
  9877. [0.6212, 0.4159, 0.8675, 0.5783, 0.4087, 0.4317, 0.5612, 0.5917],
  9878. [0.6226, 0.4098, 0.8913, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  9879. [0.6182, 0.3972, 0.8552, 0.5914, 0.3683, 0.4181, 0.5688, 0.5378],
  9880. [0.6102, 0.4001, 0.7738, 0.3583, 0.3462, 0.3800, 0.5524, 0.5689],
  9881. [0.6276, 0.4235, 0.8888, 0.5333, 0.3800, 0.3117, 0.5427, 0.6164]]],
  9882. device='cuda:0')
  9883. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  9884. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  9885. loss_train: 0.03151935857022181
  9886. step: 18
  9887. running loss: 0.001751075476123434
  9888.  
  9889. Train Steps: 18/90 Loss: 0.0018 torch.Size([8, 600, 800])
  9890. torch.Size([8, 8])
  9891. tensor([[0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650],
  9892. [0.6081, 0.3950, 0.8538, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954],
  9893. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  9894. [0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  9895. [0.6108, 0.4011, 0.8037, 0.3400, 0.3700, 0.2933, 0.5658, 0.5617],
  9896. [0.6333, 0.4037, 0.8638, 0.5733, 0.4012, 0.4717, 0.6369, 0.4938],
  9897. [0.6250, 0.4131, 0.8688, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  9898. [0.6200, 0.4098, 0.8237, 0.2917, 0.4012, 0.2967, 0.6000, 0.5683]],
  9899. device='cuda:0', dtype=torch.float64)
  9900. predictions are: tensor([[0.6285, 0.4289, 0.8369, 0.3992, 0.3666, 0.3394, 0.5768, 0.5714],
  9901. [0.5382, 0.3461, 0.9028, 0.4570, 0.3757, 0.4687, 0.5894, 0.5244],
  9902. [0.6505, 0.4122, 0.6919, 0.2711, 0.3828, 0.2432, 0.5745, 0.5503],
  9903. [0.6192, 0.3904, 0.8273, 0.2519, 0.4510, 0.1801, 0.6629, 0.5030],
  9904. [0.6138, 0.3868, 0.8164, 0.3509, 0.3933, 0.3060, 0.6069, 0.5410],
  9905. [0.6278, 0.3900, 0.8875, 0.5938, 0.4004, 0.4714, 0.6583, 0.5202],
  9906. [0.6803, 0.4378, 0.8683, 0.3065, 0.4459, 0.2154, 0.6347, 0.5347],
  9907. [0.6068, 0.3959, 0.8165, 0.3017, 0.3987, 0.2889, 0.6404, 0.5753]],
  9908. device='cuda:0', grad_fn=<AddmmBackward>)
  9909. landmarks are: tensor([[[0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650],
  9910. [0.6081, 0.3950, 0.8537, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954],
  9911. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  9912. [0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  9913. [0.6108, 0.4011, 0.8037, 0.3400, 0.3700, 0.2933, 0.5658, 0.5617],
  9914. [0.6334, 0.4037, 0.8637, 0.5733, 0.4013, 0.4717, 0.6369, 0.4938],
  9915. [0.6250, 0.4131, 0.8687, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  9916. [0.6200, 0.4098, 0.8238, 0.2917, 0.4013, 0.2967, 0.6000, 0.5683]]],
  9917. device='cuda:0')
  9918. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  9919. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  9920. loss_train: 0.032034559291787446
  9921. step: 19
  9922. running loss: 0.0016860294364098656
  9923. Train Steps: 19/90 Loss: 0.0017 torch.Size([8, 600, 800])
  9924. torch.Size([8, 8])
  9925. tensor([[0.6257, 0.4024, 0.8672, 0.5422, 0.4196, 0.5198, 0.6694, 0.5204],
  9926. [0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166],
  9927. [0.6038, 0.3946, 0.8413, 0.4883, 0.3563, 0.4550, 0.5266, 0.4693],
  9928. [0.6136, 0.4117, 0.8700, 0.5167, 0.4188, 0.5083, 0.5147, 0.5495],
  9929. [0.6250, 0.4110, 0.7238, 0.2067, 0.4263, 0.1883, 0.5625, 0.5633],
  9930. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  9931. [0.6260, 0.4120, 0.8013, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895],
  9932. [0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950]],
  9933. device='cuda:0', dtype=torch.float64)
  9934. predictions are: tensor([[0.5812, 0.3663, 0.8780, 0.5663, 0.4113, 0.5094, 0.6484, 0.5449],
  9935. [0.6205, 0.3792, 0.8964, 0.5088, 0.3694, 0.5186, 0.6588, 0.5418],
  9936. [0.5394, 0.3596, 0.8567, 0.5064, 0.3538, 0.4711, 0.5429, 0.5481],
  9937. [0.5951, 0.3821, 0.9107, 0.5731, 0.4070, 0.5004, 0.5472, 0.5524],
  9938. [0.6014, 0.3876, 0.7242, 0.2879, 0.4193, 0.1686, 0.5552, 0.5635],
  9939. [0.6758, 0.4629, 0.8608, 0.2929, 0.4783, 0.1708, 0.6264, 0.5358],
  9940. [0.7078, 0.4596, 0.8143, 0.2689, 0.4817, 0.1511, 0.6334, 0.5194],
  9941. [0.6756, 0.4203, 0.8839, 0.5290, 0.4115, 0.5091, 0.5758, 0.5183]],
  9942. device='cuda:0', grad_fn=<AddmmBackward>)
  9943. landmarks are: tensor([[[0.6257, 0.4024, 0.8672, 0.5422, 0.4196, 0.5198, 0.6694, 0.5204],
  9944. [0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166],
  9945. [0.6038, 0.3946, 0.8413, 0.4883, 0.3562, 0.4550, 0.5266, 0.4693],
  9946. [0.6136, 0.4117, 0.8700, 0.5167, 0.4187, 0.5083, 0.5147, 0.5495],
  9947. [0.6250, 0.4110, 0.7237, 0.2067, 0.4263, 0.1883, 0.5625, 0.5633],
  9948. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  9949. [0.6259, 0.4120, 0.8012, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895],
  9950. [0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950]]],
  9951. device='cuda:0')
  9952. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  9953. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  9954. loss_train: 0.03298321762122214
  9955. step: 20
  9956. running loss: 0.0016491608810611069
  9957. Train Steps: 20/90 Loss: 0.0016 torch.Size([8, 600, 800])
  9958. torch.Size([8, 8])
  9959. tensor([[ nan, nan, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  9960. [0.6097, 0.3988, 0.8650, 0.5250, 0.4213, 0.5200, 0.5675, 0.5050],
  9961. [0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  9962. [0.6152, 0.4131, 0.6863, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  9963. [0.6267, 0.4065, 0.8313, 0.2467, 0.4788, 0.1733, 0.6312, 0.5133],
  9964. [0.6205, 0.4016, 0.8350, 0.2717, 0.3987, 0.2550, 0.5787, 0.5133],
  9965. [0.6064, 0.4019, 0.8650, 0.4517, 0.4037, 0.5367, 0.5703, 0.5609],
  9966. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374]],
  9967. device='cuda:0', dtype=torch.float64)
  9968. predictions are: tensor([[0.1661, 0.0989, 0.8938, 0.3742, 0.4654, 0.1780, 0.6484, 0.5681],
  9969. [0.6539, 0.4123, 0.8830, 0.5821, 0.4425, 0.5081, 0.5056, 0.5100],
  9970. [0.6937, 0.4245, 0.8564, 0.6342, 0.4042, 0.4477, 0.6047, 0.4869],
  9971. [0.7473, 0.4797, 0.6982, 0.3150, 0.3798, 0.3226, 0.5214, 0.5388],
  9972. [0.6735, 0.4346, 0.8389, 0.2927, 0.4800, 0.1470, 0.5898, 0.5247],
  9973. [0.6678, 0.4201, 0.8428, 0.2941, 0.4032, 0.2558, 0.5571, 0.5247],
  9974. [0.6108, 0.3926, 0.8709, 0.4782, 0.4101, 0.5413, 0.5579, 0.5436],
  9975. [0.6580, 0.4124, 0.8757, 0.4204, 0.3782, 0.2969, 0.5750, 0.5194]],
  9976. device='cuda:0', grad_fn=<AddmmBackward>)
  9977. landmarks are: tensor([[[0.0000, 0.0000, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  9978. [0.6097, 0.3988, 0.8650, 0.5250, 0.4212, 0.5200, 0.5675, 0.5050],
  9979. [0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  9980. [0.6152, 0.4131, 0.6862, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  9981. [0.6266, 0.4065, 0.8313, 0.2467, 0.4787, 0.1733, 0.6313, 0.5133],
  9982. [0.6205, 0.4015, 0.8350, 0.2717, 0.3988, 0.2550, 0.5788, 0.5133],
  9983. [0.6064, 0.4019, 0.8650, 0.4517, 0.4038, 0.5367, 0.5703, 0.5609],
  9984. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374]]],
  9985. device='cuda:0')
  9986. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  9987. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  9988. loss_train: 0.034703588928095996
  9989. step: 21
  9990. running loss: 0.001652551853718857
  9991. Train Steps: 21/90 Loss: 0.0017 torch.Size([8, 600, 800])
  9992. torch.Size([8, 8])
  9993. tensor([[0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  9994. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  9995. [0.6148, 0.4076, 0.8666, 0.4820, 0.4138, 0.5067, 0.5250, 0.5767],
  9996. [0.6250, 0.4103, 0.8950, 0.4400, 0.3912, 0.5650, 0.6050, 0.5133],
  9997. [0.6057, 0.4011, 0.8750, 0.4267, 0.4400, 0.5800, 0.5845, 0.5585],
  9998. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  9999. [0.6201, 0.4082, 0.8827, 0.3715, 0.3825, 0.2712, 0.5845, 0.5412],
  10000. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986]],
  10001. device='cuda:0', dtype=torch.float64)
  10002. predictions are: tensor([[0.6097, 0.3810, 0.8444, 0.3163, 0.4960, 0.2205, 0.6758, 0.5090],
  10003. [0.6230, 0.4154, 0.7445, 0.3101, 0.4256, 0.1573, 0.5326, 0.5013],
  10004. [0.6116, 0.4013, 0.8533, 0.5269, 0.4231, 0.5081, 0.5280, 0.5487],
  10005. [0.5224, 0.3296, 0.8872, 0.4637, 0.4107, 0.5685, 0.5906, 0.5198],
  10006. [0.4906, 0.3321, 0.8419, 0.4613, 0.4652, 0.5174, 0.5737, 0.5340],
  10007. [0.6110, 0.3884, 0.8540, 0.6034, 0.4019, 0.4349, 0.5250, 0.5502],
  10008. [0.6076, 0.3926, 0.8540, 0.3914, 0.3840, 0.2692, 0.5650, 0.5152],
  10009. [0.6196, 0.4102, 0.8748, 0.4050, 0.4474, 0.2056, 0.5809, 0.5021]],
  10010. device='cuda:0', grad_fn=<AddmmBackward>)
  10011. landmarks are: tensor([[[0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  10012. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  10013. [0.6148, 0.4076, 0.8666, 0.4820, 0.4137, 0.5067, 0.5250, 0.5767],
  10014. [0.6250, 0.4103, 0.8950, 0.4400, 0.3913, 0.5650, 0.6050, 0.5133],
  10015. [0.6057, 0.4011, 0.8750, 0.4267, 0.4400, 0.5800, 0.5845, 0.5585],
  10016. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  10017. [0.6201, 0.4082, 0.8827, 0.3715, 0.3825, 0.2712, 0.5845, 0.5412],
  10018. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986]]],
  10019. device='cuda:0')
  10020. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  10021. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  10022. loss_train: 0.03575298085343093
  10023. step: 22
  10024. running loss: 0.0016251354933377695
  10025.  
  10026. Train Steps: 22/90 Loss: 0.0016 torch.Size([8, 600, 800])
  10027. torch.Size([8, 8])
  10028. tensor([[0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  10029. [0.6205, 0.4062, 0.8337, 0.2683, 0.3675, 0.4283, 0.6338, 0.5250],
  10030. [0.6310, 0.4017, 0.8563, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006],
  10031. [0.6179, 0.4008, 0.8600, 0.4015, 0.3932, 0.2515, 0.5711, 0.5438],
  10032. [0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  10033. [0.6226, 0.4098, 0.8912, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  10034. [0.6148, 0.4053, 0.8750, 0.4550, 0.4850, 0.5218, 0.5863, 0.5567],
  10035. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750]],
  10036. device='cuda:0', dtype=torch.float64)
  10037. predictions are: tensor([[0.5948, 0.3777, 0.8598, 0.5559, 0.4323, 0.4669, 0.5633, 0.5099],
  10038. [0.5748, 0.3623, 0.8276, 0.3119, 0.4020, 0.3961, 0.6031, 0.4939],
  10039. [0.5953, 0.3702, 0.8583, 0.6192, 0.4189, 0.4924, 0.5780, 0.4961],
  10040. [0.5731, 0.3803, 0.8354, 0.3893, 0.4161, 0.2656, 0.5426, 0.5256],
  10041. [0.6213, 0.3960, 0.8607, 0.4340, 0.3896, 0.4437, 0.5424, 0.4852],
  10042. [0.5919, 0.3985, 0.8903, 0.4316, 0.4365, 0.2533, 0.5274, 0.5285],
  10043. [0.5917, 0.3857, 0.8968, 0.4565, 0.4929, 0.4819, 0.5501, 0.5295],
  10044. [0.6593, 0.4371, 0.7370, 0.2946, 0.4135, 0.2788, 0.5746, 0.5380]],
  10045. device='cuda:0', grad_fn=<AddmmBackward>)
  10046. landmarks are: tensor([[[0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  10047. [0.6205, 0.4062, 0.8338, 0.2683, 0.3675, 0.4283, 0.6338, 0.5250],
  10048. [0.6310, 0.4017, 0.8562, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006],
  10049. [0.6179, 0.4008, 0.8600, 0.4015, 0.3932, 0.2515, 0.5711, 0.5438],
  10050. [0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  10051. [0.6226, 0.4098, 0.8913, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  10052. [0.6148, 0.4053, 0.8750, 0.4550, 0.4850, 0.5218, 0.5863, 0.5567],
  10053. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750]]],
  10054. device='cuda:0')
  10055. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  10056. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  10057. loss_train: 0.03661173034925014
  10058. step: 23
  10059. running loss: 0.0015918143630108755
  10060. Train Steps: 23/90 Loss: 0.0016 torch.Size([8, 600, 800])
  10061. torch.Size([8, 8])
  10062. tensor([[0.6132, 0.4037, 0.6963, 0.2217, 0.4100, 0.1950, 0.5395, 0.5175],
  10063. [0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038],
  10064. [0.6260, 0.4214, 0.8538, 0.5500, 0.3663, 0.3767, 0.5587, 0.5983],
  10065. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  10066. [ nan, nan, 0.7725, 0.2611, 0.3675, 0.2733, 0.5413, 0.5167],
  10067. [0.6200, 0.4070, 0.8938, 0.4183, 0.3538, 0.4567, 0.6175, 0.5400],
  10068. [0.6289, 0.4019, 0.8113, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332],
  10069. [0.6274, 0.4003, 0.8638, 0.5967, 0.3688, 0.4900, 0.6108, 0.4661]],
  10070. device='cuda:0', dtype=torch.float64)
  10071. predictions are: tensor([[0.6434, 0.4394, 0.7563, 0.2616, 0.4180, 0.1771, 0.5199, 0.4975],
  10072. [0.6028, 0.4039, 0.8736, 0.4948, 0.4675, 0.5166, 0.5356, 0.5194],
  10073. [0.6994, 0.4571, 0.8777, 0.5358, 0.4021, 0.4242, 0.5314, 0.5979],
  10074. [0.6870, 0.4445, 0.9097, 0.3819, 0.3974, 0.4629, 0.6863, 0.5208],
  10075. [0.1170, 0.0916, 0.7732, 0.2960, 0.3928, 0.2935, 0.5456, 0.5585],
  10076. [0.7068, 0.4634, 0.9245, 0.4283, 0.4178, 0.5128, 0.6311, 0.5296],
  10077. [0.6364, 0.4310, 0.8491, 0.5356, 0.3984, 0.5179, 0.6621, 0.5221],
  10078. [0.7051, 0.4599, 0.8635, 0.5810, 0.4164, 0.4868, 0.6125, 0.5172]],
  10079. device='cuda:0', grad_fn=<AddmmBackward>)
  10080. landmarks are: tensor([[[0.6132, 0.4037, 0.6963, 0.2217, 0.4100, 0.1950, 0.5395, 0.5175],
  10081. [0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038],
  10082. [0.6260, 0.4214, 0.8537, 0.5500, 0.3663, 0.3767, 0.5587, 0.5983],
  10083. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  10084. [0.0000, 0.0000, 0.7725, 0.2611, 0.3675, 0.2733, 0.5412, 0.5167],
  10085. [0.6200, 0.4070, 0.8938, 0.4183, 0.3537, 0.4567, 0.6175, 0.5400],
  10086. [0.6289, 0.4019, 0.8112, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332],
  10087. [0.6274, 0.4003, 0.8637, 0.5967, 0.3688, 0.4900, 0.6108, 0.4661]]],
  10088. device='cuda:0')
  10089. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  10090. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  10091. loss_train: 0.03813450282905251
  10092. step: 24
  10093. running loss: 0.001588937617877188
  10094. Train Steps: 24/90 Loss: 0.0016 torch.Size([8, 600, 800])
  10095. torch.Size([8, 8])
  10096. tensor([[0.6198, 0.4115, 0.7762, 0.2717, 0.3713, 0.3200, 0.5837, 0.5683],
  10097. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  10098. [0.6353, 0.4128, 0.9138, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  10099. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  10100. [0.6185, 0.4079, 0.8838, 0.4617, 0.4838, 0.5650, 0.6175, 0.5850],
  10101. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683],
  10102. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  10103. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167]],
  10104. device='cuda:0', dtype=torch.float64)
  10105. predictions are: tensor([[0.5181, 0.3658, 0.7687, 0.2471, 0.3332, 0.3298, 0.5577, 0.5300],
  10106. [0.5627, 0.3805, 0.8305, 0.2223, 0.4413, 0.2407, 0.6896, 0.5307],
  10107. [0.4681, 0.3387, 0.8765, 0.3318, 0.4165, 0.3352, 0.7017, 0.5461],
  10108. [0.5176, 0.3823, 0.8142, 0.4260, 0.4169, 0.3081, 0.5115, 0.5719],
  10109. [0.5179, 0.3504, 0.8527, 0.4173, 0.4445, 0.5811, 0.5924, 0.5427],
  10110. [0.5762, 0.4029, 0.8374, 0.5142, 0.3543, 0.3983, 0.5364, 0.5255],
  10111. [0.5836, 0.3958, 0.8593, 0.4602, 0.3841, 0.5372, 0.6346, 0.4635],
  10112. [0.5452, 0.3705, 0.7265, 0.3065, 0.4365, 0.2370, 0.5239, 0.5725]],
  10113. device='cuda:0', grad_fn=<AddmmBackward>)
  10114. landmarks are: tensor([[[0.6198, 0.4115, 0.7763, 0.2717, 0.3713, 0.3200, 0.5838, 0.5683],
  10115. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  10116. [0.6353, 0.4128, 0.9137, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  10117. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  10118. [0.6184, 0.4079, 0.8838, 0.4617, 0.4837, 0.5650, 0.6175, 0.5850],
  10119. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683],
  10120. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  10121. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167]]],
  10122. device='cuda:0')
  10123. loss_train_step before backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  10124. loss_train_step after backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  10125. loss_train: 0.040397314936853945
  10126. step: 25
  10127. running loss: 0.0016158925974741579
  10128. Train Steps: 25/90 Loss: 0.0016 torch.Size([8, 600, 800])
  10129. torch.Size([8, 8])
  10130. tensor([[ nan, nan, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  10131. [0.6228, 0.4119, 0.7938, 0.2233, 0.4674, 0.1773, 0.6188, 0.5433],
  10132. [0.6132, 0.4118, 0.8200, 0.3633, 0.3563, 0.5400, 0.5787, 0.5136],
  10133. [0.6251, 0.4163, 0.8662, 0.4467, 0.3625, 0.3567, 0.6038, 0.5533],
  10134. [0.6185, 0.4098, 0.8838, 0.4900, 0.4537, 0.5800, 0.6288, 0.5400],
  10135. [0.6202, 0.4064, 0.7879, 0.2179, 0.4567, 0.1725, 0.5955, 0.5478],
  10136. [0.6278, 0.4253, 0.8875, 0.5017, 0.4113, 0.2750, 0.5413, 0.6196],
  10137. [0.6145, 0.4008, 0.8750, 0.5383, 0.3975, 0.4650, 0.5563, 0.5533]],
  10138. device='cuda:0', dtype=torch.float64)
  10139. predictions are: tensor([[0.0116, 0.0246, 0.7393, 0.2538, 0.3489, 0.2734, 0.5412, 0.5697],
  10140. [0.6360, 0.4286, 0.8073, 0.2053, 0.4622, 0.1909, 0.6346, 0.5329],
  10141. [0.5926, 0.4112, 0.8407, 0.3210, 0.3518, 0.5539, 0.6553, 0.5225],
  10142. [0.6690, 0.4513, 0.8514, 0.4062, 0.3517, 0.3713, 0.6100, 0.5537],
  10143. [0.6008, 0.3862, 0.8725, 0.4734, 0.4421, 0.6024, 0.6467, 0.5385],
  10144. [0.5480, 0.3798, 0.7652, 0.2324, 0.4457, 0.1914, 0.6148, 0.5599],
  10145. [0.6214, 0.4346, 0.8467, 0.4490, 0.4235, 0.3341, 0.5924, 0.6016],
  10146. [0.6507, 0.4407, 0.8519, 0.5101, 0.3846, 0.4925, 0.6274, 0.5505]],
  10147. device='cuda:0', grad_fn=<AddmmBackward>)
  10148. landmarks are: tensor([[[0.0000, 0.0000, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  10149. [0.6228, 0.4119, 0.7937, 0.2233, 0.4674, 0.1773, 0.6187, 0.5433],
  10150. [0.6132, 0.4118, 0.8200, 0.3633, 0.3562, 0.5400, 0.5787, 0.5136],
  10151. [0.6252, 0.4162, 0.8662, 0.4467, 0.3625, 0.3567, 0.6037, 0.5533],
  10152. [0.6185, 0.4098, 0.8838, 0.4900, 0.4538, 0.5800, 0.6288, 0.5400],
  10153. [0.6202, 0.4064, 0.7879, 0.2179, 0.4567, 0.1725, 0.5955, 0.5478],
  10154. [0.6278, 0.4253, 0.8875, 0.5017, 0.4112, 0.2750, 0.5413, 0.6196],
  10155. [0.6145, 0.4008, 0.8750, 0.5383, 0.3975, 0.4650, 0.5562, 0.5533]]],
  10156. device='cuda:0')
  10157. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  10158. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  10159. loss_train: 0.04119387164246291
  10160. step: 26
  10161. running loss: 0.0015843796785562658
  10162.  
  10163. Train Steps: 26/90 Loss: 0.0016 torch.Size([8, 600, 800])
  10164. torch.Size([8, 8])
  10165. tensor([[0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  10166. [0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500],
  10167. [0.6277, 0.4083, 0.8350, 0.2717, 0.4562, 0.1800, 0.5918, 0.4878],
  10168. [0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050],
  10169. [ nan, nan, 0.8488, 0.2300, 0.5563, 0.2100, 0.7390, 0.5679],
  10170. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  10171. [ nan, nan, 0.7097, 0.2346, 0.4250, 0.1850, 0.5175, 0.5583],
  10172. [0.6263, 0.4030, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771]],
  10173. device='cuda:0', dtype=torch.float64)
  10174. predictions are: tensor([[0.7681, 0.5108, 0.8592, 0.4223, 0.3623, 0.4802, 0.6189, 0.5313],
  10175. [0.7198, 0.4836, 0.8040, 0.2694, 0.3555, 0.4219, 0.6543, 0.5635],
  10176. [0.5966, 0.3962, 0.8460, 0.2799, 0.4625, 0.2163, 0.6013, 0.5294],
  10177. [0.6678, 0.4680, 0.8998, 0.4368, 0.3742, 0.5103, 0.6602, 0.6020],
  10178. [0.1118, 0.0834, 0.8835, 0.2769, 0.5051, 0.2371, 0.7420, 0.5724],
  10179. [0.6195, 0.4208, 0.7383, 0.3169, 0.3674, 0.2995, 0.5883, 0.6082],
  10180. [0.1565, 0.1337, 0.7329, 0.2517, 0.4242, 0.1904, 0.5244, 0.5901],
  10181. [0.7288, 0.4908, 0.9129, 0.4972, 0.3733, 0.5528, 0.6372, 0.5112]],
  10182. device='cuda:0', grad_fn=<AddmmBackward>)
  10183. landmarks are: tensor([[[0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  10184. [0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500],
  10185. [0.6277, 0.4083, 0.8350, 0.2717, 0.4563, 0.1800, 0.5918, 0.4878],
  10186. [0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050],
  10187. [0.0000, 0.0000, 0.8487, 0.2300, 0.5562, 0.2100, 0.7390, 0.5679],
  10188. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  10189. [0.0000, 0.0000, 0.7097, 0.2346, 0.4250, 0.1850, 0.5175, 0.5583],
  10190. [0.6263, 0.4029, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771]]],
  10191. device='cuda:0')
  10192. loss_train_step before backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  10193. loss_train_step after backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  10194. loss_train: 0.043881157762371004
  10195. step: 27
  10196. running loss: 0.0016252280652730002
  10197. Train Steps: 27/90 Loss: 0.0016 torch.Size([8, 600, 800])
  10198. torch.Size([8, 8])
  10199. tensor([[0.6092, 0.4001, 0.8638, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  10200. [0.6304, 0.4024, 0.8925, 0.4800, 0.3937, 0.4817, 0.7485, 0.5297],
  10201. [0.6311, 0.3998, 0.7975, 0.5767, 0.3838, 0.4850, 0.7327, 0.5343],
  10202. [0.6263, 0.4030, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  10203. [0.6187, 0.4104, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  10204. [0.6102, 0.4001, 0.7738, 0.3583, 0.3463, 0.3800, 0.5524, 0.5689],
  10205. [0.6243, 0.4128, 0.7762, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417],
  10206. [0.6339, 0.4102, 0.9088, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390]],
  10207. device='cuda:0', dtype=torch.float64)
  10208. predictions are: tensor([[0.5978, 0.3811, 0.8802, 0.4775, 0.4360, 0.5232, 0.5414, 0.5480],
  10209. [0.6242, 0.3893, 0.8940, 0.4716, 0.3845, 0.4669, 0.7091, 0.5600],
  10210. [0.5777, 0.3764, 0.8253, 0.5264, 0.3796, 0.4584, 0.6991, 0.5487],
  10211. [0.5385, 0.3593, 0.9121, 0.4667, 0.3836, 0.5002, 0.6042, 0.5307],
  10212. [0.5272, 0.3532, 0.7427, 0.2166, 0.3878, 0.2256, 0.5639, 0.5856],
  10213. [0.5912, 0.3912, 0.8115, 0.3456, 0.3531, 0.3754, 0.5573, 0.6048],
  10214. [0.5342, 0.3543, 0.8327, 0.2611, 0.4037, 0.2601, 0.6445, 0.5905],
  10215. [0.5617, 0.3802, 0.9341, 0.4721, 0.4119, 0.5319, 0.7273, 0.6080]],
  10216. device='cuda:0', grad_fn=<AddmmBackward>)
  10217. landmarks are: tensor([[[0.6092, 0.4001, 0.8637, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  10218. [0.6304, 0.4024, 0.8925, 0.4800, 0.3938, 0.4817, 0.7485, 0.5297],
  10219. [0.6311, 0.3998, 0.7975, 0.5767, 0.3837, 0.4850, 0.7327, 0.5343],
  10220. [0.6263, 0.4029, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  10221. [0.6187, 0.4103, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  10222. [0.6102, 0.4001, 0.7738, 0.3583, 0.3462, 0.3800, 0.5524, 0.5689],
  10223. [0.6243, 0.4128, 0.7763, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417],
  10224. [0.6339, 0.4102, 0.9087, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390]]],
  10225. device='cuda:0')
  10226. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  10227. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  10228. loss_train: 0.04513794963713735
  10229. step: 28
  10230. running loss: 0.0016120696298977627
  10231. Train Steps: 28/90 Loss: 0.0016 torch.Size([8, 600, 800])
  10232. torch.Size([8, 8])
  10233. tensor([[0.6055, 0.4015, 0.7425, 0.2033, 0.4113, 0.1883, 0.5217, 0.4823],
  10234. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600],
  10235. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  10236. [0.6219, 0.3934, 0.8688, 0.5267, 0.4313, 0.4967, 0.5988, 0.4983],
  10237. [0.6267, 0.4080, 0.8438, 0.2633, 0.4763, 0.1800, 0.6259, 0.5240],
  10238. [0.6198, 0.4130, 0.8762, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  10239. [0.6131, 0.4037, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  10240. [0.6075, 0.4007, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280]],
  10241. device='cuda:0', dtype=torch.float64)
  10242. predictions are: tensor([[0.4372, 0.2758, 0.7354, 0.1883, 0.3909, 0.1579, 0.5633, 0.5465],
  10243. [0.6577, 0.4155, 0.8830, 0.5415, 0.3977, 0.4813, 0.6043, 0.5893],
  10244. [0.5946, 0.3737, 0.8991, 0.3180, 0.4474, 0.3089, 0.7670, 0.5524],
  10245. [0.5719, 0.3405, 0.8579, 0.5194, 0.4275, 0.4948, 0.6411, 0.5257],
  10246. [0.4898, 0.3002, 0.8607, 0.2459, 0.4668, 0.1549, 0.6598, 0.5464],
  10247. [0.5073, 0.3276, 0.8746, 0.3994, 0.3560, 0.4804, 0.6034, 0.5596],
  10248. [0.5150, 0.3232, 0.6938, 0.2471, 0.3633, 0.2815, 0.5660, 0.5930],
  10249. [0.5969, 0.3752, 0.8305, 0.4599, 0.3981, 0.4965, 0.5806, 0.5713]],
  10250. device='cuda:0', grad_fn=<AddmmBackward>)
  10251. landmarks are: tensor([[[0.6055, 0.4015, 0.7425, 0.2033, 0.4112, 0.1883, 0.5217, 0.4823],
  10252. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600],
  10253. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  10254. [0.6219, 0.3934, 0.8687, 0.5267, 0.4313, 0.4967, 0.5987, 0.4983],
  10255. [0.6267, 0.4080, 0.8438, 0.2633, 0.4762, 0.1800, 0.6259, 0.5240],
  10256. [0.6198, 0.4130, 0.8763, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  10257. [0.6131, 0.4036, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  10258. [0.6075, 0.4006, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280]]],
  10259. device='cuda:0')
  10260. loss_train_step before backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  10261. loss_train_step after backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  10262. loss_train: 0.04752955154981464
  10263. step: 29
  10264. running loss: 0.0016389500534418843
  10265. Train Steps: 29/90 Loss: 0.0016 torch.Size([8, 600, 800])
  10266. torch.Size([8, 8])
  10267. tensor([[0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  10268. [0.6190, 0.4135, 0.8000, 0.4883, 0.3566, 0.3647, 0.5613, 0.5900],
  10269. [0.6353, 0.4128, 0.8488, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  10270. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  10271. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  10272. [0.6133, 0.4066, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358],
  10273. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395],
  10274. [0.6311, 0.4008, 0.7935, 0.5746, 0.3900, 0.5033, 0.6955, 0.5366]],
  10275. device='cuda:0', dtype=torch.float64)
  10276. predictions are: tensor([[0.5865, 0.3450, 0.7975, 0.2635, 0.4103, 0.2449, 0.6346, 0.5429],
  10277. [0.6215, 0.3994, 0.8307, 0.4658, 0.3353, 0.3533, 0.5554, 0.5437],
  10278. [0.2566, 0.1509, 0.8737, 0.2176, 0.5307, 0.1564, 0.7105, 0.5645],
  10279. [0.6468, 0.3995, 0.8455, 0.4233, 0.3543, 0.4662, 0.5918, 0.5448],
  10280. [0.5884, 0.3784, 0.8735, 0.4595, 0.4237, 0.4856, 0.5844, 0.5492],
  10281. [0.5698, 0.3733, 0.7134, 0.2465, 0.3807, 0.2194, 0.5472, 0.5236],
  10282. [0.6155, 0.3682, 0.7660, 0.2156, 0.3963, 0.2318, 0.6159, 0.5211],
  10283. [0.6217, 0.3869, 0.8140, 0.5425, 0.3716, 0.5056, 0.6957, 0.5169]],
  10284. device='cuda:0', grad_fn=<AddmmBackward>)
  10285. landmarks are: tensor([[[0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  10286. [0.6190, 0.4135, 0.8000, 0.4883, 0.3566, 0.3647, 0.5612, 0.5900],
  10287. [0.6353, 0.4128, 0.8487, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  10288. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  10289. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  10290. [0.6133, 0.4065, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358],
  10291. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395],
  10292. [0.6311, 0.4008, 0.7935, 0.5746, 0.3900, 0.5033, 0.6955, 0.5366]]],
  10293. device='cuda:0')
  10294. loss_train_step before backward: tensor(0.0038, device='cuda:0', grad_fn=<MseLossBackward>)
  10295. loss_train_step after backward: tensor(0.0038, device='cuda:0', grad_fn=<MseLossBackward>)
  10296. loss_train: 0.05135433666873723
  10297. step: 30
  10298. running loss: 0.0017118112222912411
  10299.  
  10300. Train Steps: 30/90 Loss: 0.0017 torch.Size([8, 600, 800])
  10301. torch.Size([8, 8])
  10302. tensor([[0.6212, 0.4033, 0.8938, 0.4167, 0.3813, 0.4267, 0.5613, 0.5583],
  10303. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  10304. [0.6261, 0.4066, 0.8325, 0.2150, 0.4763, 0.2667, 0.7002, 0.5633],
  10305. [0.6333, 0.4037, 0.8638, 0.5733, 0.4012, 0.4717, 0.6369, 0.4938],
  10306. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  10307. [0.6200, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  10308. [ nan, nan, 0.6900, 0.1917, 0.3937, 0.2367, 0.5240, 0.5246],
  10309. [0.6102, 0.4005, 0.8688, 0.5100, 0.4813, 0.5400, 0.5404, 0.5064]],
  10310. device='cuda:0', dtype=torch.float64)
  10311. predictions are: tensor([[0.5668, 0.3629, 0.8842, 0.4258, 0.3682, 0.4059, 0.5781, 0.5503],
  10312. [0.6187, 0.3936, 0.7304, 0.2942, 0.4803, 0.2003, 0.5441, 0.5974],
  10313. [0.6314, 0.3861, 0.8064, 0.1883, 0.4717, 0.2359, 0.7089, 0.5526],
  10314. [0.6939, 0.4356, 0.8328, 0.5578, 0.3852, 0.4608, 0.6231, 0.4841],
  10315. [0.6649, 0.4234, 0.8089, 0.5228, 0.3513, 0.4833, 0.6610, 0.4989],
  10316. [0.6308, 0.4149, 0.8182, 0.2647, 0.4233, 0.2208, 0.5827, 0.5212],
  10317. [0.1004, 0.0594, 0.6898, 0.1966, 0.4079, 0.2105, 0.5226, 0.5340],
  10318. [0.6396, 0.4302, 0.8272, 0.4877, 0.4743, 0.4958, 0.5450, 0.5044]],
  10319. device='cuda:0', grad_fn=<AddmmBackward>)
  10320. landmarks are: tensor([[[0.6212, 0.4033, 0.8938, 0.4167, 0.3812, 0.4267, 0.5612, 0.5583],
  10321. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  10322. [0.6261, 0.4066, 0.8325, 0.2150, 0.4762, 0.2667, 0.7002, 0.5633],
  10323. [0.6334, 0.4037, 0.8637, 0.5733, 0.4013, 0.4717, 0.6369, 0.4938],
  10324. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  10325. [0.6199, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  10326. [0.0000, 0.0000, 0.6900, 0.1917, 0.3938, 0.2367, 0.5240, 0.5246],
  10327. [0.6102, 0.4005, 0.8687, 0.5100, 0.4812, 0.5400, 0.5404, 0.5064]]],
  10328. device='cuda:0')
  10329. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  10330. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  10331. loss_train: 0.052187160414177924
  10332. step: 31
  10333. running loss: 0.0016834567875541266
  10334. Train Steps: 31/90 Loss: 0.0017 torch.Size([8, 600, 800])
  10335. torch.Size([8, 8])
  10336. tensor([[0.6203, 0.4096, 0.8862, 0.4267, 0.3538, 0.4117, 0.6025, 0.5650],
  10337. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  10338. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5363, 0.5550],
  10339. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  10340. [0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  10341. [0.6179, 0.4118, 0.7278, 0.4237, 0.3588, 0.3400, 0.5675, 0.5917],
  10342. [0.6082, 0.4024, 0.8738, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  10343. [ nan, nan, 0.8300, 0.3150, 0.3588, 0.3383, 0.5208, 0.5194]],
  10344. device='cuda:0', dtype=torch.float64)
  10345. predictions are: tensor([[0.6223, 0.4037, 0.8574, 0.4101, 0.3940, 0.3909, 0.6264, 0.5640],
  10346. [0.6208, 0.4085, 0.8440, 0.4685, 0.4030, 0.3002, 0.5461, 0.5567],
  10347. [0.6917, 0.4606, 0.6627, 0.2580, 0.4542, 0.2239, 0.5561, 0.5403],
  10348. [0.6006, 0.3673, 0.7958, 0.3297, 0.3836, 0.3588, 0.6420, 0.5311],
  10349. [0.6163, 0.3988, 0.8610, 0.4075, 0.4042, 0.4429, 0.6372, 0.4826],
  10350. [0.6346, 0.4224, 0.7407, 0.3825, 0.3826, 0.3340, 0.5764, 0.5770],
  10351. [0.5733, 0.3838, 0.8238, 0.4076, 0.3889, 0.3602, 0.5383, 0.4948],
  10352. [0.4540, 0.2950, 0.7634, 0.2964, 0.3890, 0.3205, 0.5472, 0.5035]],
  10353. device='cuda:0', grad_fn=<AddmmBackward>)
  10354. landmarks are: tensor([[[0.6203, 0.4096, 0.8863, 0.4267, 0.3537, 0.4117, 0.6025, 0.5650],
  10355. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  10356. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5362, 0.5550],
  10357. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  10358. [0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  10359. [0.6179, 0.4118, 0.7278, 0.4237, 0.3587, 0.3400, 0.5675, 0.5917],
  10360. [0.6082, 0.4024, 0.8737, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  10361. [0.0000, 0.0000, 0.8300, 0.3150, 0.3587, 0.3383, 0.5208, 0.5194]]],
  10362. device='cuda:0')
  10363. loss_train_step before backward: tensor(0.0053, device='cuda:0', grad_fn=<MseLossBackward>)
  10364. loss_train_step after backward: tensor(0.0053, device='cuda:0', grad_fn=<MseLossBackward>)
  10365. loss_train: 0.05747255013557151
  10366. step: 32
  10367. running loss: 0.0017960171917366097
  10368. Train Steps: 32/90 Loss: 0.0018 torch.Size([8, 600, 800])
  10369. torch.Size([8, 8])
  10370. tensor([[0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896],
  10371. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  10372. [0.6275, 0.4157, 0.8337, 0.5800, 0.3763, 0.4200, 0.5547, 0.6125],
  10373. [0.6129, 0.4069, 0.8750, 0.5067, 0.3875, 0.4233, 0.5235, 0.5881],
  10374. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  10375. [0.6276, 0.4095, 0.8237, 0.2250, 0.4662, 0.1783, 0.6171, 0.4869],
  10376. [0.6156, 0.4125, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  10377. [0.6110, 0.3984, 0.8750, 0.4933, 0.4625, 0.4950, 0.5578, 0.5676]],
  10378. device='cuda:0', dtype=torch.float64)
  10379. predictions are: tensor([[0.6144, 0.4203, 0.7985, 0.3863, 0.3475, 0.2884, 0.5326, 0.5545],
  10380. [0.5494, 0.3584, 0.7314, 0.2205, 0.3959, 0.2670, 0.5991, 0.5164],
  10381. [0.6711, 0.4399, 0.7896, 0.5625, 0.3788, 0.4291, 0.5623, 0.5686],
  10382. [0.5554, 0.3771, 0.8378, 0.5346, 0.3878, 0.4062, 0.5128, 0.5303],
  10383. [0.6204, 0.4052, 0.8259, 0.4745, 0.4359, 0.4610, 0.5449, 0.5183],
  10384. [0.6332, 0.4169, 0.7866, 0.2161, 0.5003, 0.2074, 0.6308, 0.4744],
  10385. [0.5956, 0.4142, 0.8529, 0.4707, 0.4455, 0.5582, 0.5868, 0.5015],
  10386. [0.6420, 0.4306, 0.8381, 0.4839, 0.4555, 0.4854, 0.5397, 0.5353]],
  10387. device='cuda:0', grad_fn=<AddmmBackward>)
  10388. landmarks are: tensor([[[0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896],
  10389. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  10390. [0.6275, 0.4157, 0.8338, 0.5800, 0.3762, 0.4200, 0.5547, 0.6125],
  10391. [0.6129, 0.4069, 0.8750, 0.5067, 0.3875, 0.4233, 0.5235, 0.5881],
  10392. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  10393. [0.6276, 0.4095, 0.8238, 0.2250, 0.4663, 0.1783, 0.6171, 0.4869],
  10394. [0.6155, 0.4124, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  10395. [0.6110, 0.3984, 0.8750, 0.4933, 0.4625, 0.4950, 0.5578, 0.5676]]],
  10396. device='cuda:0')
  10397. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  10398. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  10399. loss_train: 0.058107397344429046
  10400. step: 33
  10401. running loss: 0.001760830222558456
  10402. Train Steps: 33/90 Loss: 0.0018 torch.Size([8, 600, 800])
  10403. torch.Size([8, 8])
  10404. tensor([[0.6190, 0.4135, 0.8000, 0.4883, 0.3566, 0.3647, 0.5613, 0.5900],
  10405. [0.6229, 0.4066, 0.8513, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  10406. [ nan, nan, 0.7515, 0.2708, 0.3987, 0.2267, 0.5162, 0.5567],
  10407. [0.6200, 0.4049, 0.8638, 0.5617, 0.4125, 0.5100, 0.6013, 0.5317],
  10408. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  10409. [0.6030, 0.3969, 0.7988, 0.3917, 0.3450, 0.3667, 0.5266, 0.4700],
  10410. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  10411. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153]],
  10412. device='cuda:0', dtype=torch.float64)
  10413. predictions are: tensor([[0.6827, 0.4496, 0.8256, 0.4827, 0.3709, 0.3676, 0.5322, 0.5571],
  10414. [0.6665, 0.4444, 0.8460, 0.5793, 0.4669, 0.4921, 0.5802, 0.5466],
  10415. [0.1378, 0.0869, 0.7551, 0.2594, 0.4420, 0.2424, 0.4932, 0.5484],
  10416. [0.6830, 0.4586, 0.8622, 0.5962, 0.4376, 0.4989, 0.5673, 0.5239],
  10417. [0.6869, 0.4526, 0.6950, 0.2222, 0.4388, 0.2058, 0.5477, 0.5366],
  10418. [0.6921, 0.4603, 0.8187, 0.4140, 0.3610, 0.3888, 0.5339, 0.5119],
  10419. [0.6835, 0.4319, 0.8662, 0.4696, 0.3985, 0.5319, 0.5821, 0.5164],
  10420. [0.6697, 0.4426, 0.7883, 0.3365, 0.3718, 0.3618, 0.5934, 0.5166]],
  10421. device='cuda:0', grad_fn=<AddmmBackward>)
  10422. landmarks are: tensor([[[0.6190, 0.4135, 0.8000, 0.4883, 0.3566, 0.3647, 0.5612, 0.5900],
  10423. [0.6229, 0.4066, 0.8512, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  10424. [0.0000, 0.0000, 0.7515, 0.2708, 0.3988, 0.2267, 0.5163, 0.5567],
  10425. [0.6199, 0.4049, 0.8637, 0.5617, 0.4125, 0.5100, 0.6012, 0.5317],
  10426. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  10427. [0.6030, 0.3969, 0.7987, 0.3917, 0.3450, 0.3667, 0.5266, 0.4700],
  10428. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  10429. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153]]],
  10430. device='cuda:0')
  10431. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  10432. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  10433. loss_train: 0.059539014648180455
  10434. step: 34
  10435. running loss: 0.0017511474896523664
  10436.  
  10437. Train Steps: 34/90 Loss: 0.0018 torch.Size([8, 600, 800])
  10438. torch.Size([8, 8])
  10439. tensor([[0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  10440. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  10441. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  10442. [0.6186, 0.4060, 0.8750, 0.5050, 0.3538, 0.4367, 0.5813, 0.6083],
  10443. [0.6307, 0.4029, 0.8650, 0.5200, 0.3763, 0.4017, 0.7311, 0.5366],
  10444. [0.6162, 0.4014, 0.8800, 0.5333, 0.3750, 0.4817, 0.5988, 0.5283],
  10445. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  10446. [0.6177, 0.4086, 0.8738, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700]],
  10447. device='cuda:0', dtype=torch.float64)
  10448. predictions are: tensor([[0.6581, 0.4109, 0.8762, 0.4965, 0.3920, 0.4964, 0.5997, 0.5183],
  10449. [0.5823, 0.3867, 0.8585, 0.4944, 0.4278, 0.4738, 0.5168, 0.5337],
  10450. [0.6553, 0.4376, 0.8685, 0.4777, 0.3748, 0.3175, 0.5004, 0.5822],
  10451. [0.6329, 0.4304, 0.8425, 0.5011, 0.4032, 0.4497, 0.5394, 0.5825],
  10452. [0.6444, 0.4426, 0.8537, 0.5210, 0.4073, 0.3890, 0.6372, 0.5359],
  10453. [0.6312, 0.4139, 0.8603, 0.5559, 0.4056, 0.4757, 0.5384, 0.5498],
  10454. [0.6495, 0.4212, 0.8286, 0.5655, 0.3784, 0.5000, 0.6258, 0.5397],
  10455. [0.6462, 0.4144, 0.8579, 0.4170, 0.3991, 0.5355, 0.5896, 0.5531]],
  10456. device='cuda:0', grad_fn=<AddmmBackward>)
  10457. landmarks are: tensor([[[0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  10458. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  10459. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  10460. [0.6186, 0.4060, 0.8750, 0.5050, 0.3537, 0.4367, 0.5813, 0.6083],
  10461. [0.6307, 0.4029, 0.8650, 0.5200, 0.3762, 0.4017, 0.7311, 0.5366],
  10462. [0.6162, 0.4014, 0.8800, 0.5333, 0.3750, 0.4817, 0.5987, 0.5283],
  10463. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  10464. [0.6177, 0.4085, 0.8737, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700]]],
  10465. device='cuda:0')
  10466. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  10467. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  10468. loss_train: 0.06021460593910888
  10469. step: 35
  10470. running loss: 0.001720417312545968
  10471. Train Steps: 35/90 Loss: 0.0017 torch.Size([8, 600, 800])
  10472. torch.Size([8, 8])
  10473. tensor([[0.6200, 0.4071, 0.7338, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  10474. [0.6261, 0.4066, 0.8325, 0.2150, 0.4763, 0.2667, 0.7002, 0.5633],
  10475. [0.6092, 0.4001, 0.8638, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  10476. [0.6230, 0.4113, 0.7213, 0.1983, 0.4325, 0.2367, 0.6262, 0.5400],
  10477. [0.6126, 0.4039, 0.8237, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138],
  10478. [0.6182, 0.4099, 0.7812, 0.3000, 0.3937, 0.2367, 0.5325, 0.5750],
  10479. [0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  10480. [0.6229, 0.4066, 0.8513, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350]],
  10481. device='cuda:0', dtype=torch.float64)
  10482. predictions are: tensor([[0.6218, 0.4113, 0.7499, 0.2620, 0.3945, 0.2552, 0.5913, 0.5554],
  10483. [0.6403, 0.4063, 0.8377, 0.2635, 0.4469, 0.2516, 0.6621, 0.5564],
  10484. [0.5643, 0.3632, 0.8689, 0.5466, 0.3856, 0.5353, 0.5056, 0.4891],
  10485. [0.6050, 0.4061, 0.7295, 0.2546, 0.4117, 0.2465, 0.6089, 0.5497],
  10486. [0.5580, 0.3757, 0.8355, 0.4600, 0.3196, 0.3810, 0.5580, 0.6032],
  10487. [0.5661, 0.3794, 0.7924, 0.3434, 0.3821, 0.2560, 0.4986, 0.5691],
  10488. [0.5372, 0.3464, 0.8553, 0.4697, 0.3184, 0.4471, 0.5288, 0.5201],
  10489. [0.5403, 0.3731, 0.8659, 0.6326, 0.4140, 0.4998, 0.5535, 0.5452]],
  10490. device='cuda:0', grad_fn=<AddmmBackward>)
  10491. landmarks are: tensor([[[0.6200, 0.4071, 0.7337, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  10492. [0.6261, 0.4066, 0.8325, 0.2150, 0.4762, 0.2667, 0.7002, 0.5633],
  10493. [0.6092, 0.4001, 0.8637, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  10494. [0.6230, 0.4113, 0.7212, 0.1983, 0.4325, 0.2367, 0.6263, 0.5400],
  10495. [0.6126, 0.4038, 0.8238, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138],
  10496. [0.6182, 0.4099, 0.7812, 0.3000, 0.3938, 0.2367, 0.5325, 0.5750],
  10497. [0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  10498. [0.6229, 0.4066, 0.8512, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350]]],
  10499. device='cuda:0')
  10500. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  10501. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  10502. loss_train: 0.06144377455348149
  10503. step: 36
  10504. running loss: 0.0017067715153744859
  10505. Train Steps: 36/90 Loss: 0.0017 torch.Size([8, 600, 800])
  10506. torch.Size([8, 8])
  10507. tensor([[0.6148, 0.4076, 0.8666, 0.4820, 0.4138, 0.5067, 0.5250, 0.5767],
  10508. [ nan, nan, 0.7612, 0.3250, 0.4037, 0.2533, 0.5438, 0.5767],
  10509. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  10510. [0.6102, 0.4005, 0.8688, 0.5100, 0.4813, 0.5400, 0.5404, 0.5064],
  10511. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  10512. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385],
  10513. [0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  10514. [0.6132, 0.4066, 0.7259, 0.2402, 0.3588, 0.3300, 0.6000, 0.5600]],
  10515. device='cuda:0', dtype=torch.float64)
  10516. predictions are: tensor([[0.5900, 0.3929, 0.9066, 0.5424, 0.3696, 0.5060, 0.5424, 0.5662],
  10517. [0.2800, 0.1911, 0.7943, 0.3393, 0.3765, 0.2683, 0.5316, 0.5769],
  10518. [0.5682, 0.3823, 0.7468, 0.2356, 0.3920, 0.2099, 0.5797, 0.5402],
  10519. [0.5947, 0.3937, 0.8851, 0.5474, 0.4331, 0.5075, 0.5386, 0.5139],
  10520. [0.6052, 0.4075, 0.9037, 0.5253, 0.4177, 0.5480, 0.5937, 0.5712],
  10521. [0.6961, 0.4666, 0.8221, 0.2895, 0.4008, 0.2437, 0.6741, 0.5568],
  10522. [0.6450, 0.4200, 0.8834, 0.5708, 0.3868, 0.5175, 0.6314, 0.5497],
  10523. [0.6174, 0.4091, 0.7355, 0.2822, 0.3222, 0.3458, 0.5709, 0.5829]],
  10524. device='cuda:0', grad_fn=<AddmmBackward>)
  10525. landmarks are: tensor([[[0.6148, 0.4076, 0.8666, 0.4820, 0.4137, 0.5067, 0.5250, 0.5767],
  10526. [0.0000, 0.0000, 0.7613, 0.3250, 0.4038, 0.2533, 0.5437, 0.5767],
  10527. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  10528. [0.6102, 0.4005, 0.8687, 0.5100, 0.4812, 0.5400, 0.5404, 0.5064],
  10529. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  10530. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385],
  10531. [0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  10532. [0.6132, 0.4066, 0.7259, 0.2402, 0.3587, 0.3300, 0.6000, 0.5600]]],
  10533. device='cuda:0')
  10534. loss_train_step before backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  10535. loss_train_step after backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  10536. loss_train: 0.0640558628947474
  10537. step: 37
  10538. running loss: 0.0017312395376958758
  10539. Train Steps: 37/90 Loss: 0.0017 torch.Size([8, 600, 800])
  10540. torch.Size([8, 8])
  10541. tensor([[0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5363, 0.5550],
  10542. [0.6058, 0.3986, 0.8324, 0.4626, 0.3838, 0.4983, 0.5147, 0.5466],
  10543. [0.6339, 0.4149, 0.8800, 0.5000, 0.3900, 0.5283, 0.7541, 0.5424],
  10544. [0.6250, 0.4103, 0.8950, 0.4400, 0.3912, 0.5650, 0.6050, 0.5133],
  10545. [0.6141, 0.4038, 0.8650, 0.4833, 0.4839, 0.5176, 0.5787, 0.5600],
  10546. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  10547. [0.6143, 0.4040, 0.8237, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  10548. [ nan, nan, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733]],
  10549. device='cuda:0', dtype=torch.float64)
  10550. predictions are: tensor([[0.6463, 0.4486, 0.7205, 0.2611, 0.3622, 0.2387, 0.5730, 0.5696],
  10551. [0.5679, 0.3821, 0.8604, 0.4857, 0.3614, 0.4933, 0.5540, 0.5588],
  10552. [0.5808, 0.3955, 0.9359, 0.5488, 0.3764, 0.5652, 0.7324, 0.5749],
  10553. [0.6251, 0.4041, 0.9236, 0.4773, 0.3702, 0.5895, 0.6485, 0.5471],
  10554. [0.6711, 0.4540, 0.9037, 0.5153, 0.4514, 0.5124, 0.6330, 0.5645],
  10555. [0.5877, 0.3802, 0.8120, 0.4371, 0.3248, 0.4588, 0.5510, 0.5464],
  10556. [0.5369, 0.3578, 0.8401, 0.3343, 0.3902, 0.2322, 0.5598, 0.5353],
  10557. [0.2642, 0.1942, 0.7214, 0.2071, 0.4413, 0.2090, 0.5914, 0.5881]],
  10558. device='cuda:0', grad_fn=<AddmmBackward>)
  10559. landmarks are: tensor([[[0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5362, 0.5550],
  10560. [0.6058, 0.3986, 0.8324, 0.4626, 0.3837, 0.4983, 0.5147, 0.5466],
  10561. [0.6339, 0.4149, 0.8800, 0.5000, 0.3900, 0.5283, 0.7541, 0.5424],
  10562. [0.6250, 0.4103, 0.8950, 0.4400, 0.3913, 0.5650, 0.6050, 0.5133],
  10563. [0.6141, 0.4038, 0.8650, 0.4833, 0.4839, 0.5176, 0.5788, 0.5600],
  10564. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  10565. [0.6143, 0.4040, 0.8238, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  10566. [0.0000, 0.0000, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733]]],
  10567. device='cuda:0')
  10568. loss_train_step before backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  10569. loss_train_step after backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  10570. loss_train: 0.0668318530661054
  10571. step: 38
  10572. running loss: 0.0017587329754238262
  10573.  
  10574. Train Steps: 38/90 Loss: 0.0018 torch.Size([8, 600, 800])
  10575. torch.Size([8, 8])
  10576. tensor([[0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  10577. [0.6268, 0.4102, 0.8938, 0.3667, 0.4025, 0.2833, 0.6275, 0.5183],
  10578. [0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  10579. [0.6137, 0.4038, 0.8563, 0.4050, 0.3813, 0.2550, 0.5106, 0.4954],
  10580. [0.6185, 0.4098, 0.8838, 0.4900, 0.4537, 0.5800, 0.6288, 0.5400],
  10581. [0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166],
  10582. [0.6201, 0.4082, 0.8827, 0.3715, 0.3825, 0.2712, 0.5845, 0.5412],
  10583. [0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431]],
  10584. device='cuda:0', dtype=torch.float64)
  10585. predictions are: tensor([[0.4808, 0.3253, 0.8920, 0.4184, 0.4204, 0.3435, 0.6587, 0.5698],
  10586. [0.5491, 0.3737, 0.8658, 0.3256, 0.4150, 0.3050, 0.6280, 0.5679],
  10587. [0.5523, 0.3764, 0.8521, 0.4818, 0.3708, 0.5154, 0.6665, 0.5470],
  10588. [0.5079, 0.3312, 0.8279, 0.3715, 0.3946, 0.2913, 0.5220, 0.5341],
  10589. [0.6193, 0.3941, 0.8742, 0.4529, 0.4648, 0.5880, 0.6097, 0.5398],
  10590. [0.6312, 0.4041, 0.8724, 0.4444, 0.4001, 0.5414, 0.6760, 0.5446],
  10591. [0.5933, 0.3861, 0.8602, 0.3223, 0.3764, 0.2990, 0.6152, 0.5545],
  10592. [0.5005, 0.3396, 0.8607, 0.4584, 0.4423, 0.4931, 0.5061, 0.5280]],
  10593. device='cuda:0', grad_fn=<AddmmBackward>)
  10594. landmarks are: tensor([[[0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  10595. [0.6268, 0.4102, 0.8938, 0.3667, 0.4025, 0.2833, 0.6275, 0.5183],
  10596. [0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  10597. [0.6137, 0.4038, 0.8562, 0.4050, 0.3812, 0.2550, 0.5106, 0.4954],
  10598. [0.6185, 0.4098, 0.8838, 0.4900, 0.4538, 0.5800, 0.6288, 0.5400],
  10599. [0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166],
  10600. [0.6201, 0.4082, 0.8827, 0.3715, 0.3825, 0.2712, 0.5845, 0.5412],
  10601. [0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431]]],
  10602. device='cuda:0')
  10603. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  10604. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  10605. loss_train: 0.06857802328886464
  10606. step: 39
  10607. running loss: 0.0017584108535606319
  10608. Train Steps: 39/90 Loss: 0.0018 torch.Size([8, 600, 800])
  10609. torch.Size([8, 8])
  10610. tensor([[0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  10611. [0.6183, 0.4076, 0.8838, 0.4517, 0.3813, 0.4483, 0.5775, 0.5633],
  10612. [0.6218, 0.4098, 0.7238, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  10613. [0.6250, 0.4131, 0.8688, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  10614. [0.6098, 0.3991, 0.8638, 0.4717, 0.4263, 0.4967, 0.5212, 0.5650],
  10615. [0.6102, 0.4001, 0.7738, 0.3583, 0.3463, 0.3800, 0.5524, 0.5689],
  10616. [0.6350, 0.4043, 0.8738, 0.5650, 0.3850, 0.4750, 0.6401, 0.4950],
  10617. [0.6227, 0.4083, 0.8938, 0.4800, 0.3800, 0.2950, 0.5737, 0.5350]],
  10618. device='cuda:0', dtype=torch.float64)
  10619. predictions are: tensor([[0.5491, 0.3530, 0.8468, 0.5286, 0.3845, 0.5185, 0.7065, 0.5212],
  10620. [0.6097, 0.3884, 0.9029, 0.4055, 0.4121, 0.4778, 0.6226, 0.5578],
  10621. [0.6146, 0.4019, 0.7406, 0.1859, 0.4536, 0.2654, 0.6394, 0.5445],
  10622. [0.6019, 0.3787, 0.8673, 0.2497, 0.4574, 0.2452, 0.6368, 0.5177],
  10623. [0.5598, 0.3475, 0.8746, 0.4580, 0.4476, 0.5110, 0.5642, 0.5375],
  10624. [0.5357, 0.3467, 0.7885, 0.3384, 0.3596, 0.4130, 0.5556, 0.5617],
  10625. [0.5348, 0.3464, 0.8776, 0.5457, 0.4168, 0.5147, 0.6626, 0.5053],
  10626. [0.5515, 0.3683, 0.8907, 0.4461, 0.4028, 0.3392, 0.6119, 0.5656]],
  10627. device='cuda:0', grad_fn=<AddmmBackward>)
  10628. landmarks are: tensor([[[0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  10629. [0.6183, 0.4076, 0.8838, 0.4517, 0.3812, 0.4483, 0.5775, 0.5633],
  10630. [0.6218, 0.4098, 0.7237, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  10631. [0.6250, 0.4131, 0.8687, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  10632. [0.6098, 0.3991, 0.8637, 0.4717, 0.4263, 0.4967, 0.5213, 0.5650],
  10633. [0.6102, 0.4001, 0.7738, 0.3583, 0.3462, 0.3800, 0.5524, 0.5689],
  10634. [0.6350, 0.4043, 0.8737, 0.5650, 0.3850, 0.4750, 0.6401, 0.4950],
  10635. [0.6227, 0.4083, 0.8938, 0.4800, 0.3800, 0.2950, 0.5738, 0.5350]]],
  10636. device='cuda:0')
  10637. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  10638. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  10639. loss_train: 0.06985682185040787
  10640. step: 40
  10641. running loss: 0.0017464205462601966
  10642. Train Steps: 40/90 Loss: 0.0017 torch.Size([8, 600, 800])
  10643. torch.Size([8, 8])
  10644. tensor([[0.6272, 0.4045, 0.8538, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  10645. [0.6104, 0.4029, 0.8738, 0.4900, 0.4088, 0.4533, 0.5070, 0.5510],
  10646. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  10647. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  10648. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  10649. [0.6201, 0.4102, 0.7288, 0.2417, 0.4150, 0.2383, 0.6100, 0.5500],
  10650. [0.6120, 0.4014, 0.6863, 0.2817, 0.3700, 0.2783, 0.5513, 0.5667],
  10651. [ nan, nan, 0.7240, 0.2722, 0.3900, 0.2567, 0.5168, 0.5933]],
  10652. device='cuda:0', dtype=torch.float64)
  10653. predictions are: tensor([[0.6196, 0.3877, 0.8643, 0.5624, 0.3899, 0.4373, 0.6491, 0.4795],
  10654. [0.5921, 0.3826, 0.9056, 0.4820, 0.4337, 0.4751, 0.5843, 0.5186],
  10655. [0.6419, 0.3869, 0.8472, 0.2046, 0.5146, 0.2001, 0.6809, 0.5151],
  10656. [0.6691, 0.4310, 0.8642, 0.5447, 0.4198, 0.4953, 0.6364, 0.5944],
  10657. [0.7202, 0.4597, 0.9144, 0.4096, 0.4746, 0.5625, 0.6505, 0.5424],
  10658. [0.6505, 0.4100, 0.7622, 0.2280, 0.4200, 0.2682, 0.6104, 0.5503],
  10659. [0.5771, 0.3638, 0.7290, 0.2275, 0.3697, 0.3242, 0.5733, 0.5296],
  10660. [0.0531, 0.0196, 0.7404, 0.2252, 0.4170, 0.2659, 0.5337, 0.5396]],
  10661. device='cuda:0', grad_fn=<AddmmBackward>)
  10662. landmarks are: tensor([[[0.6271, 0.4045, 0.8537, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  10663. [0.6104, 0.4029, 0.8737, 0.4900, 0.4087, 0.4533, 0.5070, 0.5510],
  10664. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  10665. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  10666. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  10667. [0.6201, 0.4102, 0.7287, 0.2417, 0.4150, 0.2383, 0.6100, 0.5500],
  10668. [0.6120, 0.4013, 0.6862, 0.2817, 0.3700, 0.2783, 0.5512, 0.5667],
  10669. [0.0000, 0.0000, 0.7240, 0.2722, 0.3900, 0.2567, 0.5168, 0.5933]]],
  10670. device='cuda:0')
  10671. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  10672. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  10673. loss_train: 0.07106089597800747
  10674. step: 41
  10675. running loss: 0.0017331925848294504
  10676. Train Steps: 41/90 Loss: 0.0017 torch.Size([8, 600, 800])
  10677. torch.Size([8, 8])
  10678. tensor([[0.6277, 0.4057, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  10679. [0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6138, 0.5400],
  10680. [0.6256, 0.4199, 0.8638, 0.5800, 0.3987, 0.4383, 0.5600, 0.5950],
  10681. [0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  10682. [0.6226, 0.4098, 0.8912, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  10683. [0.6090, 0.4010, 0.7838, 0.3483, 0.3538, 0.3783, 0.5462, 0.5077],
  10684. [0.6139, 0.4019, 0.7137, 0.2150, 0.4375, 0.1533, 0.5293, 0.5006],
  10685. [0.6159, 0.4085, 0.6900, 0.2283, 0.4088, 0.1950, 0.5123, 0.5397]],
  10686. device='cuda:0', dtype=torch.float64)
  10687. predictions are: tensor([[0.6513, 0.4053, 0.8174, 0.2385, 0.4802, 0.2250, 0.6510, 0.4989],
  10688. [0.6342, 0.3939, 0.8911, 0.4020, 0.3887, 0.5048, 0.6225, 0.5364],
  10689. [0.6598, 0.4216, 0.8528, 0.5737, 0.4386, 0.4707, 0.5988, 0.5928],
  10690. [0.6260, 0.4001, 0.8529, 0.5291, 0.4325, 0.5023, 0.6193, 0.5269],
  10691. [0.5934, 0.3801, 0.8919, 0.4087, 0.4370, 0.2679, 0.5986, 0.5490],
  10692. [0.5857, 0.3637, 0.7795, 0.3259, 0.3542, 0.3958, 0.5541, 0.5063],
  10693. [0.3791, 0.2412, 0.6990, 0.1641, 0.4398, 0.1908, 0.5585, 0.4823],
  10694. [0.5369, 0.3439, 0.6833, 0.2231, 0.4152, 0.2109, 0.5087, 0.5430]],
  10695. device='cuda:0', grad_fn=<AddmmBackward>)
  10696. landmarks are: tensor([[[0.6277, 0.4056, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  10697. [0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6137, 0.5400],
  10698. [0.6256, 0.4199, 0.8637, 0.5800, 0.3988, 0.4383, 0.5600, 0.5950],
  10699. [0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  10700. [0.6226, 0.4098, 0.8913, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  10701. [0.6090, 0.4010, 0.7837, 0.3483, 0.3537, 0.3783, 0.5462, 0.5077],
  10702. [0.6139, 0.4019, 0.7138, 0.2150, 0.4375, 0.1533, 0.5293, 0.5006],
  10703. [0.6159, 0.4085, 0.6900, 0.2283, 0.4087, 0.1950, 0.5123, 0.5397]]],
  10704. device='cuda:0')
  10705. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  10706. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  10707. loss_train: 0.07297768600983545
  10708. step: 42
  10709. running loss: 0.0017375639526151297
  10710.  
  10711. Train Steps: 42/90 Loss: 0.0017 torch.Size([8, 600, 800])
  10712. torch.Size([8, 8])
  10713. tensor([[0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  10714. [0.6185, 0.4079, 0.8838, 0.4617, 0.4838, 0.5650, 0.6175, 0.5850],
  10715. [0.6245, 0.4100, 0.7762, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  10716. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  10717. [0.6218, 0.4137, 0.7263, 0.2233, 0.4075, 0.2650, 0.6212, 0.5783],
  10718. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  10719. [0.6102, 0.4005, 0.8688, 0.5100, 0.4813, 0.5400, 0.5404, 0.5064],
  10720. [0.6274, 0.4003, 0.8638, 0.5967, 0.3688, 0.4900, 0.6108, 0.4661]],
  10721. device='cuda:0', dtype=torch.float64)
  10722. predictions are: tensor([[0.5348, 0.3344, 0.8918, 0.5236, 0.3754, 0.4113, 0.5386, 0.5193],
  10723. [0.6320, 0.3968, 0.8734, 0.4376, 0.4709, 0.5201, 0.6116, 0.5432],
  10724. [0.6034, 0.3844, 0.7373, 0.2353, 0.4592, 0.1258, 0.5958, 0.5342],
  10725. [0.5810, 0.3819, 0.7541, 0.2476, 0.4223, 0.1637, 0.5635, 0.5006],
  10726. [0.6223, 0.3896, 0.7139, 0.2423, 0.4031, 0.2472, 0.5769, 0.5434],
  10727. [0.5928, 0.3790, 0.8403, 0.4260, 0.3608, 0.4231, 0.5573, 0.5454],
  10728. [0.6233, 0.3942, 0.8505, 0.5002, 0.4849, 0.4924, 0.5449, 0.4821],
  10729. [0.6497, 0.4057, 0.8433, 0.5769, 0.3865, 0.4554, 0.6365, 0.4721]],
  10730. device='cuda:0', grad_fn=<AddmmBackward>)
  10731. landmarks are: tensor([[[0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  10732. [0.6184, 0.4079, 0.8838, 0.4617, 0.4837, 0.5650, 0.6175, 0.5850],
  10733. [0.6245, 0.4100, 0.7763, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  10734. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  10735. [0.6218, 0.4137, 0.7262, 0.2233, 0.4075, 0.2650, 0.6212, 0.5783],
  10736. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  10737. [0.6102, 0.4005, 0.8687, 0.5100, 0.4812, 0.5400, 0.5404, 0.5064],
  10738. [0.6274, 0.4003, 0.8637, 0.5967, 0.3688, 0.4900, 0.6108, 0.4661]]],
  10739. device='cuda:0')
  10740. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  10741. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  10742. loss_train: 0.07371883763698861
  10743. step: 43
  10744. running loss: 0.0017143915729532235
  10745. Train Steps: 43/90 Loss: 0.0017 torch.Size([8, 600, 800])
  10746. torch.Size([8, 8])
  10747. tensor([[0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367],
  10748. [0.6250, 0.3993, 0.9138, 0.4333, 0.3763, 0.5217, 0.6995, 0.5320],
  10749. [0.6151, 0.4058, 0.7068, 0.2680, 0.3400, 0.4083, 0.5775, 0.5733],
  10750. [0.6272, 0.4071, 0.8738, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  10751. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  10752. [0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  10753. [0.6276, 0.4095, 0.8237, 0.2250, 0.4662, 0.1783, 0.6171, 0.4869],
  10754. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496]],
  10755. device='cuda:0', dtype=torch.float64)
  10756. predictions are: tensor([[0.6772, 0.4490, 0.8220, 0.3402, 0.4000, 0.2432, 0.6008, 0.5333],
  10757. [0.6473, 0.4303, 0.8864, 0.4997, 0.3895, 0.5120, 0.6969, 0.5476],
  10758. [0.5202, 0.3567, 0.6952, 0.2878, 0.3395, 0.3716, 0.5097, 0.5642],
  10759. [0.5861, 0.4021, 0.8474, 0.6113, 0.3800, 0.3585, 0.5587, 0.4985],
  10760. [0.6097, 0.4207, 0.7052, 0.2364, 0.4152, 0.2074, 0.5507, 0.5604],
  10761. [0.5311, 0.3470, 0.7862, 0.2609, 0.4629, 0.1725, 0.5829, 0.5163],
  10762. [0.6151, 0.4042, 0.7916, 0.2439, 0.4920, 0.1672, 0.5747, 0.4947],
  10763. [0.6339, 0.4210, 0.8235, 0.2968, 0.4063, 0.2324, 0.5447, 0.5461]],
  10764. device='cuda:0', grad_fn=<AddmmBackward>)
  10765. landmarks are: tensor([[[0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367],
  10766. [0.6250, 0.3993, 0.9137, 0.4333, 0.3762, 0.5217, 0.6995, 0.5320],
  10767. [0.6151, 0.4058, 0.7068, 0.2680, 0.3400, 0.4083, 0.5775, 0.5733],
  10768. [0.6272, 0.4071, 0.8737, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  10769. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  10770. [0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  10771. [0.6276, 0.4095, 0.8238, 0.2250, 0.4663, 0.1783, 0.6171, 0.4869],
  10772. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496]]],
  10773. device='cuda:0')
  10774. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  10775. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  10776. loss_train: 0.07493773492751643
  10777. step: 44
  10778. running loss: 0.0017031303392617372
  10779. Train Steps: 44/90 Loss: 0.0017 torch.Size([8, 600, 800])
  10780. torch.Size([8, 8])
  10781. tensor([[ nan, nan, 0.7525, 0.2291, 0.3838, 0.3017, 0.6050, 0.5667],
  10782. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5637, 0.5633],
  10783. [0.6193, 0.4079, 0.7288, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  10784. [0.6277, 0.4057, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  10785. [ nan, nan, 0.7981, 0.3194, 0.3625, 0.3167, 0.5040, 0.5563],
  10786. [0.6276, 0.4002, 0.8800, 0.5533, 0.3575, 0.4400, 0.6132, 0.4672],
  10787. [0.6110, 0.4047, 0.8700, 0.4483, 0.3713, 0.3967, 0.5088, 0.5517],
  10788. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250]],
  10789. device='cuda:0', dtype=torch.float64)
  10790. predictions are: tensor([[0.1702, 0.1298, 0.7268, 0.2563, 0.3765, 0.2613, 0.5716, 0.5468],
  10791. [0.7073, 0.4773, 0.8548, 0.5149, 0.3481, 0.3462, 0.5311, 0.5441],
  10792. [0.7714, 0.5124, 0.7089, 0.2594, 0.4090, 0.2234, 0.5770, 0.6174],
  10793. [0.7612, 0.5006, 0.8099, 0.2706, 0.4542, 0.1748, 0.6107, 0.5040],
  10794. [0.2474, 0.1812, 0.7453, 0.3121, 0.3465, 0.2823, 0.4903, 0.5219],
  10795. [0.8275, 0.5585, 0.8661, 0.5414, 0.3792, 0.4369, 0.6221, 0.5096],
  10796. [0.6594, 0.4703, 0.8696, 0.4741, 0.3748, 0.3802, 0.4978, 0.5200],
  10797. [0.7095, 0.4710, 0.8922, 0.4212, 0.3628, 0.2994, 0.5741, 0.5112]],
  10798. device='cuda:0', grad_fn=<AddmmBackward>)
  10799. landmarks are: tensor([[[0.0000, 0.0000, 0.7525, 0.2291, 0.3837, 0.3017, 0.6050, 0.5667],
  10800. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5638, 0.5633],
  10801. [0.6193, 0.4078, 0.7287, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  10802. [0.6277, 0.4056, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  10803. [0.0000, 0.0000, 0.7981, 0.3194, 0.3625, 0.3167, 0.5040, 0.5563],
  10804. [0.6276, 0.4002, 0.8800, 0.5533, 0.3575, 0.4400, 0.6132, 0.4672],
  10805. [0.6110, 0.4047, 0.8700, 0.4483, 0.3713, 0.3967, 0.5088, 0.5517],
  10806. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250]]],
  10807. device='cuda:0')
  10808. loss_train_step before backward: tensor(0.0050, device='cuda:0', grad_fn=<MseLossBackward>)
  10809. loss_train_step after backward: tensor(0.0050, device='cuda:0', grad_fn=<MseLossBackward>)
  10810. loss_train: 0.07991487270919606
  10811. step: 45
  10812. running loss: 0.0017758860602043569
  10813. Train Steps: 45/90 Loss: 0.0018 torch.Size([8, 600, 800])
  10814. torch.Size([8, 8])
  10815. tensor([[0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  10816. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  10817. [0.6259, 0.4133, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  10818. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  10819. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  10820. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  10821. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  10822. [0.6250, 0.3993, 0.9138, 0.4333, 0.3763, 0.5217, 0.6995, 0.5320]],
  10823. device='cuda:0', dtype=torch.float64)
  10824. predictions are: tensor([[0.6370, 0.4299, 0.8304, 0.5609, 0.3799, 0.4245, 0.5558, 0.6125],
  10825. [0.6131, 0.4136, 0.7523, 0.2322, 0.3399, 0.2686, 0.5923, 0.5328],
  10826. [0.6087, 0.4224, 0.8043, 0.2357, 0.4771, 0.1322, 0.6005, 0.4889],
  10827. [0.5696, 0.3804, 0.7569, 0.2403, 0.3683, 0.2328, 0.5821, 0.5382],
  10828. [0.5883, 0.3892, 0.8575, 0.5295, 0.3315, 0.3194, 0.5310, 0.5377],
  10829. [0.6072, 0.3977, 0.8384, 0.5736, 0.4364, 0.4315, 0.4989, 0.5056],
  10830. [0.6448, 0.4398, 0.8627, 0.4838, 0.4578, 0.4607, 0.4873, 0.5472],
  10831. [0.6714, 0.4410, 0.9013, 0.4527, 0.3524, 0.4750, 0.7102, 0.5378]],
  10832. device='cuda:0', grad_fn=<AddmmBackward>)
  10833. landmarks are: tensor([[[0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  10834. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  10835. [0.6259, 0.4132, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  10836. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  10837. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  10838. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  10839. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  10840. [0.6250, 0.3993, 0.9137, 0.4333, 0.3762, 0.5217, 0.6995, 0.5320]]],
  10841. device='cuda:0')
  10842. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  10843. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  10844. loss_train: 0.08066413854248822
  10845. step: 46
  10846. running loss: 0.0017535682291845264
  10847.  
  10848. Train Steps: 46/90 Loss: 0.0018 torch.Size([8, 600, 800])
  10849. torch.Size([8, 8])
  10850. tensor([[0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  10851. [0.6276, 0.4120, 0.8738, 0.3133, 0.4225, 0.2217, 0.6203, 0.4892],
  10852. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  10853. [0.6145, 0.4008, 0.8750, 0.5383, 0.3975, 0.4650, 0.5563, 0.5533],
  10854. [0.6276, 0.4095, 0.8237, 0.2250, 0.4662, 0.1783, 0.6171, 0.4869],
  10855. [0.6058, 0.3986, 0.8324, 0.4626, 0.3838, 0.4983, 0.5147, 0.5466],
  10856. [0.6353, 0.4128, 0.9138, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  10857. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901]],
  10858. device='cuda:0', dtype=torch.float64)
  10859. predictions are: tensor([[0.5871, 0.3827, 0.7595, 0.2599, 0.4304, 0.1861, 0.5773, 0.5302],
  10860. [0.5781, 0.3907, 0.8509, 0.3644, 0.4057, 0.2110, 0.5864, 0.4940],
  10861. [0.5893, 0.3861, 0.8078, 0.2766, 0.4622, 0.1781, 0.6135, 0.5363],
  10862. [0.5620, 0.3663, 0.8583, 0.5443, 0.3683, 0.4701, 0.5473, 0.5656],
  10863. [0.6003, 0.3920, 0.8042, 0.2455, 0.4554, 0.1913, 0.5968, 0.4997],
  10864. [0.6014, 0.4023, 0.8215, 0.4718, 0.3669, 0.4867, 0.5102, 0.5505],
  10865. [0.5840, 0.3870, 0.8857, 0.3702, 0.4207, 0.2885, 0.6763, 0.5843],
  10866. [0.5322, 0.3514, 0.7657, 0.3172, 0.3484, 0.2597, 0.5061, 0.5328]],
  10867. device='cuda:0', grad_fn=<AddmmBackward>)
  10868. landmarks are: tensor([[[0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  10869. [0.6276, 0.4120, 0.8737, 0.3133, 0.4225, 0.2217, 0.6203, 0.4892],
  10870. [0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  10871. [0.6145, 0.4008, 0.8750, 0.5383, 0.3975, 0.4650, 0.5562, 0.5533],
  10872. [0.6276, 0.4095, 0.8238, 0.2250, 0.4663, 0.1783, 0.6171, 0.4869],
  10873. [0.6058, 0.3986, 0.8324, 0.4626, 0.3837, 0.4983, 0.5147, 0.5466],
  10874. [0.6353, 0.4128, 0.9137, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  10875. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901]]],
  10876. device='cuda:0')
  10877. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  10878. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  10879. loss_train: 0.08150381379527971
  10880. step: 47
  10881. running loss: 0.0017341236977719088
  10882. Train Steps: 47/90 Loss: 0.0017 torch.Size([8, 600, 800])
  10883. torch.Size([8, 8])
  10884. tensor([[0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5637, 0.5633],
  10885. [0.6085, 0.4008, 0.8588, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  10886. [0.6213, 0.4001, 0.7712, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183],
  10887. [ nan, nan, 0.7240, 0.2722, 0.3900, 0.2567, 0.5168, 0.5933],
  10888. [0.6213, 0.4131, 0.8438, 0.3550, 0.3513, 0.4400, 0.5716, 0.5123],
  10889. [0.6299, 0.4303, 0.7963, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  10890. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  10891. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483]],
  10892. device='cuda:0', dtype=torch.float64)
  10893. predictions are: tensor([[ 0.6507, 0.4312, 0.8895, 0.5101, 0.3482, 0.3958, 0.5736, 0.5568],
  10894. [ 0.6158, 0.4013, 0.8695, 0.5192, 0.4944, 0.5012, 0.5342, 0.5290],
  10895. [ 0.7051, 0.4321, 0.7647, 0.2217, 0.4441, 0.1759, 0.5999, 0.5217],
  10896. [-0.0247, 0.0039, 0.7466, 0.2767, 0.3981, 0.2568, 0.5355, 0.5709],
  10897. [ 0.7045, 0.4758, 0.8790, 0.3624, 0.3280, 0.4345, 0.5924, 0.5202],
  10898. [ 0.6245, 0.4351, 0.8443, 0.3931, 0.4787, 0.2461, 0.5647, 0.6040],
  10899. [ 0.7088, 0.4377, 0.9097, 0.4705, 0.3590, 0.4088, 0.6427, 0.5270],
  10900. [ 0.6248, 0.4281, 0.7528, 0.2074, 0.4034, 0.2537, 0.6343, 0.5507]],
  10901. device='cuda:0', grad_fn=<AddmmBackward>)
  10902. landmarks are: tensor([[[0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5638, 0.5633],
  10903. [0.6084, 0.4008, 0.8587, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  10904. [0.6213, 0.4001, 0.7713, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183],
  10905. [0.0000, 0.0000, 0.7240, 0.2722, 0.3900, 0.2567, 0.5168, 0.5933],
  10906. [0.6213, 0.4131, 0.8438, 0.3550, 0.3512, 0.4400, 0.5716, 0.5123],
  10907. [0.6299, 0.4303, 0.7962, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  10908. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  10909. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483]]],
  10910. device='cuda:0')
  10911. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  10912. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  10913. loss_train: 0.08213754411553964
  10914. step: 48
  10915. running loss: 0.001711198835740409
  10916. Train Steps: 48/90 Loss: 0.0017 torch.Size([8, 600, 800])
  10917. torch.Size([8, 8])
  10918. tensor([[0.6193, 0.4108, 0.7438, 0.2700, 0.3650, 0.3683, 0.6238, 0.5717],
  10919. [0.6128, 0.4118, 0.8638, 0.5333, 0.4625, 0.5267, 0.5193, 0.5475],
  10920. [0.6234, 0.4179, 0.7825, 0.3450, 0.3813, 0.2867, 0.5675, 0.5617],
  10921. [0.6099, 0.4030, 0.8638, 0.5117, 0.4983, 0.4965, 0.5086, 0.5388],
  10922. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6038, 0.4833],
  10923. [0.6131, 0.4064, 0.8638, 0.5200, 0.4788, 0.4783, 0.5258, 0.5867],
  10924. [0.6190, 0.4135, 0.8000, 0.4883, 0.3566, 0.3647, 0.5613, 0.5900],
  10925. [0.6133, 0.4094, 0.8495, 0.4028, 0.3588, 0.3200, 0.5003, 0.5407]],
  10926. device='cuda:0', dtype=torch.float64)
  10927. predictions are: tensor([[0.5959, 0.4000, 0.7621, 0.2510, 0.3706, 0.3683, 0.6783, 0.5539],
  10928. [0.6303, 0.4155, 0.8891, 0.5091, 0.4621, 0.4924, 0.5939, 0.5456],
  10929. [0.5692, 0.3708, 0.8184, 0.3110, 0.3984, 0.2670, 0.6072, 0.5645],
  10930. [0.5968, 0.3888, 0.8988, 0.4949, 0.5172, 0.4731, 0.5595, 0.5253],
  10931. [0.6241, 0.3917, 0.8970, 0.4466, 0.3830, 0.4809, 0.6115, 0.4860],
  10932. [0.6153, 0.4076, 0.8837, 0.4904, 0.4676, 0.4800, 0.5875, 0.5728],
  10933. [0.5677, 0.3693, 0.8548, 0.4486, 0.3512, 0.3619, 0.5806, 0.5470],
  10934. [0.5376, 0.3586, 0.8915, 0.3900, 0.3824, 0.2972, 0.5508, 0.5358]],
  10935. device='cuda:0', grad_fn=<AddmmBackward>)
  10936. landmarks are: tensor([[[0.6193, 0.4108, 0.7437, 0.2700, 0.3650, 0.3683, 0.6237, 0.5717],
  10937. [0.6128, 0.4118, 0.8637, 0.5333, 0.4625, 0.5267, 0.5193, 0.5475],
  10938. [0.6234, 0.4179, 0.7825, 0.3450, 0.3812, 0.2867, 0.5675, 0.5617],
  10939. [0.6098, 0.4030, 0.8637, 0.5117, 0.4983, 0.4965, 0.5086, 0.5388],
  10940. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6037, 0.4833],
  10941. [0.6132, 0.4063, 0.8637, 0.5200, 0.4787, 0.4783, 0.5258, 0.5867],
  10942. [0.6190, 0.4135, 0.8000, 0.4883, 0.3566, 0.3647, 0.5612, 0.5900],
  10943. [0.6133, 0.4094, 0.8495, 0.4028, 0.3587, 0.3200, 0.5003, 0.5407]]],
  10944. device='cuda:0')
  10945. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  10946. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  10947. loss_train: 0.08310104586416855
  10948. step: 49
  10949. running loss: 0.001695939711513644
  10950. Train Steps: 49/90 Loss: 0.0017 torch.Size([8, 600, 800])
  10951. torch.Size([8, 8])
  10952. tensor([[0.6109, 0.4003, 0.8650, 0.4883, 0.4775, 0.4867, 0.5175, 0.5683],
  10953. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  10954. [0.6222, 0.3937, 0.8350, 0.5617, 0.4138, 0.4600, 0.5800, 0.5233],
  10955. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783],
  10956. [0.6109, 0.4036, 0.7188, 0.1750, 0.3850, 0.2550, 0.5863, 0.5567],
  10957. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  10958. [0.6078, 0.4033, 0.8019, 0.3055, 0.3450, 0.4200, 0.6025, 0.5550],
  10959. [0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274]],
  10960. device='cuda:0', dtype=torch.float64)
  10961. predictions are: tensor([[0.5848, 0.3844, 0.8986, 0.4746, 0.4875, 0.5192, 0.5734, 0.5595],
  10962. [0.5635, 0.3480, 0.8617, 0.5607, 0.4685, 0.5106, 0.5479, 0.5005],
  10963. [0.6011, 0.3629, 0.8678, 0.5411, 0.4214, 0.4758, 0.5868, 0.5364],
  10964. [0.5875, 0.3733, 0.8663, 0.3486, 0.3731, 0.3224, 0.5392, 0.5639],
  10965. [0.6006, 0.3898, 0.7459, 0.2103, 0.4047, 0.2789, 0.5768, 0.5538],
  10966. [0.5433, 0.3653, 0.7747, 0.2095, 0.4477, 0.2071, 0.6094, 0.5367],
  10967. [0.5403, 0.3561, 0.8304, 0.2777, 0.3716, 0.4511, 0.6204, 0.5646],
  10968. [0.6354, 0.4143, 0.8928, 0.3292, 0.4869, 0.2185, 0.6488, 0.5354]],
  10969. device='cuda:0', grad_fn=<AddmmBackward>)
  10970. landmarks are: tensor([[[0.6109, 0.4003, 0.8650, 0.4883, 0.4775, 0.4867, 0.5175, 0.5683],
  10971. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  10972. [0.6222, 0.3937, 0.8350, 0.5617, 0.4137, 0.4600, 0.5800, 0.5233],
  10973. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783],
  10974. [0.6108, 0.4036, 0.7188, 0.1750, 0.3850, 0.2550, 0.5863, 0.5567],
  10975. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  10976. [0.6078, 0.4033, 0.8019, 0.3055, 0.3450, 0.4200, 0.6025, 0.5550],
  10977. [0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274]]],
  10978. device='cuda:0')
  10979. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  10980. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  10981. loss_train: 0.08384644222678617
  10982. step: 50
  10983. running loss: 0.0016769288445357233
  10984.  
  10985. Train Steps: 50/90 Loss: 0.0017 torch.Size([8, 600, 800])
  10986. torch.Size([8, 8])
  10987. tensor([[0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  10988. [0.6193, 0.4034, 0.7757, 0.2347, 0.3733, 0.2919, 0.5930, 0.4926],
  10989. [0.6257, 0.4024, 0.8672, 0.5422, 0.4196, 0.5198, 0.6694, 0.5204],
  10990. [0.6142, 0.3982, 0.8650, 0.4883, 0.3912, 0.4317, 0.5315, 0.5350],
  10991. [0.6160, 0.4093, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  10992. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  10993. [0.6187, 0.4104, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  10994. [0.6177, 0.4086, 0.8738, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700]],
  10995. device='cuda:0', dtype=torch.float64)
  10996. predictions are: tensor([[0.6018, 0.3981, 0.8972, 0.4834, 0.4861, 0.6041, 0.6385, 0.5484],
  10997. [0.5998, 0.4012, 0.7800, 0.2538, 0.4224, 0.2867, 0.6259, 0.5099],
  10998. [0.6428, 0.4122, 0.8682, 0.5459, 0.4503, 0.5400, 0.6588, 0.5392],
  10999. [0.6533, 0.3981, 0.8786, 0.5066, 0.4169, 0.4677, 0.5248, 0.5279],
  11000. [0.5899, 0.3816, 0.8542, 0.4554, 0.3995, 0.4642, 0.5371, 0.5860],
  11001. [0.7126, 0.4692, 0.8975, 0.4193, 0.3818, 0.3942, 0.5930, 0.5641],
  11002. [0.6259, 0.4239, 0.7293, 0.2223, 0.4088, 0.2643, 0.5765, 0.5782],
  11003. [0.5646, 0.3629, 0.8836, 0.3953, 0.4145, 0.5627, 0.6421, 0.5542]],
  11004. device='cuda:0', grad_fn=<AddmmBackward>)
  11005. landmarks are: tensor([[[0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  11006. [0.6193, 0.4034, 0.7757, 0.2347, 0.3733, 0.2919, 0.5930, 0.4926],
  11007. [0.6257, 0.4024, 0.8672, 0.5422, 0.4196, 0.5198, 0.6694, 0.5204],
  11008. [0.6143, 0.3982, 0.8650, 0.4883, 0.3913, 0.4317, 0.5315, 0.5350],
  11009. [0.6160, 0.4092, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  11010. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  11011. [0.6187, 0.4103, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  11012. [0.6177, 0.4085, 0.8737, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700]]],
  11013. device='cuda:0')
  11014. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  11015. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  11016. loss_train: 0.08451496018096805
  11017. step: 51
  11018. running loss: 0.0016571560819797655
  11019. Train Steps: 51/90 Loss: 0.0017 torch.Size([8, 600, 800])
  11020. torch.Size([8, 8])
  11021. tensor([[0.6339, 0.4102, 0.8588, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089],
  11022. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  11023. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  11024. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5963, 0.6217],
  11025. [0.6199, 0.4065, 0.7598, 0.2385, 0.4317, 0.1981, 0.5933, 0.5221],
  11026. [0.6307, 0.4029, 0.8988, 0.4817, 0.3937, 0.3500, 0.7311, 0.5378],
  11027. [ nan, nan, 0.8463, 0.2550, 0.5850, 0.2133, 0.7129, 0.6072],
  11028. [0.6239, 0.4174, 0.8425, 0.5733, 0.4825, 0.4500, 0.5625, 0.5933]],
  11029. device='cuda:0', dtype=torch.float64)
  11030. predictions are: tensor([[0.6809, 0.4253, 0.8568, 0.2991, 0.4494, 0.2359, 0.6058, 0.5163],
  11031. [0.5803, 0.3767, 0.8677, 0.4200, 0.4125, 0.6332, 0.5489, 0.5271],
  11032. [0.6340, 0.3932, 0.8501, 0.5158, 0.3730, 0.5077, 0.5502, 0.5099],
  11033. [0.5400, 0.3482, 0.6996, 0.2484, 0.4044, 0.2768, 0.5484, 0.6001],
  11034. [0.6741, 0.4330, 0.7321, 0.2190, 0.4196, 0.2534, 0.5463, 0.5307],
  11035. [0.6390, 0.4043, 0.8925, 0.4650, 0.3780, 0.3900, 0.6473, 0.5288],
  11036. [0.4883, 0.3077, 0.8510, 0.2756, 0.5270, 0.3071, 0.6809, 0.5730],
  11037. [0.5779, 0.3868, 0.8395, 0.5546, 0.4571, 0.4585, 0.5455, 0.6070]],
  11038. device='cuda:0', grad_fn=<AddmmBackward>)
  11039. landmarks are: tensor([[[0.6339, 0.4102, 0.8587, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089],
  11040. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  11041. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  11042. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5962, 0.6217],
  11043. [0.6199, 0.4065, 0.7598, 0.2385, 0.4317, 0.1981, 0.5933, 0.5221],
  11044. [0.6307, 0.4029, 0.8988, 0.4817, 0.3938, 0.3500, 0.7311, 0.5378],
  11045. [0.0000, 0.0000, 0.8462, 0.2550, 0.5850, 0.2133, 0.7129, 0.6072],
  11046. [0.6239, 0.4174, 0.8425, 0.5733, 0.4825, 0.4500, 0.5625, 0.5933]]],
  11047. device='cuda:0')
  11048. loss_train_step before backward: tensor(0.0064, device='cuda:0', grad_fn=<MseLossBackward>)
  11049. loss_train_step after backward: tensor(0.0064, device='cuda:0', grad_fn=<MseLossBackward>)
  11050. loss_train: 0.09093694994226098
  11051. step: 52
  11052. running loss: 0.0017487874988896342
  11053. Train Steps: 52/90 Loss: 0.0017 torch.Size([8, 600, 800])
  11054. torch.Size([8, 8])
  11055. tensor([[0.6293, 0.4024, 0.8750, 0.5000, 0.4012, 0.5733, 0.7121, 0.5633],
  11056. [0.6201, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  11057. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683],
  11058. [0.6198, 0.4130, 0.8762, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  11059. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  11060. [0.6239, 0.4061, 0.8850, 0.4600, 0.4225, 0.5200, 0.6138, 0.5450],
  11061. [0.6182, 0.4099, 0.7812, 0.3000, 0.3937, 0.2367, 0.5325, 0.5750],
  11062. [0.6182, 0.4058, 0.8738, 0.4350, 0.3563, 0.3400, 0.5290, 0.5822]],
  11063. device='cuda:0', dtype=torch.float64)
  11064. predictions are: tensor([[0.5538, 0.3485, 0.8356, 0.5000, 0.4175, 0.6157, 0.6252, 0.5638],
  11065. [0.5602, 0.3685, 0.7444, 0.2204, 0.4513, 0.2355, 0.6059, 0.5304],
  11066. [0.5542, 0.3638, 0.8654, 0.5291, 0.3889, 0.3825, 0.5489, 0.5775],
  11067. [0.6536, 0.4437, 0.8621, 0.4250, 0.3704, 0.5169, 0.5454, 0.5253],
  11068. [0.5912, 0.4080, 0.8529, 0.2716, 0.5181, 0.1879, 0.6456, 0.5222],
  11069. [0.6205, 0.4104, 0.8665, 0.4365, 0.4210, 0.5852, 0.6185, 0.5612],
  11070. [0.4969, 0.3367, 0.7763, 0.3154, 0.4192, 0.2654, 0.5452, 0.5896],
  11071. [0.6335, 0.4119, 0.8561, 0.4200, 0.3686, 0.3731, 0.5136, 0.5897]],
  11072. device='cuda:0', grad_fn=<AddmmBackward>)
  11073. landmarks are: tensor([[[0.6293, 0.4024, 0.8750, 0.5000, 0.4013, 0.5733, 0.7121, 0.5633],
  11074. [0.6202, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  11075. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683],
  11076. [0.6198, 0.4130, 0.8763, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  11077. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  11078. [0.6239, 0.4061, 0.8850, 0.4600, 0.4225, 0.5200, 0.6137, 0.5450],
  11079. [0.6182, 0.4099, 0.7812, 0.3000, 0.3938, 0.2367, 0.5325, 0.5750],
  11080. [0.6182, 0.4058, 0.8737, 0.4350, 0.3562, 0.3400, 0.5290, 0.5822]]],
  11081. device='cuda:0')
  11082. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  11083. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  11084. loss_train: 0.0920837412122637
  11085. step: 53
  11086. running loss: 0.0017374290794766737
  11087. Train Steps: 53/90 Loss: 0.0017 torch.Size([8, 600, 800])
  11088. torch.Size([8, 8])
  11089. tensor([[0.6263, 0.4233, 0.7924, 0.4626, 0.3788, 0.2883, 0.5573, 0.6047],
  11090. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6188, 0.5283],
  11091. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  11092. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  11093. [0.6179, 0.4040, 0.7412, 0.1850, 0.3825, 0.2783, 0.5837, 0.5600],
  11094. [0.6111, 0.4033, 0.8300, 0.3267, 0.3588, 0.3333, 0.5444, 0.5637],
  11095. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  11096. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153]],
  11097. device='cuda:0', dtype=torch.float64)
  11098. predictions are: tensor([[0.6361, 0.4294, 0.7938, 0.4831, 0.4019, 0.3325, 0.5618, 0.6023],
  11099. [0.6089, 0.4054, 0.8951, 0.3480, 0.4153, 0.2785, 0.6387, 0.5254],
  11100. [0.6143, 0.3938, 0.8661, 0.5296, 0.3820, 0.4893, 0.6041, 0.4902],
  11101. [0.6072, 0.4022, 0.8795, 0.4778, 0.3956, 0.4922, 0.5996, 0.5346],
  11102. [0.5502, 0.3681, 0.7329, 0.2432, 0.4047, 0.3000, 0.6031, 0.5652],
  11103. [0.5602, 0.3857, 0.8316, 0.3321, 0.3668, 0.3287, 0.5372, 0.5738],
  11104. [0.5965, 0.4166, 0.8792, 0.4442, 0.4421, 0.5145, 0.5707, 0.5443],
  11105. [0.5677, 0.3775, 0.7882, 0.3297, 0.3547, 0.3545, 0.6025, 0.5210]],
  11106. device='cuda:0', grad_fn=<AddmmBackward>)
  11107. landmarks are: tensor([[[0.6263, 0.4232, 0.7924, 0.4626, 0.3787, 0.2883, 0.5573, 0.6047],
  11108. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6187, 0.5283],
  11109. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  11110. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  11111. [0.6179, 0.4040, 0.7412, 0.1850, 0.3825, 0.2783, 0.5838, 0.5600],
  11112. [0.6111, 0.4033, 0.8300, 0.3267, 0.3587, 0.3333, 0.5444, 0.5637],
  11113. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  11114. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153]]],
  11115. device='cuda:0')
  11116. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  11117. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  11118. loss_train: 0.0925499573640991
  11119. step: 54
  11120. running loss: 0.0017138880993351686
  11121.  
  11122. Train Steps: 54/90 Loss: 0.0017 torch.Size([8, 600, 800])
  11123. torch.Size([8, 8])
  11124. tensor([[0.6151, 0.4085, 0.8750, 0.4367, 0.3887, 0.4367, 0.5066, 0.5846],
  11125. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  11126. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  11127. [0.6118, 0.4052, 0.8463, 0.3917, 0.3538, 0.3450, 0.5053, 0.5593],
  11128. [0.6271, 0.4040, 0.9138, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  11129. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  11130. [0.6264, 0.4049, 0.8988, 0.4633, 0.3813, 0.4983, 0.6326, 0.4843],
  11131. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333]],
  11132. device='cuda:0', dtype=torch.float64)
  11133. predictions are: tensor([[0.6453, 0.4232, 0.8670, 0.4669, 0.3818, 0.4092, 0.4712, 0.5853],
  11134. [0.5555, 0.3425, 0.8555, 0.3448, 0.4047, 0.3914, 0.6735, 0.5412],
  11135. [0.5939, 0.3750, 0.8760, 0.3566, 0.3731, 0.2931, 0.6047, 0.5714],
  11136. [0.5895, 0.4000, 0.8453, 0.4075, 0.3547, 0.3354, 0.4997, 0.5532],
  11137. [0.5426, 0.3417, 0.9176, 0.4015, 0.4547, 0.2686, 0.7097, 0.5357],
  11138. [0.6337, 0.4331, 0.7227, 0.2762, 0.3751, 0.2437, 0.5495, 0.5227],
  11139. [0.5736, 0.3552, 0.8865, 0.4961, 0.3587, 0.5078, 0.6134, 0.4761],
  11140. [0.5949, 0.4000, 0.7421, 0.2326, 0.4343, 0.1824, 0.5836, 0.5334]],
  11141. device='cuda:0', grad_fn=<AddmmBackward>)
  11142. landmarks are: tensor([[[0.6151, 0.4085, 0.8750, 0.4367, 0.3887, 0.4367, 0.5066, 0.5846],
  11143. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  11144. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  11145. [0.6118, 0.4052, 0.8462, 0.3917, 0.3537, 0.3450, 0.5053, 0.5593],
  11146. [0.6271, 0.4040, 0.9137, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  11147. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  11148. [0.6264, 0.4049, 0.8988, 0.4633, 0.3812, 0.4983, 0.6326, 0.4843],
  11149. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333]]],
  11150. device='cuda:0')
  11151. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  11152. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  11153. loss_train: 0.0932905454246793
  11154. step: 55
  11155. running loss: 0.001696191734994169
  11156. Train Steps: 55/90 Loss: 0.0017 torch.Size([8, 600, 800])
  11157. torch.Size([8, 8])
  11158. tensor([[0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  11159. [0.6182, 0.4099, 0.7812, 0.3000, 0.3937, 0.2367, 0.5325, 0.5750],
  11160. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  11161. [0.6109, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  11162. [0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133],
  11163. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  11164. [0.6091, 0.3997, 0.8314, 0.4334, 0.3788, 0.4550, 0.5213, 0.5656],
  11165. [0.6202, 0.4053, 0.8638, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133]],
  11166. device='cuda:0', dtype=torch.float64)
  11167. predictions are: tensor([[0.5960, 0.3916, 0.8328, 0.3137, 0.3090, 0.3448, 0.6383, 0.5323],
  11168. [0.5227, 0.3610, 0.7821, 0.2958, 0.3755, 0.2210, 0.5449, 0.5759],
  11169. [0.5617, 0.3655, 0.8606, 0.2810, 0.4852, 0.2073, 0.7063, 0.5246],
  11170. [0.6541, 0.4277, 0.8747, 0.4585, 0.3404, 0.4017, 0.5486, 0.5071],
  11171. [0.6761, 0.4527, 0.8541, 0.5255, 0.3467, 0.4580, 0.6511, 0.4972],
  11172. [0.5504, 0.3616, 0.7424, 0.3120, 0.4642, 0.1758, 0.5704, 0.5991],
  11173. [0.6524, 0.4272, 0.8397, 0.4339, 0.3420, 0.4341, 0.5560, 0.5430],
  11174. [0.6357, 0.4008, 0.8463, 0.5149, 0.4319, 0.4953, 0.5789, 0.5273]],
  11175. device='cuda:0', grad_fn=<AddmmBackward>)
  11176. landmarks are: tensor([[[0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  11177. [0.6182, 0.4099, 0.7812, 0.3000, 0.3938, 0.2367, 0.5325, 0.5750],
  11178. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  11179. [0.6108, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  11180. [0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133],
  11181. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  11182. [0.6091, 0.3997, 0.8314, 0.4334, 0.3787, 0.4550, 0.5213, 0.5656],
  11183. [0.6202, 0.4053, 0.8637, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133]]],
  11184. device='cuda:0')
  11185. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  11186. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  11187. loss_train: 0.09418759748223238
  11188. step: 56
  11189. running loss: 0.0016819213836112925
  11190. Train Steps: 56/90 Loss: 0.0017 torch.Size([8, 600, 800])
  11191. torch.Size([8, 8])
  11192. tensor([[0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309],
  11193. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317],
  11194. [0.6201, 0.4064, 0.8688, 0.5050, 0.4225, 0.5100, 0.6138, 0.5500],
  11195. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6038, 0.6167],
  11196. [0.6250, 0.4106, 0.8700, 0.3717, 0.3588, 0.4967, 0.6038, 0.5167],
  11197. [0.6224, 0.4179, 0.8700, 0.5683, 0.4037, 0.4683, 0.5650, 0.5600],
  11198. [0.6127, 0.4115, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  11199. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466]],
  11200. device='cuda:0', dtype=torch.float64)
  11201. predictions are: tensor([[0.6676, 0.4368, 0.8900, 0.4799, 0.3935, 0.3156, 0.6512, 0.5184],
  11202. [0.7548, 0.4830, 0.8803, 0.5309, 0.3699, 0.3366, 0.6227, 0.5090],
  11203. [0.6117, 0.3932, 0.8814, 0.4646, 0.4048, 0.4646, 0.6164, 0.5311],
  11204. [0.5857, 0.3969, 0.8048, 0.2933, 0.3470, 0.3339, 0.6369, 0.5901],
  11205. [0.6177, 0.4172, 0.8760, 0.3511, 0.3644, 0.4374, 0.6449, 0.5348],
  11206. [0.6561, 0.4186, 0.8780, 0.5503, 0.3889, 0.4274, 0.5882, 0.5325],
  11207. [0.6282, 0.4230, 0.7384, 0.2648, 0.3612, 0.2285, 0.5473, 0.5478],
  11208. [0.6170, 0.4053, 0.7809, 0.3479, 0.3499, 0.3602, 0.5553, 0.5295]],
  11209. device='cuda:0', grad_fn=<AddmmBackward>)
  11210. landmarks are: tensor([[[0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309],
  11211. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317],
  11212. [0.6201, 0.4064, 0.8687, 0.5050, 0.4225, 0.5100, 0.6137, 0.5500],
  11213. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6037, 0.6167],
  11214. [0.6250, 0.4105, 0.8700, 0.3717, 0.3587, 0.4967, 0.6037, 0.5167],
  11215. [0.6224, 0.4179, 0.8700, 0.5683, 0.4038, 0.4683, 0.5650, 0.5600],
  11216. [0.6127, 0.4114, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  11217. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466]]],
  11218. device='cuda:0')
  11219. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  11220. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  11221. loss_train: 0.09527627841453068
  11222. step: 57
  11223. running loss: 0.001671513656395275
  11224. Train Steps: 57/90 Loss: 0.0017 torch.Size([8, 600, 800])
  11225. torch.Size([8, 8])
  11226. tensor([[0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  11227. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  11228. [0.6178, 0.4012, 0.8900, 0.4495, 0.3891, 0.5617, 0.5972, 0.4879],
  11229. [0.6126, 0.4073, 0.8750, 0.5133, 0.3800, 0.4333, 0.4986, 0.5378],
  11230. [0.6082, 0.4024, 0.8738, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  11231. [0.6179, 0.4008, 0.7505, 0.2678, 0.4368, 0.1891, 0.5831, 0.5263],
  11232. [0.6256, 0.4199, 0.8638, 0.5800, 0.3987, 0.4383, 0.5600, 0.5950],
  11233. [0.6305, 0.3983, 0.8950, 0.4833, 0.3688, 0.4683, 0.6375, 0.5117]],
  11234. device='cuda:0', dtype=torch.float64)
  11235. predictions are: tensor([[0.5923, 0.3882, 0.7885, 0.2782, 0.3581, 0.2975, 0.6219, 0.5589],
  11236. [0.6705, 0.4474, 0.8350, 0.2856, 0.4068, 0.2243, 0.6166, 0.4929],
  11237. [0.6653, 0.4542, 0.8764, 0.4276, 0.3875, 0.5155, 0.6295, 0.5115],
  11238. [0.6140, 0.4167, 0.8751, 0.5256, 0.3806, 0.4165, 0.5473, 0.5524],
  11239. [0.6482, 0.4520, 0.8684, 0.3917, 0.3614, 0.3510, 0.5551, 0.5411],
  11240. [0.7469, 0.4962, 0.7565, 0.2374, 0.4311, 0.1452, 0.5950, 0.5401],
  11241. [0.6686, 0.4340, 0.8478, 0.5763, 0.3875, 0.4165, 0.5866, 0.6031],
  11242. [0.5929, 0.3802, 0.9192, 0.4869, 0.3647, 0.4593, 0.6552, 0.5192]],
  11243. device='cuda:0', grad_fn=<AddmmBackward>)
  11244. landmarks are: tensor([[[0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  11245. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  11246. [0.6178, 0.4012, 0.8900, 0.4495, 0.3891, 0.5617, 0.5972, 0.4879],
  11247. [0.6126, 0.4073, 0.8750, 0.5133, 0.3800, 0.4333, 0.4986, 0.5378],
  11248. [0.6082, 0.4024, 0.8737, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  11249. [0.6179, 0.4008, 0.7505, 0.2678, 0.4368, 0.1891, 0.5831, 0.5263],
  11250. [0.6256, 0.4199, 0.8637, 0.5800, 0.3988, 0.4383, 0.5600, 0.5950],
  11251. [0.6305, 0.3983, 0.8950, 0.4833, 0.3688, 0.4683, 0.6375, 0.5117]]],
  11252. device='cuda:0')
  11253. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  11254. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  11255. loss_train: 0.09628399272332899
  11256. step: 58
  11257. running loss: 0.0016600688400573965
  11258.  
  11259. Train Steps: 58/90 Loss: 0.0017 torch.Size([8, 600, 800])
  11260. torch.Size([8, 8])
  11261. tensor([[0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  11262. [0.6241, 0.4143, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  11263. [0.6107, 0.4013, 0.8700, 0.4650, 0.5049, 0.5176, 0.5850, 0.5550],
  11264. [0.6058, 0.3978, 0.8287, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461],
  11265. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6063, 0.6083],
  11266. [0.6267, 0.4065, 0.8313, 0.2467, 0.4788, 0.1733, 0.6312, 0.5133],
  11267. [0.6251, 0.4163, 0.8662, 0.4467, 0.3625, 0.3567, 0.6038, 0.5533],
  11268. [0.6148, 0.3996, 0.8488, 0.3867, 0.3488, 0.4067, 0.5863, 0.5000]],
  11269. device='cuda:0', dtype=torch.float64)
  11270. predictions are: tensor([[0.6687, 0.4266, 0.8866, 0.5246, 0.3931, 0.4508, 0.5781, 0.4987],
  11271. [0.5809, 0.3803, 0.8802, 0.4642, 0.4092, 0.5018, 0.6329, 0.5567],
  11272. [0.6141, 0.4125, 0.8633, 0.4633, 0.4655, 0.4415, 0.5980, 0.5629],
  11273. [0.6745, 0.4543, 0.8192, 0.3643, 0.3462, 0.3749, 0.5754, 0.5205],
  11274. [0.6390, 0.4397, 0.8338, 0.3683, 0.3432, 0.4207, 0.5815, 0.5846],
  11275. [0.6567, 0.4287, 0.8254, 0.2548, 0.4760, 0.1247, 0.6478, 0.5078],
  11276. [0.6230, 0.4060, 0.8469, 0.4417, 0.3595, 0.3145, 0.5798, 0.5405],
  11277. [0.7041, 0.4520, 0.8424, 0.3891, 0.3473, 0.3727, 0.5912, 0.5173]],
  11278. device='cuda:0', grad_fn=<AddmmBackward>)
  11279. landmarks are: tensor([[[0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  11280. [0.6241, 0.4142, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  11281. [0.6107, 0.4013, 0.8700, 0.4650, 0.5049, 0.5176, 0.5850, 0.5550],
  11282. [0.6058, 0.3978, 0.8288, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461],
  11283. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6062, 0.6083],
  11284. [0.6266, 0.4065, 0.8313, 0.2467, 0.4787, 0.1733, 0.6313, 0.5133],
  11285. [0.6252, 0.4162, 0.8662, 0.4467, 0.3625, 0.3567, 0.6037, 0.5533],
  11286. [0.6148, 0.3996, 0.8487, 0.3867, 0.3487, 0.4067, 0.5863, 0.5000]]],
  11287. device='cuda:0')
  11288. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  11289. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  11290. loss_train: 0.09704984255949967
  11291. step: 59
  11292. running loss: 0.0016449125857542317
  11293. Train Steps: 59/90 Loss: 0.0016 torch.Size([8, 600, 800])
  11294. torch.Size([8, 8])
  11295. tensor([[0.6099, 0.4030, 0.8638, 0.5117, 0.4983, 0.4965, 0.5086, 0.5388],
  11296. [0.6147, 0.4107, 0.8137, 0.3333, 0.3750, 0.2683, 0.5006, 0.5412],
  11297. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467],
  11298. [0.6192, 0.4128, 0.8513, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  11299. [0.6129, 0.4063, 0.8738, 0.5250, 0.4313, 0.4733, 0.5230, 0.5874],
  11300. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5787, 0.5117],
  11301. [0.6198, 0.4115, 0.7762, 0.2717, 0.3713, 0.3200, 0.5837, 0.5683],
  11302. [0.6131, 0.4037, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680]],
  11303. device='cuda:0', dtype=torch.float64)
  11304. predictions are: tensor([[0.5524, 0.3684, 0.8810, 0.5297, 0.5100, 0.4835, 0.5320, 0.5300],
  11305. [0.5466, 0.3586, 0.8001, 0.3506, 0.3732, 0.2584, 0.5257, 0.5388],
  11306. [0.6086, 0.3828, 0.8570, 0.3213, 0.3544, 0.3979, 0.5820, 0.5188],
  11307. [0.6513, 0.4311, 0.8822, 0.5697, 0.4222, 0.5225, 0.6060, 0.5511],
  11308. [0.6884, 0.4478, 0.8747, 0.5551, 0.4310, 0.4643, 0.5607, 0.5479],
  11309. [0.8111, 0.5159, 0.7555, 0.2227, 0.4257, 0.2197, 0.5950, 0.4980],
  11310. [0.6200, 0.4054, 0.7956, 0.2775, 0.3638, 0.3102, 0.5975, 0.5639],
  11311. [0.5884, 0.3865, 0.7129, 0.2693, 0.3853, 0.2834, 0.5373, 0.5499]],
  11312. device='cuda:0', grad_fn=<AddmmBackward>)
  11313. landmarks are: tensor([[[0.6098, 0.4030, 0.8637, 0.5117, 0.4983, 0.4965, 0.5086, 0.5388],
  11314. [0.6147, 0.4107, 0.8138, 0.3333, 0.3750, 0.2683, 0.5006, 0.5412],
  11315. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467],
  11316. [0.6192, 0.4128, 0.8512, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  11317. [0.6130, 0.4063, 0.8737, 0.5250, 0.4313, 0.4733, 0.5230, 0.5874],
  11318. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5788, 0.5117],
  11319. [0.6198, 0.4115, 0.7763, 0.2717, 0.3713, 0.3200, 0.5838, 0.5683],
  11320. [0.6131, 0.4036, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680]]],
  11321. device='cuda:0')
  11322. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  11323. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  11324. loss_train: 0.0983880981511902
  11325. step: 60
  11326. running loss: 0.00163980163585317
  11327. Train Steps: 60/90 Loss: 0.0016 torch.Size([8, 600, 800])
  11328. torch.Size([8, 8])
  11329. tensor([[0.6275, 0.4024, 0.8600, 0.2283, 0.5350, 0.1800, 0.7074, 0.5413],
  11330. [ nan, nan, 0.7850, 0.2700, 0.4288, 0.1717, 0.5199, 0.4999],
  11331. [0.6125, 0.4076, 0.8488, 0.3883, 0.3700, 0.3683, 0.5026, 0.5505],
  11332. [0.6210, 0.4164, 0.7202, 0.2930, 0.4025, 0.2483, 0.5687, 0.5567],
  11333. [0.6200, 0.4059, 0.8700, 0.4900, 0.4163, 0.5000, 0.6162, 0.5467],
  11334. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  11335. [0.6128, 0.4115, 0.7934, 0.3778, 0.3450, 0.4033, 0.5337, 0.5456],
  11336. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284]],
  11337. device='cuda:0', dtype=torch.float64)
  11338. predictions are: tensor([[0.5544, 0.3590, 0.8481, 0.2554, 0.5409, 0.2350, 0.6758, 0.5367],
  11339. [0.2139, 0.1506, 0.7774, 0.2641, 0.4452, 0.2224, 0.5098, 0.5522],
  11340. [0.7379, 0.4704, 0.8540, 0.4256, 0.3654, 0.3925, 0.4794, 0.5282],
  11341. [0.5526, 0.3593, 0.7588, 0.2905, 0.4331, 0.2762, 0.5581, 0.5591],
  11342. [0.6221, 0.4242, 0.8964, 0.5445, 0.4330, 0.5784, 0.5897, 0.5532],
  11343. [0.5752, 0.3930, 0.8614, 0.3583, 0.3594, 0.4973, 0.6189, 0.5515],
  11344. [0.7502, 0.4973, 0.8014, 0.4046, 0.3410, 0.4333, 0.5064, 0.5422],
  11345. [0.7757, 0.5141, 0.7650, 0.2587, 0.4693, 0.2058, 0.5800, 0.5271]],
  11346. device='cuda:0', grad_fn=<AddmmBackward>)
  11347. landmarks are: tensor([[[0.6275, 0.4024, 0.8600, 0.2283, 0.5350, 0.1800, 0.7074, 0.5413],
  11348. [0.0000, 0.0000, 0.7850, 0.2700, 0.4288, 0.1717, 0.5199, 0.4999],
  11349. [0.6125, 0.4076, 0.8487, 0.3883, 0.3700, 0.3683, 0.5026, 0.5505],
  11350. [0.6210, 0.4164, 0.7202, 0.2930, 0.4025, 0.2483, 0.5688, 0.5567],
  11351. [0.6199, 0.4059, 0.8700, 0.4900, 0.4162, 0.5000, 0.6162, 0.5467],
  11352. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  11353. [0.6128, 0.4115, 0.7934, 0.3778, 0.3450, 0.4033, 0.5337, 0.5456],
  11354. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284]]],
  11355. device='cuda:0')
  11356. loss_train_step before backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  11357. loss_train_step after backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  11358. loss_train: 0.10155432714964263
  11359. step: 61
  11360. running loss: 0.001664825035240043
  11361. Train Steps: 61/90 Loss: 0.0017 torch.Size([8, 600, 800])
  11362. torch.Size([8, 8])
  11363. tensor([[0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  11364. [0.6076, 0.3953, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954],
  11365. [ nan, nan, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  11366. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6188, 0.5283],
  11367. [ nan, nan, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  11368. [0.6143, 0.4040, 0.8237, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  11369. [0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  11370. [ nan, nan, 0.8938, 0.2850, 0.4662, 0.3117, 0.7406, 0.5528]],
  11371. device='cuda:0', dtype=torch.float64)
  11372. predictions are: tensor([[0.7333, 0.5004, 0.7735, 0.3322, 0.3780, 0.3378, 0.4660, 0.5727],
  11373. [0.7130, 0.4630, 0.7946, 0.3756, 0.3519, 0.4460, 0.5214, 0.5315],
  11374. [0.3009, 0.2057, 0.8649, 0.3455, 0.5145, 0.2872, 0.6320, 0.5746],
  11375. [0.7436, 0.4856, 0.8686, 0.3659, 0.4083, 0.3257, 0.5701, 0.5233],
  11376. [0.2093, 0.1384, 0.6915, 0.2328, 0.4616, 0.2953, 0.5221, 0.5782],
  11377. [0.6982, 0.4350, 0.7919, 0.3409, 0.4342, 0.2717, 0.4762, 0.5167],
  11378. [0.8061, 0.5239, 0.8521, 0.3967, 0.3773, 0.4158, 0.5907, 0.5020],
  11379. [0.1672, 0.1115, 0.8612, 0.2978, 0.4808, 0.3562, 0.6760, 0.5824]],
  11380. device='cuda:0', grad_fn=<AddmmBackward>)
  11381. landmarks are: tensor([[[0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  11382. [0.6076, 0.3952, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954],
  11383. [0.0000, 0.0000, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  11384. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6187, 0.5283],
  11385. [0.0000, 0.0000, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  11386. [0.6143, 0.4040, 0.8238, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  11387. [0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  11388. [0.0000, 0.0000, 0.8938, 0.2850, 0.4663, 0.3117, 0.7406, 0.5528]]],
  11389. device='cuda:0')
  11390. loss_train_step before backward: tensor(0.0064, device='cuda:0', grad_fn=<MseLossBackward>)
  11391. loss_train_step after backward: tensor(0.0064, device='cuda:0', grad_fn=<MseLossBackward>)
  11392. loss_train: 0.10794588641147129
  11393. step: 62
  11394. running loss: 0.0017410626840559885
  11395.  
  11396. Train Steps: 62/90 Loss: 0.0017 torch.Size([8, 600, 800])
  11397. torch.Size([8, 8])
  11398. tensor([[0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  11399. [0.6236, 0.3966, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  11400. [0.6179, 0.4118, 0.7278, 0.4237, 0.3588, 0.3400, 0.5675, 0.5917],
  11401. [0.6098, 0.3991, 0.8638, 0.4717, 0.4263, 0.4967, 0.5212, 0.5650],
  11402. [ nan, nan, 0.6469, 0.1943, 0.4025, 0.2000, 0.5125, 0.5533],
  11403. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  11404. [0.6058, 0.3986, 0.8324, 0.4626, 0.3838, 0.4983, 0.5147, 0.5466],
  11405. [0.6072, 0.4029, 0.7037, 0.2150, 0.3912, 0.2267, 0.5516, 0.5507]],
  11406. device='cuda:0', dtype=torch.float64)
  11407. predictions are: tensor([[0.6443, 0.3973, 0.9186, 0.4363, 0.3844, 0.4203, 0.6107, 0.5025],
  11408. [0.5870, 0.3576, 0.9018, 0.4988, 0.3745, 0.4480, 0.5780, 0.5176],
  11409. [0.6250, 0.4147, 0.7870, 0.4136, 0.3721, 0.3787, 0.5733, 0.5922],
  11410. [0.5771, 0.3501, 0.8886, 0.4988, 0.4560, 0.5279, 0.5326, 0.5574],
  11411. [0.1619, 0.0922, 0.7203, 0.2126, 0.4333, 0.2041, 0.5526, 0.5699],
  11412. [0.4896, 0.3245, 0.9115, 0.4716, 0.4796, 0.5833, 0.5782, 0.5692],
  11413. [0.5880, 0.3797, 0.8613, 0.4694, 0.4195, 0.5210, 0.5192, 0.5485],
  11414. [0.6626, 0.4191, 0.7250, 0.2456, 0.4148, 0.2477, 0.5574, 0.5377]],
  11415. device='cuda:0', grad_fn=<AddmmBackward>)
  11416. landmarks are: tensor([[[0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  11417. [0.6236, 0.3965, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  11418. [0.6179, 0.4118, 0.7278, 0.4237, 0.3587, 0.3400, 0.5675, 0.5917],
  11419. [0.6098, 0.3991, 0.8637, 0.4717, 0.4263, 0.4967, 0.5213, 0.5650],
  11420. [0.0000, 0.0000, 0.6469, 0.1943, 0.4025, 0.2000, 0.5125, 0.5533],
  11421. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  11422. [0.6058, 0.3986, 0.8324, 0.4626, 0.3837, 0.4983, 0.5147, 0.5466],
  11423. [0.6072, 0.4029, 0.7038, 0.2150, 0.3913, 0.2267, 0.5516, 0.5507]]],
  11424. device='cuda:0')
  11425. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  11426. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  11427. loss_train: 0.1094864918559324
  11428. step: 63
  11429. running loss: 0.001737880823110038
  11430. Train Steps: 63/90 Loss: 0.0017 torch.Size([8, 600, 800])
  11431. torch.Size([8, 8])
  11432. tensor([[0.6202, 0.4066, 0.8746, 0.3376, 0.3717, 0.3090, 0.5842, 0.5165],
  11433. [0.6236, 0.3967, 0.8675, 0.5400, 0.3862, 0.4517, 0.5825, 0.5200],
  11434. [0.6200, 0.3978, 0.8900, 0.4550, 0.3775, 0.5200, 0.6150, 0.5367],
  11435. [0.6204, 0.4007, 0.7838, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  11436. [0.6246, 0.4028, 0.8738, 0.4867, 0.4088, 0.5667, 0.6362, 0.5200],
  11437. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  11438. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  11439. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650]],
  11440. device='cuda:0', dtype=torch.float64)
  11441. predictions are: tensor([[0.5513, 0.3592, 0.8662, 0.3485, 0.3772, 0.3205, 0.5762, 0.5387],
  11442. [0.4817, 0.3048, 0.8697, 0.5463, 0.3880, 0.4739, 0.5632, 0.5283],
  11443. [0.5399, 0.3348, 0.9029, 0.4546, 0.4078, 0.5403, 0.5872, 0.5457],
  11444. [0.4647, 0.2731, 0.7782, 0.2394, 0.4507, 0.1822, 0.5812, 0.5226],
  11445. [0.5025, 0.3091, 0.8910, 0.4677, 0.4300, 0.5890, 0.6320, 0.5428],
  11446. [0.5606, 0.3719, 0.8676, 0.5222, 0.4119, 0.4951, 0.6652, 0.5893],
  11447. [0.4988, 0.3030, 0.7705, 0.3064, 0.3635, 0.3410, 0.5255, 0.5430],
  11448. [0.4922, 0.3373, 0.8392, 0.3999, 0.3699, 0.3585, 0.5458, 0.5868]],
  11449. device='cuda:0', grad_fn=<AddmmBackward>)
  11450. landmarks are: tensor([[[0.6202, 0.4066, 0.8746, 0.3376, 0.3717, 0.3090, 0.5842, 0.5165],
  11451. [0.6236, 0.3967, 0.8675, 0.5400, 0.3862, 0.4517, 0.5825, 0.5200],
  11452. [0.6199, 0.3978, 0.8900, 0.4550, 0.3775, 0.5200, 0.6150, 0.5367],
  11453. [0.6204, 0.4007, 0.7837, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  11454. [0.6246, 0.4028, 0.8737, 0.4867, 0.4087, 0.5667, 0.6363, 0.5200],
  11455. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  11456. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  11457. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650]]],
  11458. device='cuda:0')
  11459. loss_train_step before backward: tensor(0.0029, device='cuda:0', grad_fn=<MseLossBackward>)
  11460. loss_train_step after backward: tensor(0.0029, device='cuda:0', grad_fn=<MseLossBackward>)
  11461. loss_train: 0.11233921625535004
  11462. step: 64
  11463. running loss: 0.0017553002539898444
  11464. Train Steps: 64/90 Loss: 0.0018 torch.Size([8, 600, 800])
  11465. torch.Size([8, 8])
  11466. tensor([[0.6204, 0.4055, 0.8438, 0.5733, 0.4574, 0.4801, 0.5487, 0.5617],
  11467. [0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  11468. [0.6193, 0.4079, 0.7288, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  11469. [0.6151, 0.4058, 0.7068, 0.2680, 0.3400, 0.4083, 0.5775, 0.5733],
  11470. [0.6276, 0.4120, 0.8738, 0.3133, 0.4225, 0.2217, 0.6203, 0.4892],
  11471. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467],
  11472. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  11473. [0.6257, 0.4167, 0.8775, 0.3433, 0.3563, 0.4133, 0.6200, 0.5667]],
  11474. device='cuda:0', dtype=torch.float64)
  11475. predictions are: tensor([[0.4944, 0.3168, 0.8474, 0.5816, 0.4411, 0.4373, 0.5721, 0.5677],
  11476. [0.5039, 0.3218, 0.8871, 0.5070, 0.4029, 0.5259, 0.5760, 0.5276],
  11477. [0.4978, 0.3210, 0.7504, 0.2609, 0.4081, 0.2607, 0.6218, 0.6058],
  11478. [0.5300, 0.3422, 0.7354, 0.2737, 0.3297, 0.3932, 0.5870, 0.5635],
  11479. [0.4631, 0.3005, 0.8866, 0.3447, 0.4234, 0.2134, 0.6178, 0.4854],
  11480. [0.5129, 0.3247, 0.8528, 0.3257, 0.3440, 0.3944, 0.5765, 0.5320],
  11481. [0.4362, 0.2714, 0.8773, 0.4883, 0.3986, 0.4727, 0.5806, 0.5211],
  11482. [0.4654, 0.2932, 0.8854, 0.3447, 0.3446, 0.4048, 0.6510, 0.5646]],
  11483. device='cuda:0', grad_fn=<AddmmBackward>)
  11484. landmarks are: tensor([[[0.6204, 0.4055, 0.8438, 0.5733, 0.4574, 0.4801, 0.5487, 0.5617],
  11485. [0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  11486. [0.6193, 0.4078, 0.7287, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  11487. [0.6151, 0.4058, 0.7068, 0.2680, 0.3400, 0.4083, 0.5775, 0.5733],
  11488. [0.6276, 0.4120, 0.8737, 0.3133, 0.4225, 0.2217, 0.6203, 0.4892],
  11489. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467],
  11490. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  11491. [0.6257, 0.4167, 0.8775, 0.3433, 0.3562, 0.4133, 0.6200, 0.5667]]],
  11492. device='cuda:0')
  11493. loss_train_step before backward: tensor(0.0036, device='cuda:0', grad_fn=<MseLossBackward>)
  11494. loss_train_step after backward: tensor(0.0036, device='cuda:0', grad_fn=<MseLossBackward>)
  11495. loss_train: 0.11592077181558125
  11496. step: 65
  11497. running loss: 0.0017833964894704807
  11498. Train Steps: 65/90 Loss: 0.0018 torch.Size([8, 600, 800])
  11499. torch.Size([8, 8])
  11500. tensor([[0.6310, 0.4017, 0.8563, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006],
  11501. [ nan, nan, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  11502. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373],
  11503. [0.6115, 0.3998, 0.7063, 0.2383, 0.4037, 0.1950, 0.5320, 0.4993],
  11504. [0.6157, 0.4102, 0.8513, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  11505. [0.6304, 0.4024, 0.8925, 0.4800, 0.3937, 0.4817, 0.7485, 0.5297],
  11506. [0.6173, 0.4114, 0.7325, 0.2500, 0.4213, 0.1917, 0.5338, 0.5700],
  11507. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389]],
  11508. device='cuda:0', dtype=torch.float64)
  11509. predictions are: tensor([[ 0.5804, 0.3857, 0.8771, 0.5844, 0.3679, 0.4868, 0.6360, 0.5012],
  11510. [-0.0425, -0.0367, 0.8551, 0.2400, 0.4732, 0.2295, 0.6997, 0.5585],
  11511. [ 0.5929, 0.4005, 0.8401, 0.5809, 0.3690, 0.4718, 0.5810, 0.5505],
  11512. [ 0.5663, 0.3888, 0.7067, 0.2198, 0.3865, 0.2174, 0.5131, 0.5074],
  11513. [ 0.5684, 0.3669, 0.8644, 0.3607, 0.3334, 0.3551, 0.5328, 0.5657],
  11514. [ 0.5612, 0.3696, 0.8814, 0.4700, 0.3682, 0.4813, 0.6840, 0.5249],
  11515. [ 0.4214, 0.2801, 0.7307, 0.2213, 0.4080, 0.1993, 0.5412, 0.5526],
  11516. [ 0.5440, 0.3590, 0.7677, 0.2257, 0.3728, 0.2917, 0.5940, 0.5315]],
  11517. device='cuda:0', grad_fn=<AddmmBackward>)
  11518. landmarks are: tensor([[[0.6310, 0.4017, 0.8562, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006],
  11519. [0.0000, 0.0000, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  11520. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373],
  11521. [0.6115, 0.3998, 0.7063, 0.2383, 0.4038, 0.1950, 0.5320, 0.4993],
  11522. [0.6157, 0.4102, 0.8512, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  11523. [0.6304, 0.4024, 0.8925, 0.4800, 0.3938, 0.4817, 0.7485, 0.5297],
  11524. [0.6173, 0.4114, 0.7325, 0.2500, 0.4212, 0.1917, 0.5337, 0.5700],
  11525. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389]]],
  11526. device='cuda:0')
  11527. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  11528. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  11529. loss_train: 0.11747965749236755
  11530. step: 66
  11531. running loss: 0.0017799948104904174
  11532.  
  11533. Train Steps: 66/90 Loss: 0.0018 torch.Size([8, 600, 800])
  11534. torch.Size([8, 8])
  11535. tensor([[0.6128, 0.4084, 0.8738, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  11536. [0.6129, 0.4063, 0.8738, 0.5250, 0.4313, 0.4733, 0.5230, 0.5874],
  11537. [0.6261, 0.3987, 0.9045, 0.4208, 0.3600, 0.4633, 0.6570, 0.5162],
  11538. [0.6170, 0.4102, 0.7468, 0.3695, 0.3463, 0.3767, 0.5238, 0.5823],
  11539. [0.6300, 0.4133, 0.8538, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  11540. [0.6205, 0.4062, 0.8337, 0.2683, 0.3675, 0.4283, 0.6338, 0.5250],
  11541. [0.6239, 0.4061, 0.8850, 0.4600, 0.4225, 0.5200, 0.6138, 0.5450],
  11542. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895]],
  11543. device='cuda:0', dtype=torch.float64)
  11544. predictions are: tensor([[0.6251, 0.4033, 0.8557, 0.5007, 0.3239, 0.3519, 0.5428, 0.5372],
  11545. [0.6137, 0.4180, 0.8449, 0.5570, 0.3967, 0.4510, 0.5643, 0.5701],
  11546. [0.5641, 0.3861, 0.8715, 0.4379, 0.3183, 0.4774, 0.6737, 0.5093],
  11547. [0.5425, 0.3741, 0.7356, 0.3609, 0.3198, 0.3816, 0.5545, 0.5847],
  11548. [0.3630, 0.2469, 0.8479, 0.2203, 0.5101, 0.2250, 0.7429, 0.5562],
  11549. [0.5277, 0.3549, 0.7876, 0.2757, 0.3350, 0.4166, 0.6158, 0.5284],
  11550. [0.5695, 0.3940, 0.8755, 0.4750, 0.3733, 0.5183, 0.6559, 0.5532],
  11551. [0.4964, 0.3130, 0.7487, 0.1767, 0.4193, 0.1578, 0.6286, 0.4901]],
  11552. device='cuda:0', grad_fn=<AddmmBackward>)
  11553. landmarks are: tensor([[[0.6127, 0.4084, 0.8737, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  11554. [0.6130, 0.4063, 0.8737, 0.5250, 0.4313, 0.4733, 0.5230, 0.5874],
  11555. [0.6261, 0.3987, 0.9045, 0.4208, 0.3600, 0.4633, 0.6570, 0.5162],
  11556. [0.6170, 0.4102, 0.7468, 0.3695, 0.3462, 0.3767, 0.5238, 0.5823],
  11557. [0.6300, 0.4133, 0.8537, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  11558. [0.6205, 0.4062, 0.8338, 0.2683, 0.3675, 0.4283, 0.6338, 0.5250],
  11559. [0.6239, 0.4061, 0.8850, 0.4600, 0.4225, 0.5200, 0.6137, 0.5450],
  11560. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895]]],
  11561. device='cuda:0')
  11562. loss_train_step before backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  11563. loss_train_step after backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  11564. loss_train: 0.1202581440738868
  11565. step: 67
  11566. running loss: 0.001794897672744579
  11567. Train Steps: 67/90 Loss: 0.0018 torch.Size([8, 600, 800])
  11568. torch.Size([8, 8])
  11569. tensor([[0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  11570. [0.6250, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6088, 0.5183],
  11571. [0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  11572. [0.6151, 0.4125, 0.8738, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  11573. [0.6193, 0.4034, 0.7757, 0.2347, 0.3733, 0.2919, 0.5930, 0.4926],
  11574. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  11575. [0.6100, 0.4071, 0.7601, 0.3444, 0.3400, 0.4117, 0.5625, 0.5617],
  11576. [0.6339, 0.4118, 0.7988, 0.5800, 0.3912, 0.4583, 0.7343, 0.5760]],
  11577. device='cuda:0', dtype=torch.float64)
  11578. predictions are: tensor([[0.6612, 0.4562, 0.8680, 0.4165, 0.4319, 0.5259, 0.6222, 0.5346],
  11579. [0.5574, 0.3880, 0.8802, 0.4529, 0.4421, 0.5177, 0.6578, 0.5421],
  11580. [0.5582, 0.3715, 0.8900, 0.4436, 0.3643, 0.3686, 0.6026, 0.5147],
  11581. [0.5936, 0.4037, 0.8495, 0.4358, 0.3438, 0.3282, 0.5475, 0.5556],
  11582. [0.5840, 0.4030, 0.7509, 0.2079, 0.3725, 0.2356, 0.6260, 0.4932],
  11583. [0.4665, 0.3168, 0.7085, 0.2098, 0.3752, 0.2044, 0.5738, 0.5676],
  11584. [0.6158, 0.4249, 0.7574, 0.3089, 0.3399, 0.4049, 0.5947, 0.5401],
  11585. [0.5303, 0.3628, 0.7998, 0.4980, 0.3657, 0.4049, 0.6914, 0.5614]],
  11586. device='cuda:0', grad_fn=<AddmmBackward>)
  11587. landmarks are: tensor([[[0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  11588. [0.6251, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6087, 0.5183],
  11589. [0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  11590. [0.6151, 0.4125, 0.8737, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  11591. [0.6193, 0.4034, 0.7757, 0.2347, 0.3733, 0.2919, 0.5930, 0.4926],
  11592. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  11593. [0.6100, 0.4071, 0.7601, 0.3444, 0.3400, 0.4117, 0.5625, 0.5617],
  11594. [0.6339, 0.4118, 0.7987, 0.5800, 0.3913, 0.4583, 0.7343, 0.5760]]],
  11595. device='cuda:0')
  11596. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  11597. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  11598. loss_train: 0.12180890384479426
  11599. step: 68
  11600. running loss: 0.0017913074094822684
  11601. Train Steps: 68/90 Loss: 0.0018 torch.Size([8, 600, 800])
  11602. torch.Size([8, 8])
  11603. tensor([[0.6185, 0.4129, 0.8900, 0.4567, 0.3937, 0.5417, 0.5734, 0.5110],
  11604. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  11605. [0.6112, 0.4029, 0.8638, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  11606. [0.6189, 0.4029, 0.8375, 0.5767, 0.4745, 0.4829, 0.5551, 0.5598],
  11607. [0.6293, 0.4097, 0.8800, 0.2517, 0.5262, 0.2600, 0.7430, 0.5378],
  11608. [0.6272, 0.4045, 0.8538, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  11609. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  11610. [0.6196, 0.4088, 0.8888, 0.4583, 0.4500, 0.5683, 0.6138, 0.5883]],
  11611. device='cuda:0', dtype=torch.float64)
  11612. predictions are: tensor([[0.6241, 0.4234, 0.8645, 0.3934, 0.3759, 0.5098, 0.5802, 0.5197],
  11613. [0.6142, 0.4060, 0.7109, 0.2683, 0.4481, 0.1873, 0.5449, 0.5926],
  11614. [0.5609, 0.3858, 0.8440, 0.4253, 0.4511, 0.4539, 0.5817, 0.5489],
  11615. [0.6234, 0.4384, 0.8018, 0.4867, 0.4361, 0.4341, 0.5862, 0.5596],
  11616. [0.5419, 0.3619, 0.8450, 0.2001, 0.4939, 0.2217, 0.7119, 0.5358],
  11617. [0.6548, 0.4339, 0.8147, 0.5305, 0.3407, 0.3900, 0.5933, 0.4804],
  11618. [0.6723, 0.4529, 0.8368, 0.3916, 0.3305, 0.3355, 0.6153, 0.4993],
  11619. [0.6250, 0.4366, 0.8396, 0.3740, 0.4185, 0.5401, 0.6053, 0.5594]],
  11620. device='cuda:0', grad_fn=<AddmmBackward>)
  11621. landmarks are: tensor([[[0.6186, 0.4129, 0.8900, 0.4567, 0.3938, 0.5417, 0.5734, 0.5110],
  11622. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  11623. [0.6112, 0.4029, 0.8637, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  11624. [0.6189, 0.4029, 0.8375, 0.5767, 0.4745, 0.4829, 0.5551, 0.5598],
  11625. [0.6293, 0.4097, 0.8800, 0.2517, 0.5263, 0.2600, 0.7430, 0.5378],
  11626. [0.6271, 0.4045, 0.8537, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  11627. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  11628. [0.6196, 0.4088, 0.8888, 0.4583, 0.4500, 0.5683, 0.6137, 0.5883]]],
  11629. device='cuda:0')
  11630. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  11631. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  11632. loss_train: 0.12344274375936948
  11633. step: 69
  11634. running loss: 0.00178902527187492
  11635. Train Steps: 69/90 Loss: 0.0018 torch.Size([8, 600, 800])
  11636. torch.Size([8, 8])
  11637. tensor([[0.6177, 0.4086, 0.8738, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700],
  11638. [0.6271, 0.4040, 0.9138, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  11639. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864],
  11640. [0.6140, 0.4070, 0.8700, 0.5000, 0.4612, 0.4900, 0.5260, 0.5852],
  11641. [0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5988, 0.5700],
  11642. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  11643. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  11644. [0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921]],
  11645. device='cuda:0', dtype=torch.float64)
  11646. predictions are: tensor([[0.7387, 0.4992, 0.8326, 0.3715, 0.3934, 0.5256, 0.6269, 0.5265],
  11647. [0.6438, 0.4278, 0.9021, 0.3559, 0.4733, 0.2499, 0.6971, 0.5084],
  11648. [0.6759, 0.4547, 0.8371, 0.5015, 0.4044, 0.4597, 0.6859, 0.5400],
  11649. [0.6668, 0.4625, 0.8443, 0.4836, 0.4699, 0.4555, 0.5151, 0.5617],
  11650. [0.5322, 0.3633, 0.7571, 0.2769, 0.3974, 0.3349, 0.5683, 0.5541],
  11651. [0.6474, 0.4479, 0.8151, 0.3181, 0.3610, 0.4445, 0.6198, 0.5369],
  11652. [0.6664, 0.4622, 0.8580, 0.4708, 0.4462, 0.5022, 0.6261, 0.5201],
  11653. [0.7322, 0.4684, 0.8168, 0.2213, 0.4268, 0.2554, 0.6238, 0.4816]],
  11654. device='cuda:0', grad_fn=<AddmmBackward>)
  11655. landmarks are: tensor([[[0.6177, 0.4085, 0.8737, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700],
  11656. [0.6271, 0.4040, 0.9137, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  11657. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864],
  11658. [0.6140, 0.4070, 0.8700, 0.5000, 0.4613, 0.4900, 0.5260, 0.5852],
  11659. [0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5987, 0.5700],
  11660. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  11661. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  11662. [0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921]]],
  11663. device='cuda:0')
  11664. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  11665. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  11666. loss_train: 0.12500527323572896
  11667. step: 70
  11668. running loss: 0.0017857896176532708
  11669.  
  11670. Train Steps: 70/90 Loss: 0.0018 torch.Size([8, 600, 800])
  11671. torch.Size([8, 8])
  11672. tensor([[0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  11673. [0.6236, 0.3966, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  11674. [0.6147, 0.4107, 0.8137, 0.3333, 0.3750, 0.2683, 0.5006, 0.5412],
  11675. [0.6248, 0.4185, 0.8500, 0.5767, 0.4463, 0.4550, 0.5613, 0.5917],
  11676. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  11677. [ nan, nan, 0.8850, 0.3000, 0.5363, 0.2250, 0.7343, 0.5771],
  11678. [0.6222, 0.4169, 0.8638, 0.5650, 0.4313, 0.4783, 0.5637, 0.5633],
  11679. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533]],
  11680. device='cuda:0', dtype=torch.float64)
  11681. predictions are: tensor([[0.6740, 0.4580, 0.7269, 0.2416, 0.3939, 0.2809, 0.5639, 0.5228],
  11682. [0.6931, 0.4447, 0.8656, 0.4627, 0.3803, 0.4282, 0.6171, 0.5161],
  11683. [0.7225, 0.4748, 0.7947, 0.3181, 0.3930, 0.2757, 0.5183, 0.5355],
  11684. [0.7884, 0.5138, 0.8540, 0.5704, 0.4757, 0.4530, 0.5667, 0.5913],
  11685. [0.7516, 0.4976, 0.8647, 0.4521, 0.4303, 0.4952, 0.5614, 0.5281],
  11686. [0.2397, 0.1529, 0.8816, 0.2423, 0.5373, 0.2346, 0.7495, 0.5470],
  11687. [0.7616, 0.5046, 0.8529, 0.5398, 0.4444, 0.4879, 0.6154, 0.5691],
  11688. [0.7566, 0.5095, 0.8891, 0.4354, 0.4529, 0.5162, 0.6032, 0.5257]],
  11689. device='cuda:0', grad_fn=<AddmmBackward>)
  11690. landmarks are: tensor([[[0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  11691. [0.6236, 0.3965, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  11692. [0.6147, 0.4107, 0.8138, 0.3333, 0.3750, 0.2683, 0.5006, 0.5412],
  11693. [0.6248, 0.4185, 0.8500, 0.5767, 0.4462, 0.4550, 0.5612, 0.5917],
  11694. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  11695. [0.0000, 0.0000, 0.8850, 0.3000, 0.5362, 0.2250, 0.7343, 0.5771],
  11696. [0.6222, 0.4169, 0.8637, 0.5650, 0.4313, 0.4783, 0.5638, 0.5633],
  11697. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533]]],
  11698. device='cuda:0')
  11699. loss_train_step before backward: tensor(0.0039, device='cuda:0', grad_fn=<MseLossBackward>)
  11700. loss_train_step after backward: tensor(0.0039, device='cuda:0', grad_fn=<MseLossBackward>)
  11701. loss_train: 0.1288923279789742
  11702. step: 71
  11703. running loss: 0.0018153849011123128
  11704. Train Steps: 71/90 Loss: 0.0018 torch.Size([8, 600, 800])
  11705. torch.Size([8, 8])
  11706. tensor([[0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153],
  11707. [0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364],
  11708. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200],
  11709. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  11710. [0.6336, 0.4154, 0.8900, 0.2767, 0.4988, 0.2867, 0.7422, 0.5540],
  11711. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  11712. [0.6263, 0.4065, 0.9038, 0.4317, 0.3588, 0.4550, 0.6325, 0.5250],
  11713. [0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633]],
  11714. device='cuda:0', dtype=torch.float64)
  11715. predictions are: tensor([[0.6854, 0.4290, 0.8553, 0.5348, 0.4771, 0.5437, 0.5710, 0.5228],
  11716. [0.7214, 0.4744, 0.7999, 0.2458, 0.4623, 0.2501, 0.6241, 0.5392],
  11717. [0.6655, 0.4104, 0.8681, 0.3181, 0.4673, 0.2397, 0.6046, 0.5253],
  11718. [0.7321, 0.4662, 0.8887, 0.5671, 0.4110, 0.4492, 0.5422, 0.5732],
  11719. [0.5737, 0.3805, 0.8622, 0.3099, 0.5229, 0.3049, 0.6988, 0.5457],
  11720. [0.6894, 0.4463, 0.8710, 0.4764, 0.4769, 0.5825, 0.5848, 0.5603],
  11721. [0.7157, 0.4586, 0.8985, 0.4550, 0.4095, 0.4748, 0.6335, 0.5365],
  11722. [0.6618, 0.4263, 0.8856, 0.5448, 0.3901, 0.4869, 0.5635, 0.5772]],
  11723. device='cuda:0', grad_fn=<AddmmBackward>)
  11724. landmarks are: tensor([[[0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153],
  11725. [0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364],
  11726. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200],
  11727. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  11728. [0.6336, 0.4154, 0.8900, 0.2767, 0.4988, 0.2867, 0.7422, 0.5540],
  11729. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  11730. [0.6263, 0.4065, 0.9038, 0.4317, 0.3587, 0.4550, 0.6325, 0.5250],
  11731. [0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633]]],
  11732. device='cuda:0')
  11733. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  11734. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  11735. loss_train: 0.1301426654390525
  11736. step: 72
  11737. running loss: 0.0018075370199868404
  11738. Train Steps: 72/90 Loss: 0.0018 torch.Size([8, 600, 800])
  11739. torch.Size([8, 8])
  11740. tensor([[0.6263, 0.4233, 0.7924, 0.4626, 0.3788, 0.2883, 0.5573, 0.6047],
  11741. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  11742. [0.6085, 0.4005, 0.8400, 0.4317, 0.3763, 0.4750, 0.5476, 0.5058],
  11743. [0.6262, 0.4163, 0.8850, 0.5183, 0.3763, 0.4150, 0.6025, 0.5500],
  11744. [0.6198, 0.4076, 0.7945, 0.2453, 0.4475, 0.1871, 0.5897, 0.5350],
  11745. [0.6124, 0.4075, 0.7696, 0.4153, 0.3475, 0.3767, 0.5157, 0.5427],
  11746. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  11747. [0.6125, 0.4076, 0.8488, 0.3883, 0.3700, 0.3683, 0.5026, 0.5505]],
  11748. device='cuda:0', dtype=torch.float64)
  11749. predictions are: tensor([[0.6557, 0.4289, 0.8310, 0.4977, 0.4275, 0.3516, 0.5835, 0.6059],
  11750. [0.6078, 0.3892, 0.8621, 0.5863, 0.4557, 0.5347, 0.6241, 0.6111],
  11751. [0.6815, 0.4376, 0.8850, 0.4496, 0.4109, 0.5026, 0.5828, 0.5285],
  11752. [0.6125, 0.3737, 0.9020, 0.4991, 0.4120, 0.4265, 0.5849, 0.5473],
  11753. [0.7450, 0.4814, 0.8346, 0.2803, 0.4986, 0.2320, 0.6264, 0.5490],
  11754. [0.6894, 0.4417, 0.8345, 0.4276, 0.3791, 0.4259, 0.5487, 0.5544],
  11755. [0.6251, 0.3859, 0.8485, 0.2816, 0.4423, 0.3012, 0.6581, 0.5162],
  11756. [0.6389, 0.3790, 0.8842, 0.4048, 0.3998, 0.4133, 0.5169, 0.5496]],
  11757. device='cuda:0', grad_fn=<AddmmBackward>)
  11758. landmarks are: tensor([[[0.6263, 0.4232, 0.7924, 0.4626, 0.3787, 0.2883, 0.5573, 0.6047],
  11759. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  11760. [0.6084, 0.4005, 0.8400, 0.4317, 0.3762, 0.4750, 0.5476, 0.5058],
  11761. [0.6262, 0.4163, 0.8850, 0.5183, 0.3762, 0.4150, 0.6025, 0.5500],
  11762. [0.6198, 0.4076, 0.7945, 0.2453, 0.4475, 0.1871, 0.5897, 0.5350],
  11763. [0.6124, 0.4075, 0.7696, 0.4153, 0.3475, 0.3767, 0.5157, 0.5427],
  11764. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  11765. [0.6125, 0.4076, 0.8487, 0.3883, 0.3700, 0.3683, 0.5026, 0.5505]]],
  11766. device='cuda:0')
  11767. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  11768. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  11769. loss_train: 0.13164239979232661
  11770. step: 73
  11771. running loss: 0.0018033205451003646
  11772. Train Steps: 73/90 Loss: 0.0018 torch.Size([8, 600, 800])
  11773. torch.Size([8, 8])
  11774. tensor([[0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  11775. [ nan, nan, 0.8850, 0.3000, 0.5363, 0.2250, 0.7343, 0.5771],
  11776. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  11777. [0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  11778. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050],
  11779. [ nan, nan, 0.7612, 0.3250, 0.4037, 0.2533, 0.5438, 0.5767],
  11780. [0.6272, 0.4071, 0.8738, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  11781. [0.6264, 0.4248, 0.8938, 0.4183, 0.3875, 0.4100, 0.5400, 0.5350]],
  11782. device='cuda:0', dtype=torch.float64)
  11783. predictions are: tensor([[0.7152, 0.4496, 0.8840, 0.5291, 0.4344, 0.5146, 0.5446, 0.5550],
  11784. [0.2075, 0.1350, 0.8997, 0.3122, 0.5219, 0.2621, 0.7280, 0.5755],
  11785. [0.7042, 0.4605, 0.8933, 0.3257, 0.4916, 0.2425, 0.6603, 0.5587],
  11786. [0.7824, 0.4959, 0.8963, 0.5029, 0.3762, 0.4046, 0.5804, 0.6031],
  11787. [0.7340, 0.4657, 0.8780, 0.5537, 0.4509, 0.5468, 0.5906, 0.5369],
  11788. [0.2747, 0.1706, 0.8041, 0.3679, 0.4138, 0.2918, 0.5423, 0.6006],
  11789. [0.7430, 0.4759, 0.8953, 0.5835, 0.3869, 0.4296, 0.5988, 0.5045],
  11790. [0.7167, 0.4848, 0.8927, 0.4600, 0.3611, 0.4525, 0.5368, 0.5691]],
  11791. device='cuda:0', grad_fn=<AddmmBackward>)
  11792. landmarks are: tensor([[[0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  11793. [0.0000, 0.0000, 0.8850, 0.3000, 0.5362, 0.2250, 0.7343, 0.5771],
  11794. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  11795. [0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  11796. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050],
  11797. [0.0000, 0.0000, 0.7613, 0.3250, 0.4038, 0.2533, 0.5437, 0.5767],
  11798. [0.6272, 0.4071, 0.8737, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  11799. [0.6264, 0.4248, 0.8938, 0.4183, 0.3875, 0.4100, 0.5400, 0.5350]]],
  11800. device='cuda:0')
  11801. loss_train_step before backward: tensor(0.0047, device='cuda:0', grad_fn=<MseLossBackward>)
  11802. loss_train_step after backward: tensor(0.0047, device='cuda:0', grad_fn=<MseLossBackward>)
  11803. loss_train: 0.13630416334490292
  11804. step: 74
  11805. running loss: 0.001841948153309499
  11806.  
  11807. Train Steps: 74/90 Loss: 0.0018 torch.Size([8, 600, 800])
  11808. torch.Size([8, 8])
  11809. tensor([[ nan, nan, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  11810. [0.6250, 0.4103, 0.8950, 0.4400, 0.3912, 0.5650, 0.6050, 0.5133],
  11811. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  11812. [0.6147, 0.4112, 0.7988, 0.3200, 0.3775, 0.2767, 0.5150, 0.5550],
  11813. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284],
  11814. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  11815. [0.6109, 0.4015, 0.7668, 0.3639, 0.3513, 0.3667, 0.5200, 0.5641],
  11816. [0.6151, 0.4125, 0.8738, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483]],
  11817. device='cuda:0', dtype=torch.float64)
  11818. predictions are: tensor([[0.0903, 0.0535, 0.9101, 0.3129, 0.5345, 0.2583, 0.7578, 0.5770],
  11819. [0.6861, 0.4237, 0.9318, 0.5016, 0.4171, 0.6002, 0.6311, 0.5333],
  11820. [0.6454, 0.4002, 0.8520, 0.3570, 0.4135, 0.2628, 0.6123, 0.5308],
  11821. [0.6206, 0.4030, 0.8520, 0.3886, 0.3837, 0.2987, 0.5037, 0.5705],
  11822. [0.6418, 0.3951, 0.8160, 0.2760, 0.4705, 0.2130, 0.6188, 0.5424],
  11823. [0.6373, 0.3980, 0.8801, 0.6439, 0.4178, 0.4880, 0.5721, 0.6193],
  11824. [0.6372, 0.4049, 0.8112, 0.4223, 0.3590, 0.3866, 0.5287, 0.5580],
  11825. [0.6187, 0.3832, 0.9024, 0.5265, 0.3736, 0.4140, 0.5123, 0.5496]],
  11826. device='cuda:0', grad_fn=<AddmmBackward>)
  11827. landmarks are: tensor([[[0.0000, 0.0000, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  11828. [0.6250, 0.4103, 0.8950, 0.4400, 0.3913, 0.5650, 0.6050, 0.5133],
  11829. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  11830. [0.6147, 0.4112, 0.7987, 0.3200, 0.3775, 0.2767, 0.5150, 0.5550],
  11831. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284],
  11832. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  11833. [0.6109, 0.4015, 0.7668, 0.3639, 0.3512, 0.3667, 0.5200, 0.5641],
  11834. [0.6151, 0.4125, 0.8737, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483]]],
  11835. device='cuda:0')
  11836. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  11837. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  11838. loss_train: 0.13764630918740295
  11839. step: 75
  11840. running loss: 0.001835284122498706
  11841. Train Steps: 75/90 Loss: 0.0018 torch.Size([8, 600, 800])
  11842. torch.Size([8, 8])
  11843. tensor([[0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921],
  11844. [ nan, nan, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  11845. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  11846. [0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  11847. [ nan, nan, 0.6688, 0.2513, 0.4113, 0.2117, 0.5193, 0.5933],
  11848. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  11849. [ nan, nan, 0.6900, 0.1917, 0.3937, 0.2367, 0.5240, 0.5246],
  11850. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672]],
  11851. device='cuda:0', dtype=torch.float64)
  11852. predictions are: tensor([[0.6749, 0.4133, 0.8857, 0.2980, 0.4145, 0.2898, 0.6248, 0.5178],
  11853. [0.1693, 0.0882, 0.9349, 0.3785, 0.5117, 0.2394, 0.6746, 0.5773],
  11854. [0.7113, 0.4677, 0.8635, 0.3449, 0.4212, 0.2230, 0.5482, 0.5491],
  11855. [0.6812, 0.4329, 0.9257, 0.5342, 0.3713, 0.3869, 0.5587, 0.6029],
  11856. [0.1745, 0.1099, 0.7082, 0.3025, 0.4149, 0.2373, 0.5410, 0.5802],
  11857. [0.6236, 0.4131, 0.7854, 0.2501, 0.4213, 0.2076, 0.5841, 0.5242],
  11858. [0.0469, 0.0247, 0.7511, 0.2891, 0.4237, 0.2496, 0.5318, 0.5650],
  11859. [0.7042, 0.4386, 0.9340, 0.5136, 0.3466, 0.3730, 0.6089, 0.5196]],
  11860. device='cuda:0', grad_fn=<AddmmBackward>)
  11861. landmarks are: tensor([[[0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921],
  11862. [0.0000, 0.0000, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  11863. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  11864. [0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  11865. [0.0000, 0.0000, 0.6688, 0.2513, 0.4112, 0.2117, 0.5193, 0.5933],
  11866. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  11867. [0.0000, 0.0000, 0.6900, 0.1917, 0.3938, 0.2367, 0.5240, 0.5246],
  11868. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672]]],
  11869. device='cuda:0')
  11870. loss_train_step before backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  11871. loss_train_step after backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  11872. loss_train: 0.14028748925193213
  11873. step: 76
  11874. running loss: 0.0018458880164727912
  11875. Train Steps: 76/90 Loss: 0.0018 torch.Size([8, 600, 800])
  11876. torch.Size([8, 8])
  11877. tensor([[0.6277, 0.4118, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  11878. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  11879. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5637, 0.5633],
  11880. [0.6109, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  11881. [ nan, nan, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  11882. [0.6257, 0.4167, 0.8775, 0.3433, 0.3563, 0.4133, 0.6200, 0.5667],
  11883. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  11884. [0.6346, 0.4092, 0.7712, 0.5917, 0.4037, 0.4767, 0.7343, 0.5725]],
  11885. device='cuda:0', dtype=torch.float64)
  11886. predictions are: tensor([[0.5165, 0.3344, 0.8865, 0.3937, 0.3729, 0.2577, 0.5661, 0.4983],
  11887. [0.5008, 0.3059, 0.8650, 0.5040, 0.4121, 0.4633, 0.5061, 0.5278],
  11888. [0.5801, 0.3869, 0.8603, 0.5037, 0.3318, 0.3774, 0.5080, 0.5630],
  11889. [0.4800, 0.3126, 0.8583, 0.4781, 0.3460, 0.3829, 0.5058, 0.5076],
  11890. [0.0449, 0.0263, 0.9105, 0.3653, 0.4914, 0.2295, 0.6939, 0.5560],
  11891. [0.5305, 0.3265, 0.8573, 0.3543, 0.3323, 0.3917, 0.5863, 0.5516],
  11892. [0.5581, 0.3568, 0.8780, 0.3115, 0.4651, 0.2026, 0.6345, 0.5329],
  11893. [0.5063, 0.3323, 0.8051, 0.5204, 0.3592, 0.4526, 0.6360, 0.5617]],
  11894. device='cuda:0', grad_fn=<AddmmBackward>)
  11895. landmarks are: tensor([[[0.6277, 0.4117, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  11896. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  11897. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5638, 0.5633],
  11898. [0.6108, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  11899. [0.0000, 0.0000, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  11900. [0.6257, 0.4167, 0.8775, 0.3433, 0.3562, 0.4133, 0.6200, 0.5667],
  11901. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  11902. [0.6346, 0.4092, 0.7713, 0.5917, 0.4038, 0.4767, 0.7343, 0.5725]]],
  11903. device='cuda:0')
  11904. loss_train_step before backward: tensor(0.0025, device='cuda:0', grad_fn=<MseLossBackward>)
  11905. loss_train_step after backward: tensor(0.0025, device='cuda:0', grad_fn=<MseLossBackward>)
  11906. loss_train: 0.14280669260188006
  11907. step: 77
  11908. running loss: 0.0018546323714529879
  11909. Train Steps: 77/90 Loss: 0.0019 torch.Size([8, 600, 800])
  11910. torch.Size([8, 8])
  11911. tensor([[0.6224, 0.4097, 0.7438, 0.2267, 0.3850, 0.2850, 0.5988, 0.5250],
  11912. [0.6057, 0.4011, 0.8750, 0.4267, 0.4400, 0.5800, 0.5845, 0.5585],
  11913. [0.6212, 0.4033, 0.8938, 0.4167, 0.3813, 0.4267, 0.5613, 0.5583],
  11914. [0.6126, 0.4073, 0.8750, 0.5133, 0.3800, 0.4333, 0.4986, 0.5378],
  11915. [0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332],
  11916. [0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  11917. [0.6175, 0.4091, 0.7863, 0.2800, 0.3638, 0.3583, 0.6188, 0.5433],
  11918. [0.6111, 0.4019, 0.8350, 0.3867, 0.3500, 0.4283, 0.5480, 0.5719]],
  11919. device='cuda:0', dtype=torch.float64)
  11920. predictions are: tensor([[0.5059, 0.3345, 0.7602, 0.2278, 0.3854, 0.2539, 0.6262, 0.5230],
  11921. [0.4361, 0.2765, 0.8633, 0.4468, 0.4724, 0.5165, 0.5810, 0.5478],
  11922. [0.4907, 0.3082, 0.9063, 0.4370, 0.3787, 0.3822, 0.5449, 0.5404],
  11923. [0.4585, 0.2929, 0.8790, 0.5238, 0.3940, 0.4032, 0.4904, 0.5269],
  11924. [0.4985, 0.3049, 0.9053, 0.4899, 0.3699, 0.3515, 0.6380, 0.5238],
  11925. [0.4820, 0.3069, 0.8524, 0.5386, 0.3745, 0.4547, 0.6722, 0.5328],
  11926. [0.5487, 0.3470, 0.7873, 0.2789, 0.3601, 0.3334, 0.6035, 0.5362],
  11927. [0.4773, 0.2953, 0.8434, 0.4209, 0.3548, 0.3967, 0.5461, 0.5606]],
  11928. device='cuda:0', grad_fn=<AddmmBackward>)
  11929. landmarks are: tensor([[[0.6224, 0.4097, 0.7437, 0.2267, 0.3850, 0.2850, 0.5987, 0.5250],
  11930. [0.6057, 0.4011, 0.8750, 0.4267, 0.4400, 0.5800, 0.5845, 0.5585],
  11931. [0.6212, 0.4033, 0.8938, 0.4167, 0.3812, 0.4267, 0.5612, 0.5583],
  11932. [0.6126, 0.4073, 0.8750, 0.5133, 0.3800, 0.4333, 0.4986, 0.5378],
  11933. [0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332],
  11934. [0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  11935. [0.6175, 0.4091, 0.7862, 0.2800, 0.3638, 0.3583, 0.6187, 0.5433],
  11936. [0.6111, 0.4019, 0.8350, 0.3867, 0.3500, 0.4283, 0.5480, 0.5719]]],
  11937. device='cuda:0')
  11938. loss_train_step before backward: tensor(0.0039, device='cuda:0', grad_fn=<MseLossBackward>)
  11939. loss_train_step after backward: tensor(0.0039, device='cuda:0', grad_fn=<MseLossBackward>)
  11940. loss_train: 0.14673250037594698
  11941. step: 78
  11942. running loss: 0.0018811859022557305
  11943.  
  11944. Train Steps: 78/90 Loss: 0.0019 torch.Size([8, 600, 800])
  11945. torch.Size([8, 8])
  11946. tensor([[0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250],
  11947. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600],
  11948. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  11949. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  11950. [0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  11951. [0.6170, 0.4102, 0.7468, 0.3695, 0.3463, 0.3767, 0.5238, 0.5823],
  11952. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116],
  11953. [0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921]],
  11954. device='cuda:0', dtype=torch.float64)
  11955. predictions are: tensor([[0.4087, 0.2881, 0.8793, 0.3682, 0.3442, 0.2797, 0.6288, 0.5055],
  11956. [0.4526, 0.2877, 0.8537, 0.5116, 0.3869, 0.4292, 0.5877, 0.5334],
  11957. [0.4888, 0.3502, 0.7298, 0.2433, 0.3949, 0.1769, 0.5642, 0.5608],
  11958. [0.4653, 0.2900, 0.8206, 0.5220, 0.3491, 0.4446, 0.6688, 0.4878],
  11959. [0.4368, 0.2940, 0.8106, 0.4830, 0.4568, 0.4552, 0.5322, 0.5297],
  11960. [0.4606, 0.2997, 0.7351, 0.3265, 0.3292, 0.3455, 0.5507, 0.5565],
  11961. [0.4449, 0.2896, 0.8538, 0.3646, 0.3796, 0.5156, 0.5988, 0.5139],
  11962. [0.4937, 0.3109, 0.8220, 0.1960, 0.3931, 0.2451, 0.6621, 0.4891]],
  11963. device='cuda:0', grad_fn=<AddmmBackward>)
  11964. landmarks are: tensor([[[0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250],
  11965. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600],
  11966. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  11967. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  11968. [0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  11969. [0.6170, 0.4102, 0.7468, 0.3695, 0.3462, 0.3767, 0.5238, 0.5823],
  11970. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116],
  11971. [0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921]]],
  11972. device='cuda:0')
  11973. loss_train_step before backward: tensor(0.0059, device='cuda:0', grad_fn=<MseLossBackward>)
  11974. loss_train_step after backward: tensor(0.0059, device='cuda:0', grad_fn=<MseLossBackward>)
  11975. loss_train: 0.15266501865698956
  11976. step: 79
  11977. running loss: 0.0019324685905948045
  11978. Train Steps: 79/90 Loss: 0.0019 torch.Size([8, 600, 800])
  11979. torch.Size([8, 8])
  11980. tensor([[0.6193, 0.4050, 0.7313, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  11981. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650],
  11982. [0.6200, 0.4071, 0.7338, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  11983. [ nan, nan, 0.8300, 0.3150, 0.3588, 0.3383, 0.5208, 0.5194],
  11984. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  11985. [0.6179, 0.3993, 0.8925, 0.4789, 0.3879, 0.4900, 0.6041, 0.5279],
  11986. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  11987. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395]],
  11988. device='cuda:0', dtype=torch.float64)
  11989. predictions are: tensor([[0.5307, 0.3548, 0.7096, 0.2187, 0.4055, 0.2188, 0.5669, 0.5435],
  11990. [0.4671, 0.3458, 0.8251, 0.3786, 0.3618, 0.3229, 0.5673, 0.5475],
  11991. [0.4901, 0.3563, 0.7372, 0.1861, 0.4123, 0.2281, 0.6333, 0.5263],
  11992. [0.0950, 0.0853, 0.7791, 0.2953, 0.3558, 0.3173, 0.5706, 0.5129],
  11993. [0.5282, 0.3683, 0.6496, 0.2782, 0.3489, 0.2879, 0.5350, 0.5610],
  11994. [0.5229, 0.3438, 0.8743, 0.4487, 0.3905, 0.4842, 0.6071, 0.5113],
  11995. [0.4942, 0.3315, 0.7959, 0.2371, 0.3885, 0.2666, 0.6473, 0.4877],
  11996. [0.5228, 0.3476, 0.7379, 0.1952, 0.3836, 0.2282, 0.6192, 0.5012]],
  11997. device='cuda:0', grad_fn=<AddmmBackward>)
  11998. landmarks are: tensor([[[0.6193, 0.4050, 0.7312, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  11999. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650],
  12000. [0.6200, 0.4071, 0.7337, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  12001. [0.0000, 0.0000, 0.8300, 0.3150, 0.3587, 0.3383, 0.5208, 0.5194],
  12002. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  12003. [0.6179, 0.3993, 0.8925, 0.4789, 0.3879, 0.4900, 0.6041, 0.5279],
  12004. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  12005. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395]]],
  12006. device='cuda:0')
  12007. loss_train_step before backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  12008. loss_train_step after backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  12009. loss_train: 0.15501353502622806
  12010. step: 80
  12011. running loss: 0.0019376691878278506
  12012. Train Steps: 80/90 Loss: 0.0019 torch.Size([8, 600, 800])
  12013. torch.Size([8, 8])
  12014. tensor([[0.6257, 0.4167, 0.8775, 0.3433, 0.3563, 0.4133, 0.6200, 0.5667],
  12015. [0.6212, 0.4171, 0.7875, 0.3633, 0.3813, 0.2933, 0.5675, 0.5700],
  12016. [ nan, nan, 0.7981, 0.3194, 0.3625, 0.3167, 0.5040, 0.5563],
  12017. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  12018. [0.6296, 0.4076, 0.8400, 0.5583, 0.3700, 0.4367, 0.6876, 0.5494],
  12019. [ nan, nan, 0.7425, 0.2117, 0.3937, 0.2433, 0.5438, 0.5567],
  12020. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  12021. [0.6149, 0.4054, 0.6713, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695]],
  12022. device='cuda:0', dtype=torch.float64)
  12023. predictions are: tensor([[0.6282, 0.4381, 0.8355, 0.2734, 0.3580, 0.4126, 0.6383, 0.5545],
  12024. [0.6837, 0.4846, 0.7717, 0.3039, 0.3803, 0.3083, 0.5922, 0.5750],
  12025. [0.0733, 0.0855, 0.7373, 0.2484, 0.3598, 0.3237, 0.5362, 0.5699],
  12026. [0.6103, 0.4397, 0.8611, 0.3480, 0.3714, 0.3243, 0.5854, 0.5111],
  12027. [0.6272, 0.4369, 0.8275, 0.4628, 0.3831, 0.4366, 0.6994, 0.5166],
  12028. [0.0427, 0.0495, 0.7091, 0.1944, 0.4089, 0.2482, 0.5838, 0.5537],
  12029. [0.6725, 0.4609, 0.7970, 0.5225, 0.3743, 0.5087, 0.5892, 0.4707],
  12030. [0.5609, 0.4096, 0.6612, 0.1675, 0.4200, 0.2040, 0.5145, 0.5588]],
  12031. device='cuda:0', grad_fn=<AddmmBackward>)
  12032. landmarks are: tensor([[[0.6257, 0.4167, 0.8775, 0.3433, 0.3562, 0.4133, 0.6200, 0.5667],
  12033. [0.6212, 0.4171, 0.7875, 0.3633, 0.3812, 0.2933, 0.5675, 0.5700],
  12034. [0.0000, 0.0000, 0.7981, 0.3194, 0.3625, 0.3167, 0.5040, 0.5563],
  12035. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  12036. [0.6296, 0.4076, 0.8400, 0.5583, 0.3700, 0.4367, 0.6876, 0.5494],
  12037. [0.0000, 0.0000, 0.7425, 0.2117, 0.3938, 0.2433, 0.5437, 0.5567],
  12038. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  12039. [0.6149, 0.4054, 0.6712, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695]]],
  12040. device='cuda:0')
  12041. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  12042. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  12043. loss_train: 0.15654646544135176
  12044. step: 81
  12045. running loss: 0.0019326724128561946
  12046. Train Steps: 81/90 Loss: 0.0019 torch.Size([8, 600, 800])
  12047. torch.Size([8, 8])
  12048. tensor([[0.6250, 0.3993, 0.9138, 0.4333, 0.3763, 0.5217, 0.6995, 0.5320],
  12049. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  12050. [0.6224, 0.4097, 0.7438, 0.2267, 0.3850, 0.2850, 0.5988, 0.5250],
  12051. [0.6203, 0.4021, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  12052. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  12053. [ nan, nan, 0.7981, 0.3194, 0.3625, 0.3167, 0.5040, 0.5563],
  12054. [0.6200, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  12055. [0.6140, 0.4070, 0.8700, 0.5000, 0.4612, 0.4900, 0.5260, 0.5852]],
  12056. device='cuda:0', dtype=torch.float64)
  12057. predictions are: tensor([[0.6120, 0.4161, 0.8576, 0.3980, 0.3587, 0.5396, 0.7251, 0.5467],
  12058. [0.5780, 0.4041, 0.8139, 0.4110, 0.3942, 0.5006, 0.5479, 0.5276],
  12059. [0.6369, 0.4528, 0.6879, 0.1798, 0.3727, 0.2992, 0.6297, 0.5245],
  12060. [0.7181, 0.4893, 0.8373, 0.4649, 0.3389, 0.4007, 0.5916, 0.5105],
  12061. [0.6550, 0.4485, 0.8356, 0.3766, 0.3475, 0.3610, 0.6218, 0.5490],
  12062. [0.0318, 0.0481, 0.7165, 0.2537, 0.3442, 0.3365, 0.5431, 0.5739],
  12063. [0.6474, 0.4496, 0.7977, 0.2374, 0.4135, 0.2600, 0.5980, 0.5348],
  12064. [0.5745, 0.4167, 0.8119, 0.4426, 0.4552, 0.4981, 0.5370, 0.5871]],
  12065. device='cuda:0', grad_fn=<AddmmBackward>)
  12066. landmarks are: tensor([[[0.6250, 0.3993, 0.9137, 0.4333, 0.3762, 0.5217, 0.6995, 0.5320],
  12067. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  12068. [0.6224, 0.4097, 0.7437, 0.2267, 0.3850, 0.2850, 0.5987, 0.5250],
  12069. [0.6203, 0.4020, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  12070. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  12071. [0.0000, 0.0000, 0.7981, 0.3194, 0.3625, 0.3167, 0.5040, 0.5563],
  12072. [0.6199, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  12073. [0.6140, 0.4070, 0.8700, 0.5000, 0.4613, 0.4900, 0.5260, 0.5852]]],
  12074. device='cuda:0')
  12075. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  12076. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  12077. loss_train: 0.1579744432528969
  12078. step: 82
  12079. running loss: 0.001926517600645084
  12080.  
  12081. Train Steps: 82/90 Loss: 0.0019 torch.Size([8, 600, 800])
  12082. torch.Size([8, 8])
  12083. tensor([[0.6249, 0.4142, 0.8350, 0.3283, 0.3613, 0.3700, 0.6188, 0.5400],
  12084. [0.6193, 0.4050, 0.7313, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  12085. [0.6200, 0.4071, 0.7338, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  12086. [ nan, nan, 0.8888, 0.3100, 0.5262, 0.2817, 0.7145, 0.6003],
  12087. [0.6250, 0.4103, 0.8950, 0.4400, 0.3912, 0.5650, 0.6050, 0.5133],
  12088. [0.6182, 0.3930, 0.8841, 0.3892, 0.3556, 0.4967, 0.6222, 0.5279],
  12089. [0.6129, 0.4069, 0.8750, 0.5067, 0.3875, 0.4233, 0.5235, 0.5881],
  12090. [0.6197, 0.4050, 0.7527, 0.2000, 0.4042, 0.2249, 0.5895, 0.4995]],
  12091. device='cuda:0', dtype=torch.float64)
  12092. predictions are: tensor([[0.6289, 0.4397, 0.7994, 0.3143, 0.3484, 0.3620, 0.6211, 0.5480],
  12093. [0.7012, 0.4679, 0.6978, 0.2448, 0.4033, 0.2456, 0.5379, 0.5728],
  12094. [0.6532, 0.4595, 0.7230, 0.2023, 0.4137, 0.2560, 0.6047, 0.5566],
  12095. [0.0559, 0.0454, 0.8784, 0.3179, 0.4937, 0.2871, 0.7252, 0.5912],
  12096. [0.7048, 0.4770, 0.8636, 0.4523, 0.3856, 0.6014, 0.5997, 0.5220],
  12097. [0.5908, 0.3908, 0.8337, 0.4236, 0.3456, 0.4994, 0.6183, 0.5246],
  12098. [0.6762, 0.4809, 0.8649, 0.5304, 0.3847, 0.4387, 0.4774, 0.5786],
  12099. [0.6734, 0.4679, 0.7195, 0.1926, 0.3988, 0.2341, 0.5801, 0.5138]],
  12100. device='cuda:0', grad_fn=<AddmmBackward>)
  12101. landmarks are: tensor([[[0.6249, 0.4142, 0.8350, 0.3283, 0.3613, 0.3700, 0.6187, 0.5400],
  12102. [0.6193, 0.4050, 0.7312, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  12103. [0.6200, 0.4071, 0.7337, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  12104. [0.0000, 0.0000, 0.8888, 0.3100, 0.5263, 0.2817, 0.7145, 0.6003],
  12105. [0.6250, 0.4103, 0.8950, 0.4400, 0.3913, 0.5650, 0.6050, 0.5133],
  12106. [0.6182, 0.3930, 0.8841, 0.3892, 0.3556, 0.4967, 0.6222, 0.5279],
  12107. [0.6129, 0.4069, 0.8750, 0.5067, 0.3875, 0.4233, 0.5235, 0.5881],
  12108. [0.6197, 0.4050, 0.7527, 0.2000, 0.4042, 0.2249, 0.5895, 0.4995]]],
  12109. device='cuda:0')
  12110. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  12111. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  12112. loss_train: 0.15900916929240339
  12113. step: 83
  12114. running loss: 0.0019157731240048602
  12115. Train Steps: 83/90 Loss: 0.0019 torch.Size([8, 600, 800])
  12116. torch.Size([8, 8])
  12117. tensor([[0.6135, 0.4115, 0.8838, 0.4667, 0.4288, 0.6050, 0.5778, 0.5097],
  12118. [0.6264, 0.3972, 0.8853, 0.4771, 0.3853, 0.4511, 0.6293, 0.5334],
  12119. [0.6250, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6088, 0.5183],
  12120. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6063, 0.6083],
  12121. [0.6202, 0.4064, 0.7879, 0.2179, 0.4567, 0.1725, 0.5955, 0.5478],
  12122. [0.6185, 0.4079, 0.8838, 0.4617, 0.4838, 0.5650, 0.6175, 0.5850],
  12123. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  12124. [0.6167, 0.4048, 0.6831, 0.3639, 0.3763, 0.3017, 0.5700, 0.5883]],
  12125. device='cuda:0', dtype=torch.float64)
  12126. predictions are: tensor([[0.6739, 0.4526, 0.8910, 0.4254, 0.4190, 0.5681, 0.5595, 0.5604],
  12127. [0.6833, 0.4371, 0.8709, 0.4826, 0.3629, 0.4658, 0.6200, 0.5308],
  12128. [0.6546, 0.4241, 0.8739, 0.4820, 0.4487, 0.5598, 0.6135, 0.5418],
  12129. [0.6296, 0.4391, 0.8380, 0.3504, 0.3509, 0.4107, 0.5820, 0.6039],
  12130. [0.5579, 0.3854, 0.7547, 0.2289, 0.4464, 0.1491, 0.5852, 0.5591],
  12131. [0.6199, 0.3966, 0.8858, 0.4370, 0.4615, 0.5330, 0.6153, 0.5827],
  12132. [0.6631, 0.4274, 0.7721, 0.4041, 0.3427, 0.4250, 0.5122, 0.5284],
  12133. [0.5237, 0.3468, 0.7270, 0.3245, 0.3668, 0.2981, 0.5689, 0.5847]],
  12134. device='cuda:0', grad_fn=<AddmmBackward>)
  12135. landmarks are: tensor([[[0.6135, 0.4115, 0.8838, 0.4667, 0.4288, 0.6050, 0.5778, 0.5097],
  12136. [0.6264, 0.3972, 0.8853, 0.4771, 0.3853, 0.4511, 0.6293, 0.5334],
  12137. [0.6251, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6087, 0.5183],
  12138. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6062, 0.6083],
  12139. [0.6202, 0.4064, 0.7879, 0.2179, 0.4567, 0.1725, 0.5955, 0.5478],
  12140. [0.6184, 0.4079, 0.8838, 0.4617, 0.4837, 0.5650, 0.6175, 0.5850],
  12141. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  12142. [0.6167, 0.4048, 0.6831, 0.3639, 0.3762, 0.3017, 0.5700, 0.5883]]],
  12143. device='cuda:0')
  12144. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  12145. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  12146. loss_train: 0.15980508815846406
  12147. step: 84
  12148. running loss: 0.0019024415256960008
  12149. Train Steps: 84/90 Loss: 0.0019 torch.Size([8, 600, 800])
  12150. torch.Size([8, 8])
  12151. tensor([[0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  12152. [0.6245, 0.4100, 0.7762, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  12153. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  12154. [0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500],
  12155. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  12156. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  12157. [0.6124, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  12158. [0.6185, 0.4079, 0.8838, 0.4617, 0.4838, 0.5650, 0.6175, 0.5850]],
  12159. device='cuda:0', dtype=torch.float64)
  12160. predictions are: tensor([[0.6171, 0.3926, 0.8730, 0.5601, 0.3915, 0.4549, 0.5773, 0.5414],
  12161. [0.6341, 0.4047, 0.7808, 0.2310, 0.4821, 0.1496, 0.5899, 0.5704],
  12162. [0.7446, 0.4741, 0.8923, 0.5481, 0.3716, 0.4588, 0.5961, 0.5184],
  12163. [0.6517, 0.4116, 0.7889, 0.2948, 0.3571, 0.4012, 0.5962, 0.5773],
  12164. [0.6654, 0.4209, 0.9023, 0.4934, 0.4563, 0.5767, 0.6043, 0.5427],
  12165. [0.6486, 0.4106, 0.8826, 0.4646, 0.3893, 0.5222, 0.5684, 0.5229],
  12166. [0.7261, 0.4736, 0.8319, 0.5206, 0.3985, 0.4609, 0.5244, 0.5686],
  12167. [0.6244, 0.3855, 0.9119, 0.4555, 0.4923, 0.5583, 0.5995, 0.5956]],
  12168. device='cuda:0', grad_fn=<AddmmBackward>)
  12169. landmarks are: tensor([[[0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  12170. [0.6245, 0.4100, 0.7763, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  12171. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  12172. [0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500],
  12173. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  12174. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  12175. [0.6123, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  12176. [0.6184, 0.4079, 0.8838, 0.4617, 0.4837, 0.5650, 0.6175, 0.5850]]],
  12177. device='cuda:0')
  12178. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  12179. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  12180. loss_train: 0.16067833415581845
  12181. step: 85
  12182. running loss: 0.0018903333430096288
  12183. Train Steps: 85/90 Loss: 0.0019 torch.Size([8, 600, 800])
  12184. torch.Size([8, 8])
  12185. tensor([[0.6120, 0.4014, 0.6863, 0.2817, 0.3700, 0.2783, 0.5513, 0.5667],
  12186. [0.6215, 0.4119, 0.7688, 0.2300, 0.4200, 0.2283, 0.5925, 0.5317],
  12187. [0.6138, 0.4101, 0.8800, 0.5083, 0.4637, 0.5950, 0.5587, 0.5077],
  12188. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5787, 0.5117],
  12189. [0.6308, 0.3990, 0.8688, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  12190. [0.6142, 0.3982, 0.8650, 0.4883, 0.3912, 0.4317, 0.5315, 0.5350],
  12191. [0.6277, 0.4029, 0.8250, 0.2433, 0.4325, 0.2100, 0.6366, 0.5207],
  12192. [0.6219, 0.4097, 0.8738, 0.3400, 0.3563, 0.4117, 0.5975, 0.5683]],
  12193. device='cuda:0', dtype=torch.float64)
  12194. predictions are: tensor([[0.6802, 0.4376, 0.7374, 0.3089, 0.3743, 0.2980, 0.5208, 0.5690],
  12195. [0.4057, 0.2461, 0.7853, 0.2706, 0.4651, 0.2423, 0.5779, 0.5631],
  12196. [0.6088, 0.3779, 0.9175, 0.5766, 0.4961, 0.5800, 0.5669, 0.5666],
  12197. [0.7017, 0.4283, 0.7751, 0.2334, 0.4296, 0.2204, 0.5703, 0.5378],
  12198. [0.6362, 0.3832, 0.8968, 0.5974, 0.4216, 0.5279, 0.6449, 0.5289],
  12199. [0.6274, 0.3839, 0.9022, 0.5466, 0.4141, 0.4581, 0.5066, 0.5398],
  12200. [0.6415, 0.3878, 0.8510, 0.3102, 0.4527, 0.2484, 0.6664, 0.5180],
  12201. [0.7252, 0.4803, 0.9016, 0.4021, 0.3710, 0.4009, 0.5595, 0.5757]],
  12202. device='cuda:0', grad_fn=<AddmmBackward>)
  12203. landmarks are: tensor([[[0.6120, 0.4013, 0.6862, 0.2817, 0.3700, 0.2783, 0.5512, 0.5667],
  12204. [0.6215, 0.4119, 0.7688, 0.2300, 0.4200, 0.2283, 0.5925, 0.5317],
  12205. [0.6138, 0.4101, 0.8800, 0.5083, 0.4638, 0.5950, 0.5587, 0.5077],
  12206. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5788, 0.5117],
  12207. [0.6308, 0.3990, 0.8687, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  12208. [0.6143, 0.3982, 0.8650, 0.4883, 0.3913, 0.4317, 0.5315, 0.5350],
  12209. [0.6277, 0.4029, 0.8250, 0.2433, 0.4325, 0.2100, 0.6366, 0.5207],
  12210. [0.6219, 0.4097, 0.8737, 0.3400, 0.3562, 0.4117, 0.5975, 0.5683]]],
  12211. device='cuda:0')
  12212. loss_train_step before backward: tensor(0.0025, device='cuda:0', grad_fn=<MseLossBackward>)
  12213. loss_train_step after backward: tensor(0.0025, device='cuda:0', grad_fn=<MseLossBackward>)
  12214. loss_train: 0.1631579902896192
  12215. step: 86
  12216. running loss: 0.0018971859336002232
  12217.  
  12218. Train Steps: 86/90 Loss: 0.0019 torch.Size([8, 600, 800])
  12219. torch.Size([8, 8])
  12220. tensor([[0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  12221. [ nan, nan, 0.7425, 0.2117, 0.3937, 0.2433, 0.5438, 0.5567],
  12222. [0.6274, 0.4099, 0.8625, 0.3233, 0.4400, 0.1983, 0.5876, 0.4869],
  12223. [0.6125, 0.3999, 0.8750, 0.4883, 0.4750, 0.4700, 0.5533, 0.5617],
  12224. [0.6188, 0.4099, 0.7400, 0.2433, 0.3962, 0.2750, 0.6162, 0.5467],
  12225. [0.6199, 0.4102, 0.8950, 0.4417, 0.4012, 0.5367, 0.6112, 0.5967],
  12226. [ nan, nan, 0.7850, 0.2700, 0.4288, 0.1717, 0.5199, 0.4999],
  12227. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301]],
  12228. device='cuda:0', dtype=torch.float64)
  12229. predictions are: tensor([[ 6.5615e-01, 4.3238e-01, 8.7637e-01, 3.0122e-01, 4.6985e-01,
  12230. 2.7266e-01, 6.8688e-01, 5.6284e-01],
  12231. [ 5.5656e-02, -1.0218e-04, 7.6715e-01, 2.8529e-01, 4.1903e-01,
  12232. 2.4294e-01, 5.3250e-01, 5.5664e-01],
  12233. [ 7.3798e-01, 4.6379e-01, 8.9042e-01, 3.5780e-01, 4.6348e-01,
  12234. 2.0265e-01, 5.6376e-01, 5.0279e-01],
  12235. [ 8.2466e-01, 5.1510e-01, 8.9817e-01, 5.4252e-01, 4.8617e-01,
  12236. 4.8849e-01, 5.3407e-01, 5.4021e-01],
  12237. [ 7.7672e-01, 4.9967e-01, 7.5344e-01, 2.7726e-01, 3.9371e-01,
  12238. 2.8717e-01, 6.0870e-01, 5.4520e-01],
  12239. [ 7.9712e-01, 4.9606e-01, 9.1677e-01, 5.0082e-01, 4.2626e-01,
  12240. 5.5632e-01, 5.9605e-01, 5.7758e-01],
  12241. [ 6.8887e-02, 1.3605e-02, 8.0388e-01, 2.7323e-01, 4.5281e-01,
  12242. 1.8833e-01, 5.1119e-01, 5.3303e-01],
  12243. [ 7.6316e-01, 4.6732e-01, 8.8208e-01, 5.5342e-01, 4.2469e-01,
  12244. 5.1457e-01, 6.1878e-01, 5.2243e-01]], device='cuda:0',
  12245. grad_fn=<AddmmBackward>)
  12246. landmarks are: tensor([[[0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  12247. [0.0000, 0.0000, 0.7425, 0.2117, 0.3938, 0.2433, 0.5437, 0.5567],
  12248. [0.6274, 0.4099, 0.8625, 0.3233, 0.4400, 0.1983, 0.5876, 0.4869],
  12249. [0.6125, 0.3999, 0.8750, 0.4883, 0.4750, 0.4700, 0.5533, 0.5617],
  12250. [0.6188, 0.4099, 0.7400, 0.2433, 0.3963, 0.2750, 0.6162, 0.5467],
  12251. [0.6199, 0.4102, 0.8950, 0.4417, 0.4013, 0.5367, 0.6112, 0.5967],
  12252. [0.0000, 0.0000, 0.7850, 0.2700, 0.4288, 0.1717, 0.5199, 0.4999],
  12253. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301]]],
  12254. device='cuda:0')
  12255. loss_train_step before backward: tensor(0.0033, device='cuda:0', grad_fn=<MseLossBackward>)
  12256. loss_train_step after backward: tensor(0.0033, device='cuda:0', grad_fn=<MseLossBackward>)
  12257. loss_train: 0.16645036087720655
  12258. step: 87
  12259. running loss: 0.001913222538818466
  12260. Train Steps: 87/90 Loss: 0.0019 torch.Size([8, 600, 800])
  12261. torch.Size([8, 8])
  12262. tensor([[0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  12263. [0.6243, 0.4128, 0.7762, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417],
  12264. [0.6339, 0.4123, 0.8638, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  12265. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  12266. [0.6250, 0.4236, 0.8638, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  12267. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  12268. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  12269. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103]],
  12270. device='cuda:0', dtype=torch.float64)
  12271. predictions are: tensor([[0.6109, 0.3798, 0.9016, 0.5194, 0.4766, 0.5684, 0.6182, 0.5540],
  12272. [0.6222, 0.3921, 0.7879, 0.2932, 0.4149, 0.2974, 0.6207, 0.5324],
  12273. [0.5676, 0.3614, 0.8883, 0.5625, 0.4245, 0.5565, 0.7536, 0.5352],
  12274. [0.5486, 0.3695, 0.8816, 0.4359, 0.3645, 0.3880, 0.5915, 0.5291],
  12275. [0.5466, 0.3531, 0.8790, 0.4196, 0.4184, 0.3233, 0.5451, 0.5796],
  12276. [0.5856, 0.3652, 0.7095, 0.2472, 0.4402, 0.1400, 0.5354, 0.5184],
  12277. [0.6644, 0.4115, 0.8798, 0.4273, 0.3535, 0.3874, 0.6053, 0.5026],
  12278. [0.5934, 0.3878, 0.8912, 0.4718, 0.4126, 0.4666, 0.5440, 0.5000]],
  12279. device='cuda:0', grad_fn=<AddmmBackward>)
  12280. landmarks are: tensor([[[0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  12281. [0.6243, 0.4128, 0.7763, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417],
  12282. [0.6339, 0.4123, 0.8637, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  12283. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  12284. [0.6250, 0.4236, 0.8637, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  12285. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  12286. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  12287. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103]]],
  12288. device='cuda:0')
  12289. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  12290. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  12291. loss_train: 0.1671832787396852
  12292. step: 88
  12293. running loss: 0.0018998099856782408
  12294. Train Steps: 88/90 Loss: 0.0019 torch.Size([8, 600, 800])
  12295. torch.Size([8, 8])
  12296. tensor([[0.6265, 0.4088, 0.8025, 0.1850, 0.4163, 0.2500, 0.6290, 0.4947],
  12297. [0.6268, 0.4029, 0.8500, 0.2683, 0.3937, 0.3500, 0.6860, 0.5297],
  12298. [0.6075, 0.4000, 0.8513, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  12299. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  12300. [0.6197, 0.4118, 0.8688, 0.5517, 0.4037, 0.5233, 0.5875, 0.5600],
  12301. [0.6267, 0.4080, 0.8438, 0.2633, 0.4763, 0.1800, 0.6259, 0.5240],
  12302. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5513, 0.5750],
  12303. [0.6159, 0.4085, 0.6900, 0.2283, 0.4088, 0.1950, 0.5123, 0.5397]],
  12304. device='cuda:0', dtype=torch.float64)
  12305. predictions are: tensor([[0.5255, 0.3113, 0.8251, 0.2187, 0.4319, 0.2542, 0.6413, 0.5076],
  12306. [0.6474, 0.4166, 0.8514, 0.3105, 0.3775, 0.3558, 0.6948, 0.5177],
  12307. [0.6135, 0.3969, 0.8811, 0.5613, 0.4470, 0.5547, 0.5503, 0.5267],
  12308. [0.5737, 0.3469, 0.7954, 0.2748, 0.4297, 0.2041, 0.5773, 0.5026],
  12309. [0.5767, 0.3556, 0.9012, 0.5930, 0.4184, 0.5899, 0.6142, 0.5412],
  12310. [0.5345, 0.3243, 0.8682, 0.2854, 0.4664, 0.2013, 0.6259, 0.5182],
  12311. [0.4622, 0.2721, 0.7097, 0.2735, 0.4363, 0.2165, 0.5343, 0.5638],
  12312. [0.5916, 0.3473, 0.7170, 0.2494, 0.4071, 0.2011, 0.5076, 0.5404]],
  12313. device='cuda:0', grad_fn=<AddmmBackward>)
  12314. landmarks are: tensor([[[0.6265, 0.4088, 0.8025, 0.1850, 0.4162, 0.2500, 0.6290, 0.4947],
  12315. [0.6268, 0.4029, 0.8500, 0.2683, 0.3938, 0.3500, 0.6860, 0.5297],
  12316. [0.6075, 0.4000, 0.8512, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  12317. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  12318. [0.6197, 0.4118, 0.8687, 0.5517, 0.4038, 0.5233, 0.5875, 0.5600],
  12319. [0.6267, 0.4080, 0.8438, 0.2633, 0.4762, 0.1800, 0.6259, 0.5240],
  12320. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5512, 0.5750],
  12321. [0.6159, 0.4085, 0.6900, 0.2283, 0.4087, 0.1950, 0.5123, 0.5397]]],
  12322. device='cuda:0')
  12323. loss_train_step before backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  12324. loss_train_step after backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  12325. loss_train: 0.16899637822643854
  12326. step: 89
  12327. running loss: 0.0018988357104094218
  12328. Train Steps: 89/90 Loss: 0.0019 torch.Size([8, 600, 800])
  12329. torch.Size([8, 8])
  12330. tensor([[0.6118, 0.4052, 0.8463, 0.3917, 0.3538, 0.3450, 0.5053, 0.5593],
  12331. [0.6226, 0.4103, 0.8575, 0.3450, 0.4388, 0.2067, 0.5787, 0.5383],
  12332. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  12333. [ nan, nan, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  12334. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  12335. [0.6277, 0.4118, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  12336. [0.6165, 0.4106, 0.7575, 0.1733, 0.3838, 0.2650, 0.5680, 0.5116],
  12337. [0.6129, 0.4069, 0.8750, 0.5067, 0.3875, 0.4233, 0.5235, 0.5881]],
  12338. device='cuda:0', dtype=torch.float64)
  12339. predictions are: tensor([[0.4714, 0.3092, 0.8766, 0.3644, 0.3797, 0.3765, 0.5317, 0.5521],
  12340. [0.7109, 0.4638, 0.8667, 0.3510, 0.4728, 0.2814, 0.5916, 0.5561],
  12341. [0.5956, 0.3719, 0.9270, 0.5358, 0.3928, 0.4968, 0.6874, 0.4935],
  12342. [0.2441, 0.1478, 0.7006, 0.2561, 0.4080, 0.2624, 0.5646, 0.5555],
  12343. [0.6287, 0.4122, 0.7137, 0.2069, 0.4231, 0.2120, 0.5734, 0.5442],
  12344. [0.6276, 0.4006, 0.9260, 0.3551, 0.4172, 0.2875, 0.6585, 0.5085],
  12345. [0.6717, 0.4245, 0.7733, 0.1955, 0.4045, 0.2666, 0.6203, 0.5008],
  12346. [0.5436, 0.3620, 0.9240, 0.5174, 0.4223, 0.4559, 0.5638, 0.5646]],
  12347. device='cuda:0', grad_fn=<AddmmBackward>)
  12348. landmarks are: tensor([[[0.6118, 0.4052, 0.8462, 0.3917, 0.3537, 0.3450, 0.5053, 0.5593],
  12349. [0.6226, 0.4103, 0.8575, 0.3450, 0.4387, 0.2067, 0.5788, 0.5383],
  12350. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  12351. [0.0000, 0.0000, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  12352. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  12353. [0.6277, 0.4117, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  12354. [0.6165, 0.4106, 0.7575, 0.1733, 0.3837, 0.2650, 0.5680, 0.5116],
  12355. [0.6129, 0.4069, 0.8750, 0.5067, 0.3875, 0.4233, 0.5235, 0.5881]]],
  12356. device='cuda:0')
  12357. loss_train_step before backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  12358. loss_train_step after backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  12359. loss_train: 0.17162802946404554
  12360. step: 90
  12361. running loss: 0.0019069781051560615
  12362.  
  12363. Valid Steps: 10/10 Loss: nan 19
  12364. --------------------------------------------------
  12365. Epoch: 4 Train Loss: 0.0019 Valid Loss: nan
  12366. --------------------------------------------------
  12367. size of train loader is: 90
  12368. torch.Size([8, 600, 800])
  12369. torch.Size([8, 8])
  12370. tensor([[0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6038, 0.4833],
  12371. [ nan, nan, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  12372. [0.6353, 0.4128, 0.8488, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  12373. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  12374. [0.6267, 0.4065, 0.8313, 0.2467, 0.4788, 0.1733, 0.6312, 0.5133],
  12375. [0.6109, 0.4036, 0.7188, 0.1750, 0.3850, 0.2550, 0.5863, 0.5567],
  12376. [0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533],
  12377. [0.6200, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391]],
  12378. device='cuda:0', dtype=torch.float64)
  12379. predictions are: tensor([[0.5069, 0.3172, 0.9035, 0.4719, 0.3578, 0.5011, 0.6231, 0.4727],
  12380. [0.2427, 0.1428, 0.7823, 0.2831, 0.3590, 0.3039, 0.5319, 0.5369],
  12381. [0.4679, 0.2951, 0.8326, 0.2327, 0.5136, 0.2198, 0.6668, 0.5393],
  12382. [0.6167, 0.3997, 0.7749, 0.2314, 0.4379, 0.2040, 0.6198, 0.5140],
  12383. [0.6602, 0.4364, 0.8328, 0.2305, 0.4575, 0.1908, 0.6440, 0.5086],
  12384. [0.6117, 0.3958, 0.7204, 0.2080, 0.3732, 0.2822, 0.5813, 0.5445],
  12385. [0.6157, 0.3828, 0.7254, 0.2383, 0.3530, 0.3197, 0.6175, 0.5472],
  12386. [0.6747, 0.4466, 0.8350, 0.2806, 0.4019, 0.2544, 0.5892, 0.5197]],
  12387. device='cuda:0', grad_fn=<AddmmBackward>)
  12388. landmarks are: tensor([[[0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6037, 0.4833],
  12389. [0.0000, 0.0000, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  12390. [0.6353, 0.4128, 0.8487, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  12391. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  12392. [0.6266, 0.4065, 0.8313, 0.2467, 0.4787, 0.1733, 0.6313, 0.5133],
  12393. [0.6108, 0.4036, 0.7188, 0.1750, 0.3850, 0.2550, 0.5863, 0.5567],
  12394. [0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533],
  12395. [0.6199, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391]]],
  12396. device='cuda:0')
  12397. loss_train_step before backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  12398. loss_train_step after backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  12399. loss_train: 0.002571904333308339
  12400. step: 1
  12401. running loss: 0.002571904333308339
  12402. Train Steps: 1/90 Loss: 0.0026 torch.Size([8, 600, 800])
  12403. torch.Size([8, 8])
  12404. tensor([[0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5413, 0.5717],
  12405. [0.6310, 0.4017, 0.8563, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006],
  12406. [0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510],
  12407. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  12408. [0.6114, 0.4018, 0.7213, 0.1967, 0.3763, 0.2700, 0.5875, 0.5533],
  12409. [0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  12410. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  12411. [0.6212, 0.4171, 0.7875, 0.3633, 0.3813, 0.2933, 0.5675, 0.5700]],
  12412. device='cuda:0', dtype=torch.float64)
  12413. predictions are: tensor([[0.6312, 0.4185, 0.8692, 0.4429, 0.4182, 0.4553, 0.5748, 0.5425],
  12414. [0.5685, 0.3769, 0.8712, 0.5344, 0.3641, 0.4595, 0.6565, 0.5075],
  12415. [0.4979, 0.3211, 0.7142, 0.2441, 0.3522, 0.2886, 0.5398, 0.5386],
  12416. [0.6197, 0.4256, 0.8434, 0.2249, 0.4742, 0.1780, 0.6291, 0.4914],
  12417. [0.6333, 0.4119, 0.7307, 0.1883, 0.3516, 0.2513, 0.6118, 0.5479],
  12418. [0.5472, 0.3613, 0.8589, 0.4923, 0.3900, 0.4429, 0.6165, 0.5297],
  12419. [0.6158, 0.4280, 0.8171, 0.2819, 0.3469, 0.3697, 0.5905, 0.5223],
  12420. [0.6046, 0.4057, 0.7983, 0.3157, 0.3680, 0.2756, 0.5851, 0.5720]],
  12421. device='cuda:0', grad_fn=<AddmmBackward>)
  12422. landmarks are: tensor([[[0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5412, 0.5717],
  12423. [0.6310, 0.4017, 0.8562, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006],
  12424. [0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510],
  12425. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  12426. [0.6114, 0.4018, 0.7212, 0.1967, 0.3762, 0.2700, 0.5875, 0.5533],
  12427. [0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  12428. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  12429. [0.6212, 0.4171, 0.7875, 0.3633, 0.3812, 0.2933, 0.5675, 0.5700]]],
  12430. device='cuda:0')
  12431. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  12432. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  12433. loss_train: 0.003422100911848247
  12434. step: 2
  12435. running loss: 0.0017110504559241235
  12436. Train Steps: 2/90 Loss: 0.0017 torch.Size([8, 600, 800])
  12437. torch.Size([8, 8])
  12438. tensor([[0.6091, 0.3997, 0.8314, 0.4334, 0.3788, 0.4550, 0.5213, 0.5656],
  12439. [0.6128, 0.4115, 0.7934, 0.3778, 0.3450, 0.4033, 0.5337, 0.5456],
  12440. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467],
  12441. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  12442. [0.6156, 0.4125, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  12443. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  12444. [0.6286, 0.4086, 0.8408, 0.2801, 0.4163, 0.2800, 0.6725, 0.5393],
  12445. [0.6229, 0.4198, 0.7662, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783]],
  12446. device='cuda:0', dtype=torch.float64)
  12447. predictions are: tensor([[0.5767, 0.4007, 0.8143, 0.3931, 0.3484, 0.3931, 0.5638, 0.5402],
  12448. [0.5164, 0.3594, 0.7774, 0.3241, 0.3213, 0.3582, 0.5171, 0.5565],
  12449. [0.5055, 0.3489, 0.8258, 0.2838, 0.3339, 0.3657, 0.5966, 0.5359],
  12450. [0.5999, 0.4185, 0.8611, 0.4580, 0.3542, 0.3977, 0.5300, 0.5404],
  12451. [0.5769, 0.4194, 0.8771, 0.4247, 0.4371, 0.5407, 0.5959, 0.5125],
  12452. [0.5649, 0.3986, 0.8371, 0.3267, 0.3532, 0.2654, 0.6139, 0.5156],
  12453. [0.6567, 0.4417, 0.8134, 0.2386, 0.4005, 0.2341, 0.6993, 0.5478],
  12454. [0.6901, 0.4905, 0.7182, 0.2260, 0.4397, 0.1728, 0.5948, 0.5641]],
  12455. device='cuda:0', grad_fn=<AddmmBackward>)
  12456. landmarks are: tensor([[[0.6091, 0.3997, 0.8314, 0.4334, 0.3787, 0.4550, 0.5213, 0.5656],
  12457. [0.6128, 0.4115, 0.7934, 0.3778, 0.3450, 0.4033, 0.5337, 0.5456],
  12458. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467],
  12459. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  12460. [0.6155, 0.4124, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  12461. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  12462. [0.6286, 0.4086, 0.8408, 0.2801, 0.4162, 0.2800, 0.6725, 0.5393],
  12463. [0.6229, 0.4198, 0.7663, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783]]],
  12464. device='cuda:0')
  12465. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  12466. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  12467. loss_train: 0.004888763302005827
  12468. step: 3
  12469. running loss: 0.0016295877673352759
  12470. Train Steps: 3/90 Loss: 0.0016 torch.Size([8, 600, 800])
  12471. torch.Size([8, 8])
  12472. tensor([[0.6239, 0.4123, 0.8313, 0.2550, 0.4500, 0.2050, 0.6175, 0.5400],
  12473. [ nan, nan, 0.7648, 0.2722, 0.3962, 0.2183, 0.5060, 0.5422],
  12474. [0.6048, 0.3928, 0.8538, 0.5433, 0.3875, 0.5117, 0.5266, 0.4719],
  12475. [0.6079, 0.3964, 0.7420, 0.2958, 0.3563, 0.2917, 0.5351, 0.4980],
  12476. [0.6179, 0.4008, 0.7505, 0.2678, 0.4368, 0.1891, 0.5831, 0.5263],
  12477. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  12478. [0.6143, 0.4034, 0.8800, 0.4833, 0.4512, 0.5367, 0.5289, 0.5097],
  12479. [0.6201, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044]],
  12480. device='cuda:0', dtype=torch.float64)
  12481. predictions are: tensor([[0.6569, 0.4591, 0.8035, 0.2211, 0.4456, 0.1612, 0.6344, 0.5325],
  12482. [0.0221, 0.0467, 0.7281, 0.2313, 0.3881, 0.1999, 0.4928, 0.5521],
  12483. [0.5861, 0.4172, 0.8259, 0.5173, 0.3827, 0.4669, 0.5640, 0.5321],
  12484. [0.6071, 0.4254, 0.7368, 0.2558, 0.3392, 0.2520, 0.5336, 0.5193],
  12485. [0.6452, 0.4404, 0.7274, 0.2249, 0.4207, 0.1520, 0.5773, 0.5537],
  12486. [0.6508, 0.4609, 0.8622, 0.3731, 0.3328, 0.3513, 0.6068, 0.5783],
  12487. [0.5856, 0.4143, 0.8582, 0.4284, 0.4329, 0.4699, 0.5170, 0.5379],
  12488. [0.6478, 0.4482, 0.7278, 0.1706, 0.4110, 0.1783, 0.5851, 0.5236]],
  12489. device='cuda:0', grad_fn=<AddmmBackward>)
  12490. landmarks are: tensor([[[0.6239, 0.4123, 0.8313, 0.2550, 0.4500, 0.2050, 0.6175, 0.5400],
  12491. [0.0000, 0.0000, 0.7648, 0.2722, 0.3963, 0.2183, 0.5060, 0.5422],
  12492. [0.6048, 0.3928, 0.8537, 0.5433, 0.3875, 0.5117, 0.5266, 0.4719],
  12493. [0.6079, 0.3964, 0.7420, 0.2958, 0.3562, 0.2917, 0.5351, 0.4980],
  12494. [0.6179, 0.4008, 0.7505, 0.2678, 0.4368, 0.1891, 0.5831, 0.5263],
  12495. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  12496. [0.6143, 0.4034, 0.8800, 0.4833, 0.4512, 0.5367, 0.5289, 0.5097],
  12497. [0.6202, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044]]],
  12498. device='cuda:0')
  12499. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  12500. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  12501. loss_train: 0.005824644234962761
  12502. step: 4
  12503. running loss: 0.0014561610587406904
  12504.  
  12505. Train Steps: 4/90 Loss: 0.0015 torch.Size([8, 600, 800])
  12506. torch.Size([8, 8])
  12507. tensor([[0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  12508. [0.6286, 0.4055, 0.9000, 0.4717, 0.3763, 0.4683, 0.7018, 0.5494],
  12509. [0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533],
  12510. [0.6058, 0.3986, 0.8324, 0.4626, 0.3838, 0.4983, 0.5147, 0.5466],
  12511. [0.6102, 0.4001, 0.7738, 0.3583, 0.3463, 0.3800, 0.5524, 0.5689],
  12512. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  12513. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  12514. [0.6151, 0.4085, 0.8750, 0.4367, 0.3887, 0.4367, 0.5066, 0.5846]],
  12515. device='cuda:0', dtype=torch.float64)
  12516. predictions are: tensor([[0.6226, 0.4463, 0.8285, 0.5319, 0.3688, 0.4619, 0.6373, 0.5457],
  12517. [0.6219, 0.4211, 0.8754, 0.4207, 0.3663, 0.4380, 0.6535, 0.5222],
  12518. [0.6355, 0.4175, 0.7015, 0.2082, 0.3690, 0.2846, 0.5753, 0.5700],
  12519. [0.5749, 0.4067, 0.8070, 0.4229, 0.3914, 0.4444, 0.5041, 0.5446],
  12520. [0.5692, 0.3908, 0.7660, 0.3264, 0.3533, 0.3449, 0.5118, 0.5689],
  12521. [0.6845, 0.4591, 0.8447, 0.4739, 0.4148, 0.4918, 0.5787, 0.5144],
  12522. [0.6625, 0.4511, 0.9037, 0.4784, 0.3745, 0.3178, 0.5695, 0.4884],
  12523. [0.5560, 0.3884, 0.8544, 0.3927, 0.4004, 0.3863, 0.4718, 0.5870]],
  12524. device='cuda:0', grad_fn=<AddmmBackward>)
  12525. landmarks are: tensor([[[0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  12526. [0.6286, 0.4055, 0.9000, 0.4717, 0.3762, 0.4683, 0.7018, 0.5494],
  12527. [0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533],
  12528. [0.6058, 0.3986, 0.8324, 0.4626, 0.3837, 0.4983, 0.5147, 0.5466],
  12529. [0.6102, 0.4001, 0.7738, 0.3583, 0.3462, 0.3800, 0.5524, 0.5689],
  12530. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  12531. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  12532. [0.6151, 0.4085, 0.8750, 0.4367, 0.3887, 0.4367, 0.5066, 0.5846]]],
  12533. device='cuda:0')
  12534. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  12535. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  12536. loss_train: 0.0068395029520615935
  12537. step: 5
  12538. running loss: 0.0013679005904123187
  12539. Train Steps: 5/90 Loss: 0.0014 torch.Size([8, 600, 800])
  12540. torch.Size([8, 8])
  12541. tensor([[0.6223, 0.4028, 0.8988, 0.4200, 0.3763, 0.5733, 0.6375, 0.5167],
  12542. [0.6179, 0.3993, 0.8925, 0.4789, 0.3879, 0.4900, 0.6041, 0.5279],
  12543. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  12544. [ nan, nan, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  12545. [0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431],
  12546. [0.6263, 0.4065, 0.9038, 0.4317, 0.3588, 0.4550, 0.6325, 0.5250],
  12547. [0.6308, 0.3990, 0.8688, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  12548. [0.6134, 0.4090, 0.6926, 0.2819, 0.3538, 0.3233, 0.5563, 0.5667]],
  12549. device='cuda:0', dtype=torch.float64)
  12550. predictions are: tensor([[0.6524, 0.4446, 0.8739, 0.4292, 0.3901, 0.5684, 0.5870, 0.5248],
  12551. [0.7536, 0.5074, 0.8733, 0.4629, 0.3983, 0.4563, 0.5661, 0.5410],
  12552. [0.6150, 0.4130, 0.8696, 0.4876, 0.4082, 0.5064, 0.6075, 0.4961],
  12553. [0.1934, 0.1508, 0.6788, 0.2390, 0.4383, 0.1802, 0.4801, 0.5844],
  12554. [0.6349, 0.4411, 0.8529, 0.4943, 0.4403, 0.4505, 0.4819, 0.5348],
  12555. [0.6732, 0.4605, 0.8986, 0.4346, 0.3841, 0.4376, 0.6269, 0.5354],
  12556. [0.6724, 0.4510, 0.8377, 0.5237, 0.4102, 0.4649, 0.6140, 0.5327],
  12557. [0.6730, 0.4634, 0.6818, 0.2965, 0.3441, 0.3061, 0.5348, 0.5776]],
  12558. device='cuda:0', grad_fn=<AddmmBackward>)
  12559. landmarks are: tensor([[[0.6223, 0.4028, 0.8988, 0.4200, 0.3762, 0.5733, 0.6375, 0.5167],
  12560. [0.6179, 0.3993, 0.8925, 0.4789, 0.3879, 0.4900, 0.6041, 0.5279],
  12561. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  12562. [0.0000, 0.0000, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  12563. [0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431],
  12564. [0.6263, 0.4065, 0.9038, 0.4317, 0.3587, 0.4550, 0.6325, 0.5250],
  12565. [0.6308, 0.3990, 0.8687, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  12566. [0.6134, 0.4090, 0.6926, 0.2819, 0.3537, 0.3233, 0.5562, 0.5667]]],
  12567. device='cuda:0')
  12568. loss_train_step before backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  12569. loss_train_step after backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  12570. loss_train: 0.008879333385266364
  12571. step: 6
  12572. running loss: 0.001479888897544394
  12573. Train Steps: 6/90 Loss: 0.0015 torch.Size([8, 600, 800])
  12574. torch.Size([8, 8])
  12575. tensor([[0.6257, 0.4024, 0.8672, 0.5422, 0.4196, 0.5198, 0.6694, 0.5204],
  12576. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  12577. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  12578. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  12579. [0.6058, 0.3986, 0.8324, 0.4626, 0.3838, 0.4983, 0.5147, 0.5466],
  12580. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  12581. [0.6268, 0.4102, 0.8938, 0.3667, 0.4025, 0.2833, 0.6275, 0.5183],
  12582. [0.6329, 0.4055, 0.9050, 0.4783, 0.3613, 0.3917, 0.6464, 0.5019]],
  12583. device='cuda:0', dtype=torch.float64)
  12584. predictions are: tensor([[0.6325, 0.4042, 0.8374, 0.5384, 0.4076, 0.5202, 0.6245, 0.5165],
  12585. [0.5927, 0.3926, 0.8575, 0.4811, 0.4553, 0.5659, 0.5778, 0.5579],
  12586. [0.5796, 0.3891, 0.6593, 0.2784, 0.3838, 0.2641, 0.5286, 0.5727],
  12587. [0.5931, 0.3895, 0.8627, 0.5164, 0.4052, 0.5231, 0.5720, 0.5328],
  12588. [0.5840, 0.3913, 0.8053, 0.4573, 0.3862, 0.4881, 0.5126, 0.5421],
  12589. [0.5666, 0.4013, 0.8484, 0.4060, 0.3668, 0.3963, 0.5045, 0.5135],
  12590. [0.5671, 0.3753, 0.8731, 0.3701, 0.4138, 0.2802, 0.5981, 0.5359],
  12591. [0.6675, 0.4155, 0.8711, 0.4727, 0.3791, 0.4025, 0.6250, 0.5075]],
  12592. device='cuda:0', grad_fn=<AddmmBackward>)
  12593. landmarks are: tensor([[[0.6257, 0.4024, 0.8672, 0.5422, 0.4196, 0.5198, 0.6694, 0.5204],
  12594. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  12595. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  12596. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  12597. [0.6058, 0.3986, 0.8324, 0.4626, 0.3837, 0.4983, 0.5147, 0.5466],
  12598. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  12599. [0.6268, 0.4102, 0.8938, 0.3667, 0.4025, 0.2833, 0.6275, 0.5183],
  12600. [0.6329, 0.4055, 0.9050, 0.4783, 0.3613, 0.3917, 0.6464, 0.5019]]],
  12601. device='cuda:0')
  12602. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  12603. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  12604. loss_train: 0.009341607365058735
  12605. step: 7
  12606. running loss: 0.0013345153378655336
  12607. Train Steps: 7/90 Loss: 0.0013 torch.Size([8, 600, 800])
  12608. torch.Size([8, 8])
  12609. tensor([[0.6336, 0.4154, 0.8900, 0.2767, 0.4988, 0.2867, 0.7422, 0.5540],
  12610. [0.6185, 0.4079, 0.8838, 0.4617, 0.4838, 0.5650, 0.6175, 0.5850],
  12611. [0.6248, 0.4032, 0.7738, 0.1900, 0.4813, 0.1400, 0.5941, 0.4904],
  12612. [0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  12613. [0.6245, 0.4100, 0.7762, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  12614. [0.6206, 0.4001, 0.8900, 0.3933, 0.3588, 0.3567, 0.5837, 0.5083],
  12615. [0.6286, 0.4055, 0.9000, 0.4717, 0.3763, 0.4683, 0.7018, 0.5494],
  12616. [0.6231, 0.3973, 0.8650, 0.3950, 0.3625, 0.3183, 0.5837, 0.5167]],
  12617. device='cuda:0', dtype=torch.float64)
  12618. predictions are: tensor([[0.5499, 0.3594, 0.8369, 0.3185, 0.4948, 0.3402, 0.6740, 0.5420],
  12619. [0.5998, 0.3705, 0.8692, 0.4855, 0.4617, 0.6227, 0.5686, 0.5542],
  12620. [0.5499, 0.3415, 0.7362, 0.2629, 0.4629, 0.1953, 0.5573, 0.5133],
  12621. [0.4837, 0.2881, 0.7961, 0.2749, 0.4318, 0.2609, 0.5862, 0.5048],
  12622. [0.6002, 0.3903, 0.7413, 0.2989, 0.4616, 0.2055, 0.5607, 0.5631],
  12623. [0.5945, 0.3634, 0.8931, 0.4460, 0.3458, 0.4164, 0.5333, 0.5032],
  12624. [0.5868, 0.3631, 0.8820, 0.5066, 0.3499, 0.5411, 0.6449, 0.5156],
  12625. [0.5235, 0.3172, 0.8431, 0.4431, 0.3745, 0.3671, 0.5454, 0.5285]],
  12626. device='cuda:0', grad_fn=<AddmmBackward>)
  12627. landmarks are: tensor([[[0.6336, 0.4154, 0.8900, 0.2767, 0.4988, 0.2867, 0.7422, 0.5540],
  12628. [0.6184, 0.4079, 0.8838, 0.4617, 0.4837, 0.5650, 0.6175, 0.5850],
  12629. [0.6248, 0.4032, 0.7738, 0.1900, 0.4812, 0.1400, 0.5941, 0.4904],
  12630. [0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  12631. [0.6245, 0.4100, 0.7763, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  12632. [0.6206, 0.4001, 0.8900, 0.3933, 0.3587, 0.3567, 0.5838, 0.5083],
  12633. [0.6286, 0.4055, 0.9000, 0.4717, 0.3762, 0.4683, 0.7018, 0.5494],
  12634. [0.6231, 0.3973, 0.8650, 0.3950, 0.3625, 0.3183, 0.5838, 0.5167]]],
  12635. device='cuda:0')
  12636. loss_train_step before backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  12637. loss_train_step after backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  12638. loss_train: 0.011782382178353146
  12639. step: 8
  12640. running loss: 0.0014727977722941432
  12641.  
  12642. Train Steps: 8/90 Loss: 0.0015 torch.Size([8, 600, 800])
  12643. torch.Size([8, 8])
  12644. tensor([[0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950],
  12645. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6362, 0.5200],
  12646. [0.6151, 0.4125, 0.8738, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  12647. [0.6200, 0.3993, 0.8519, 0.4923, 0.3962, 0.4717, 0.6013, 0.5433],
  12648. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200],
  12649. [0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050],
  12650. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317],
  12651. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274]],
  12652. device='cuda:0', dtype=torch.float64)
  12653. predictions are: tensor([[0.6115, 0.3676, 0.8520, 0.5439, 0.4247, 0.5460, 0.5940, 0.4784],
  12654. [0.6112, 0.3801, 0.8665, 0.4029, 0.3777, 0.5389, 0.6466, 0.5161],
  12655. [0.5991, 0.3805, 0.8630, 0.4775, 0.3677, 0.4075, 0.5308, 0.5507],
  12656. [0.5744, 0.3545, 0.8599, 0.5122, 0.3941, 0.4950, 0.6332, 0.5469],
  12657. [0.5568, 0.3427, 0.8693, 0.3457, 0.4553, 0.2653, 0.6197, 0.5071],
  12658. [0.6292, 0.4066, 0.8632, 0.4395, 0.3943, 0.5198, 0.5969, 0.5634],
  12659. [0.6212, 0.3967, 0.8706, 0.5548, 0.4042, 0.3961, 0.6012, 0.5386],
  12660. [0.5783, 0.3662, 0.8935, 0.4913, 0.3843, 0.4621, 0.6692, 0.5343]],
  12661. device='cuda:0', grad_fn=<AddmmBackward>)
  12662. landmarks are: tensor([[[0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950],
  12663. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6363, 0.5200],
  12664. [0.6151, 0.4125, 0.8737, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  12665. [0.6200, 0.3993, 0.8519, 0.4923, 0.3963, 0.4717, 0.6012, 0.5433],
  12666. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200],
  12667. [0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050],
  12668. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317],
  12669. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274]]],
  12670. device='cuda:0')
  12671. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  12672. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  12673. loss_train: 0.012471756344893947
  12674. step: 9
  12675. running loss: 0.0013857507049882163
  12676. Train Steps: 9/90 Loss: 0.0014 torch.Size([8, 600, 800])
  12677. torch.Size([8, 8])
  12678. tensor([[0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  12679. [ nan, nan, 0.7335, 0.2569, 0.3788, 0.2667, 0.5066, 0.5578],
  12680. [0.6110, 0.4047, 0.8700, 0.4483, 0.3713, 0.3967, 0.5088, 0.5517],
  12681. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  12682. [0.6371, 0.4092, 0.8337, 0.5850, 0.3950, 0.5117, 0.6559, 0.5262],
  12683. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  12684. [0.6296, 0.4008, 0.9150, 0.4317, 0.4263, 0.3050, 0.7256, 0.5413],
  12685. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383]],
  12686. device='cuda:0', dtype=torch.float64)
  12687. predictions are: tensor([[ 0.6793, 0.4314, 0.9187, 0.5516, 0.3941, 0.4325, 0.6246, 0.5432],
  12688. [ 0.0100, -0.0118, 0.7616, 0.2766, 0.4076, 0.2870, 0.5129, 0.5589],
  12689. [ 0.6446, 0.4051, 0.9034, 0.4762, 0.3717, 0.4353, 0.5629, 0.5327],
  12690. [ 0.7062, 0.4283, 0.8897, 0.4961, 0.4185, 0.5016, 0.5951, 0.5012],
  12691. [ 0.7383, 0.4840, 0.8713, 0.5639, 0.3891, 0.5187, 0.7153, 0.5261],
  12692. [ 0.6300, 0.3864, 0.8894, 0.5457, 0.4119, 0.4570, 0.5903, 0.4735],
  12693. [ 0.6149, 0.3706, 0.9511, 0.4093, 0.4241, 0.3094, 0.7230, 0.5335],
  12694. [ 0.6544, 0.4049, 0.9248, 0.4669, 0.4432, 0.5979, 0.6524, 0.5007]],
  12695. device='cuda:0', grad_fn=<AddmmBackward>)
  12696. landmarks are: tensor([[[0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  12697. [0.0000, 0.0000, 0.7335, 0.2569, 0.3787, 0.2667, 0.5066, 0.5578],
  12698. [0.6110, 0.4047, 0.8700, 0.4483, 0.3713, 0.3967, 0.5088, 0.5517],
  12699. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  12700. [0.6371, 0.4092, 0.8338, 0.5850, 0.3950, 0.5117, 0.6559, 0.5262],
  12701. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  12702. [0.6296, 0.4008, 0.9150, 0.4317, 0.4263, 0.3050, 0.7256, 0.5413],
  12703. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383]]],
  12704. device='cuda:0')
  12705. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  12706. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  12707. loss_train: 0.013627727428684011
  12708. step: 10
  12709. running loss: 0.001362772742868401
  12710. Train Steps: 10/90 Loss: 0.0014 torch.Size([8, 600, 800])
  12711. torch.Size([8, 8])
  12712. tensor([[0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  12713. [0.6332, 0.4165, 0.9100, 0.3350, 0.4188, 0.3683, 0.7438, 0.5528],
  12714. [0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038],
  12715. [0.6275, 0.4111, 0.8463, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  12716. [0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5988, 0.5700],
  12717. [0.6219, 0.4114, 0.8175, 0.2817, 0.3925, 0.2783, 0.5900, 0.5350],
  12718. [0.6026, 0.3979, 0.8550, 0.4233, 0.3613, 0.5233, 0.5582, 0.4967],
  12719. [0.6250, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6088, 0.5183]],
  12720. device='cuda:0', dtype=torch.float64)
  12721. predictions are: tensor([[0.6525, 0.4029, 0.8657, 0.3708, 0.3643, 0.3567, 0.6704, 0.5558],
  12722. [0.4966, 0.3262, 0.9405, 0.4151, 0.4271, 0.3549, 0.7339, 0.5399],
  12723. [0.6338, 0.3980, 0.8844, 0.5107, 0.4377, 0.5056, 0.5811, 0.4925],
  12724. [0.5620, 0.3533, 0.8747, 0.3214, 0.4716, 0.2212, 0.6513, 0.4920],
  12725. [0.4764, 0.3062, 0.8148, 0.3349, 0.3829, 0.3372, 0.6139, 0.5654],
  12726. [0.5407, 0.3467, 0.8335, 0.3137, 0.4079, 0.2607, 0.6321, 0.5385],
  12727. [0.6389, 0.3857, 0.9060, 0.4683, 0.3464, 0.5044, 0.6220, 0.5043],
  12728. [0.6267, 0.3907, 0.9125, 0.5158, 0.4654, 0.5760, 0.6488, 0.5189]],
  12729. device='cuda:0', grad_fn=<AddmmBackward>)
  12730. landmarks are: tensor([[[0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  12731. [0.6332, 0.4165, 0.9100, 0.3350, 0.4187, 0.3683, 0.7438, 0.5528],
  12732. [0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038],
  12733. [0.6275, 0.4111, 0.8462, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  12734. [0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5987, 0.5700],
  12735. [0.6219, 0.4114, 0.8175, 0.2817, 0.3925, 0.2783, 0.5900, 0.5350],
  12736. [0.6026, 0.3979, 0.8550, 0.4233, 0.3613, 0.5233, 0.5582, 0.4967],
  12737. [0.6251, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6087, 0.5183]]],
  12738. device='cuda:0')
  12739. loss_train_step before backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  12740. loss_train_step after backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  12741. loss_train: 0.015669901302317157
  12742. step: 11
  12743. running loss: 0.0014245364820288325
  12744. Train Steps: 11/90 Loss: 0.0014 torch.Size([8, 600, 800])
  12745. torch.Size([8, 8])
  12746. tensor([[0.6212, 0.4033, 0.8938, 0.4167, 0.3813, 0.4267, 0.5613, 0.5583],
  12747. [ nan, nan, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  12748. [0.6102, 0.4020, 0.8638, 0.3717, 0.3625, 0.5017, 0.6038, 0.5500],
  12749. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  12750. [0.6147, 0.4081, 0.8538, 0.3400, 0.3663, 0.3150, 0.5142, 0.4875],
  12751. [0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  12752. [0.6201, 0.4102, 0.7288, 0.2417, 0.4150, 0.2383, 0.6100, 0.5500],
  12753. [0.6196, 0.4094, 0.7562, 0.2817, 0.3937, 0.3183, 0.6013, 0.6183]],
  12754. device='cuda:0', dtype=torch.float64)
  12755. predictions are: tensor([[0.6875, 0.4300, 0.9321, 0.4452, 0.3982, 0.4376, 0.6041, 0.5300],
  12756. [0.1043, 0.0346, 0.9332, 0.3317, 0.5154, 0.2238, 0.6927, 0.5586],
  12757. [0.6547, 0.4155, 0.9068, 0.4049, 0.3867, 0.5096, 0.6642, 0.5283],
  12758. [0.6674, 0.4435, 0.9025, 0.4268, 0.3800, 0.3965, 0.5753, 0.4976],
  12759. [0.6864, 0.4318, 0.9109, 0.3977, 0.3846, 0.3165, 0.5818, 0.5175],
  12760. [0.7053, 0.4394, 0.9215, 0.5413, 0.3811, 0.4696, 0.6207, 0.5601],
  12761. [0.6223, 0.3908, 0.7892, 0.2697, 0.4158, 0.2622, 0.6334, 0.5526],
  12762. [0.6602, 0.4094, 0.8130, 0.3379, 0.4130, 0.3217, 0.6394, 0.6013]],
  12763. device='cuda:0', grad_fn=<AddmmBackward>)
  12764. landmarks are: tensor([[[0.6212, 0.4033, 0.8938, 0.4167, 0.3812, 0.4267, 0.5612, 0.5583],
  12765. [0.0000, 0.0000, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  12766. [0.6102, 0.4020, 0.8637, 0.3717, 0.3625, 0.5017, 0.6037, 0.5500],
  12767. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  12768. [0.6147, 0.4081, 0.8537, 0.3400, 0.3663, 0.3150, 0.5142, 0.4875],
  12769. [0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  12770. [0.6201, 0.4102, 0.7287, 0.2417, 0.4150, 0.2383, 0.6100, 0.5500],
  12771. [0.6196, 0.4094, 0.7563, 0.2817, 0.3938, 0.3183, 0.6012, 0.6183]]],
  12772. device='cuda:0')
  12773. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  12774. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  12775. loss_train: 0.017088345863157883
  12776. step: 12
  12777. running loss: 0.0014240288219298236
  12778.  
  12779. Train Steps: 12/90 Loss: 0.0014 torch.Size([8, 600, 800])
  12780. torch.Size([8, 8])
  12781. tensor([[0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  12782. [0.6069, 0.3975, 0.8625, 0.5083, 0.4388, 0.5483, 0.5650, 0.4967],
  12783. [0.6083, 0.3957, 0.8638, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  12784. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  12785. [0.6189, 0.4049, 0.8888, 0.4417, 0.4213, 0.5200, 0.5988, 0.5633],
  12786. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5963, 0.6217],
  12787. [0.6193, 0.4034, 0.7757, 0.2347, 0.3733, 0.2919, 0.5930, 0.4926],
  12788. [0.6125, 0.3974, 0.7725, 0.2517, 0.3538, 0.3317, 0.5887, 0.5500]],
  12789. device='cuda:0', dtype=torch.float64)
  12790. predictions are: tensor([[0.6531, 0.4223, 0.9179, 0.4122, 0.3533, 0.3904, 0.6084, 0.5556],
  12791. [0.6554, 0.4147, 0.8955, 0.5243, 0.4580, 0.5364, 0.5704, 0.5223],
  12792. [0.6251, 0.3943, 0.9011, 0.4907, 0.4358, 0.4927, 0.5703, 0.4922],
  12793. [0.6102, 0.3983, 0.9248, 0.4518, 0.3971, 0.4422, 0.5567, 0.5213],
  12794. [0.6523, 0.4251, 0.9362, 0.4552, 0.4239, 0.5249, 0.6157, 0.5596],
  12795. [0.6320, 0.3951, 0.7458, 0.2448, 0.4413, 0.2426, 0.6295, 0.6072],
  12796. [0.5919, 0.3868, 0.8213, 0.2366, 0.3866, 0.2564, 0.6227, 0.5115],
  12797. [0.6126, 0.3744, 0.8217, 0.2681, 0.3642, 0.3385, 0.6378, 0.5634]],
  12798. device='cuda:0', grad_fn=<AddmmBackward>)
  12799. landmarks are: tensor([[[0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  12800. [0.6069, 0.3975, 0.8625, 0.5083, 0.4387, 0.5483, 0.5650, 0.4967],
  12801. [0.6083, 0.3957, 0.8637, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  12802. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  12803. [0.6189, 0.4049, 0.8888, 0.4417, 0.4212, 0.5200, 0.5987, 0.5633],
  12804. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5962, 0.6217],
  12805. [0.6193, 0.4034, 0.7757, 0.2347, 0.3733, 0.2919, 0.5930, 0.4926],
  12806. [0.6125, 0.3974, 0.7725, 0.2517, 0.3537, 0.3317, 0.5888, 0.5500]]],
  12807. device='cuda:0')
  12808. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  12809. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  12810. loss_train: 0.01765251744654961
  12811. step: 13
  12812. running loss: 0.001357885957426893
  12813. Train Steps: 13/90 Loss: 0.0014 torch.Size([8, 600, 800])
  12814. torch.Size([8, 8])
  12815. tensor([[0.6200, 0.4059, 0.8700, 0.4900, 0.4163, 0.5000, 0.6162, 0.5467],
  12816. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  12817. [0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  12818. [0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364],
  12819. [0.6293, 0.4024, 0.8750, 0.5000, 0.4012, 0.5733, 0.7121, 0.5633],
  12820. [0.6246, 0.4028, 0.8738, 0.4867, 0.4088, 0.5667, 0.6362, 0.5200],
  12821. [0.6254, 0.3993, 0.8988, 0.4767, 0.3987, 0.5517, 0.6955, 0.5285],
  12822. [0.6200, 0.4118, 0.8287, 0.4017, 0.3775, 0.2833, 0.5391, 0.5799]],
  12823. device='cuda:0', dtype=torch.float64)
  12824. predictions are: tensor([[0.6367, 0.4247, 0.8898, 0.4514, 0.4261, 0.5047, 0.5741, 0.5611],
  12825. [0.5919, 0.4006, 0.8134, 0.2667, 0.3914, 0.2788, 0.5439, 0.5663],
  12826. [0.6725, 0.4472, 0.9134, 0.4913, 0.3964, 0.4697, 0.5581, 0.5180],
  12827. [0.6398, 0.4302, 0.8322, 0.2005, 0.4468, 0.2317, 0.6397, 0.5476],
  12828. [0.6671, 0.4439, 0.9001, 0.4696, 0.4171, 0.5686, 0.6304, 0.5652],
  12829. [0.7045, 0.4583, 0.8897, 0.4612, 0.4242, 0.5617, 0.5993, 0.5274],
  12830. [0.6130, 0.4137, 0.9090, 0.4562, 0.4153, 0.5455, 0.6352, 0.5445],
  12831. [0.4926, 0.3300, 0.8518, 0.3653, 0.3680, 0.2756, 0.5241, 0.5692]],
  12832. device='cuda:0', grad_fn=<AddmmBackward>)
  12833. landmarks are: tensor([[[0.6199, 0.4059, 0.8700, 0.4900, 0.4162, 0.5000, 0.6162, 0.5467],
  12834. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  12835. [0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  12836. [0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364],
  12837. [0.6293, 0.4024, 0.8750, 0.5000, 0.4013, 0.5733, 0.7121, 0.5633],
  12838. [0.6246, 0.4028, 0.8737, 0.4867, 0.4087, 0.5667, 0.6363, 0.5200],
  12839. [0.6254, 0.3993, 0.8988, 0.4767, 0.3988, 0.5517, 0.6955, 0.5285],
  12840. [0.6200, 0.4118, 0.8288, 0.4017, 0.3775, 0.2833, 0.5391, 0.5799]]],
  12841. device='cuda:0')
  12842. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  12843. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  12844. loss_train: 0.018795535230310634
  12845. step: 14
  12846. running loss: 0.0013425382307364739
  12847. Train Steps: 14/90 Loss: 0.0013 torch.Size([8, 600, 800])
  12848. torch.Size([8, 8])
  12849. tensor([[0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  12850. [0.6072, 0.4029, 0.7037, 0.2150, 0.3912, 0.2267, 0.5516, 0.5507],
  12851. [0.6026, 0.3979, 0.8550, 0.4233, 0.3613, 0.5233, 0.5582, 0.4967],
  12852. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  12853. [0.6275, 0.4081, 0.8063, 0.2017, 0.4825, 0.1583, 0.6156, 0.4869],
  12854. [0.6229, 0.4198, 0.7662, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783],
  12855. [ nan, nan, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  12856. [0.6250, 0.4131, 0.8688, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367]],
  12857. device='cuda:0', dtype=torch.float64)
  12858. predictions are: tensor([[0.7229, 0.4724, 0.8251, 0.5350, 0.3827, 0.4518, 0.5353, 0.5470],
  12859. [0.6008, 0.4034, 0.6870, 0.1803, 0.3881, 0.2368, 0.5260, 0.5752],
  12860. [0.6170, 0.4062, 0.8478, 0.3946, 0.3342, 0.5308, 0.5184, 0.5252],
  12861. [0.6327, 0.4191, 0.8496, 0.2966, 0.4430, 0.3363, 0.6651, 0.5467],
  12862. [0.6231, 0.4200, 0.8101, 0.1875, 0.4821, 0.1824, 0.5935, 0.4996],
  12863. [0.5662, 0.3985, 0.7558, 0.2272, 0.4465, 0.2259, 0.5278, 0.5828],
  12864. [0.0313, 0.0332, 0.8566, 0.1995, 0.4916, 0.2640, 0.6464, 0.5660],
  12865. [0.6997, 0.4572, 0.8597, 0.2787, 0.4170, 0.2444, 0.5558, 0.5374]],
  12866. device='cuda:0', grad_fn=<AddmmBackward>)
  12867. landmarks are: tensor([[[0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  12868. [0.6072, 0.4029, 0.7038, 0.2150, 0.3913, 0.2267, 0.5516, 0.5507],
  12869. [0.6026, 0.3979, 0.8550, 0.4233, 0.3613, 0.5233, 0.5582, 0.4967],
  12870. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  12871. [0.6275, 0.4081, 0.8062, 0.2017, 0.4825, 0.1583, 0.6156, 0.4869],
  12872. [0.6229, 0.4198, 0.7663, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783],
  12873. [0.0000, 0.0000, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  12874. [0.6250, 0.4131, 0.8687, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367]]],
  12875. device='cuda:0')
  12876. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  12877. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  12878. loss_train: 0.019952477450715378
  12879. step: 15
  12880. running loss: 0.0013301651633810252
  12881. Train Steps: 15/90 Loss: 0.0013 torch.Size([8, 600, 800])
  12882. torch.Size([8, 8])
  12883. tensor([[ nan, nan, 0.8888, 0.3100, 0.5262, 0.2817, 0.7145, 0.6003],
  12884. [0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  12885. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  12886. [0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367],
  12887. [0.6339, 0.4102, 0.9088, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  12888. [0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038],
  12889. [0.6200, 0.4071, 0.7338, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  12890. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323]],
  12891. device='cuda:0', dtype=torch.float64)
  12892. predictions are: tensor([[0.1825, 0.1387, 0.8564, 0.2657, 0.4663, 0.2815, 0.6554, 0.5692],
  12893. [0.5838, 0.4080, 0.8435, 0.4816, 0.3514, 0.4680, 0.5071, 0.5772],
  12894. [0.6640, 0.4447, 0.8520, 0.3452, 0.4214, 0.2683, 0.5725, 0.5439],
  12895. [0.6804, 0.4549, 0.8350, 0.2710, 0.3790, 0.2731, 0.6021, 0.5237],
  12896. [0.6128, 0.4227, 0.8599, 0.4379, 0.3931, 0.5277, 0.6612, 0.5508],
  12897. [0.5641, 0.3913, 0.8124, 0.4318, 0.4103, 0.5226, 0.4824, 0.5051],
  12898. [0.6186, 0.4344, 0.7072, 0.1699, 0.4006, 0.2641, 0.5666, 0.5555],
  12899. [0.6137, 0.4181, 0.7518, 0.1825, 0.4275, 0.2110, 0.5735, 0.5496]],
  12900. device='cuda:0', grad_fn=<AddmmBackward>)
  12901. landmarks are: tensor([[[0.0000, 0.0000, 0.8888, 0.3100, 0.5263, 0.2817, 0.7145, 0.6003],
  12902. [0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  12903. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  12904. [0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367],
  12905. [0.6339, 0.4102, 0.9087, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  12906. [0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038],
  12907. [0.6200, 0.4071, 0.7337, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  12908. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323]]],
  12909. device='cuda:0')
  12910. loss_train_step before backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  12911. loss_train_step after backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  12912. loss_train: 0.021972051385091618
  12913. step: 16
  12914. running loss: 0.001373253211568226
  12915.  
  12916. Train Steps: 16/90 Loss: 0.0014 torch.Size([8, 600, 800])
  12917. torch.Size([8, 8])
  12918. tensor([[0.6111, 0.4019, 0.8350, 0.3867, 0.3500, 0.4283, 0.5480, 0.5719],
  12919. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  12920. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  12921. [0.6082, 0.4024, 0.8738, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  12922. [0.6339, 0.4102, 0.9088, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  12923. [0.6371, 0.4092, 0.8337, 0.5850, 0.3950, 0.5117, 0.6559, 0.5262],
  12924. [0.6203, 0.4021, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  12925. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082]],
  12926. device='cuda:0', dtype=torch.float64)
  12927. predictions are: tensor([[0.6298, 0.4151, 0.8153, 0.3393, 0.3600, 0.4291, 0.5379, 0.5698],
  12928. [0.6456, 0.4376, 0.8645, 0.4351, 0.3727, 0.4266, 0.6316, 0.5136],
  12929. [0.5620, 0.3916, 0.8048, 0.4880, 0.3995, 0.4805, 0.5509, 0.6130],
  12930. [0.6626, 0.4515, 0.8499, 0.3342, 0.3620, 0.3796, 0.5293, 0.5116],
  12931. [0.5962, 0.3994, 0.8743, 0.4134, 0.4052, 0.5115, 0.7128, 0.5608],
  12932. [0.6262, 0.4437, 0.8169, 0.4994, 0.3806, 0.5024, 0.6557, 0.5502],
  12933. [0.6586, 0.4178, 0.8642, 0.4511, 0.3658, 0.3891, 0.5872, 0.5320],
  12934. [0.6053, 0.3949, 0.8285, 0.4581, 0.4370, 0.5583, 0.6096, 0.5243]],
  12935. device='cuda:0', grad_fn=<AddmmBackward>)
  12936. landmarks are: tensor([[[0.6111, 0.4019, 0.8350, 0.3867, 0.3500, 0.4283, 0.5480, 0.5719],
  12937. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  12938. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  12939. [0.6082, 0.4024, 0.8737, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  12940. [0.6339, 0.4102, 0.9087, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  12941. [0.6371, 0.4092, 0.8338, 0.5850, 0.3950, 0.5117, 0.6559, 0.5262],
  12942. [0.6203, 0.4020, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  12943. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082]]],
  12944. device='cuda:0')
  12945. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  12946. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  12947. loss_train: 0.022940666152862832
  12948. step: 17
  12949. running loss: 0.001349450950168402
  12950. Train Steps: 17/90 Loss: 0.0013 torch.Size([8, 600, 800])
  12951. torch.Size([8, 8])
  12952. tensor([[0.6098, 0.3991, 0.8638, 0.4717, 0.4263, 0.4967, 0.5212, 0.5650],
  12953. [0.6097, 0.4000, 0.7325, 0.2667, 0.3450, 0.3517, 0.5284, 0.5045],
  12954. [0.6182, 0.4058, 0.8738, 0.4350, 0.3563, 0.3400, 0.5290, 0.5822],
  12955. [0.6151, 0.4058, 0.7068, 0.2680, 0.3400, 0.4083, 0.5775, 0.5733],
  12956. [ nan, nan, 0.7268, 0.2333, 0.4125, 0.1933, 0.5112, 0.5383],
  12957. [ nan, nan, 0.7515, 0.2708, 0.3987, 0.2267, 0.5162, 0.5567],
  12958. [0.6193, 0.4079, 0.7288, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  12959. [0.6189, 0.4049, 0.8888, 0.4417, 0.4213, 0.5200, 0.5988, 0.5633]],
  12960. device='cuda:0', dtype=torch.float64)
  12961. predictions are: tensor([[0.6877, 0.4486, 0.8495, 0.5078, 0.4421, 0.4954, 0.5811, 0.5105],
  12962. [0.5926, 0.3816, 0.7462, 0.2623, 0.3489, 0.3333, 0.5847, 0.5139],
  12963. [0.7007, 0.4712, 0.8677, 0.4622, 0.3739, 0.3210, 0.5782, 0.5522],
  12964. [0.6480, 0.4357, 0.7169, 0.2784, 0.3488, 0.3984, 0.6073, 0.5504],
  12965. [0.1842, 0.1286, 0.7313, 0.2295, 0.4224, 0.1709, 0.5615, 0.5122],
  12966. [0.0856, 0.0792, 0.7627, 0.2752, 0.4135, 0.2344, 0.5505, 0.5307],
  12967. [0.7567, 0.5055, 0.7287, 0.2531, 0.4355, 0.2495, 0.6603, 0.5967],
  12968. [0.6695, 0.4580, 0.8895, 0.4715, 0.4324, 0.5336, 0.6547, 0.5361]],
  12969. device='cuda:0', grad_fn=<AddmmBackward>)
  12970. landmarks are: tensor([[[0.6098, 0.3991, 0.8637, 0.4717, 0.4263, 0.4967, 0.5213, 0.5650],
  12971. [0.6097, 0.4000, 0.7325, 0.2667, 0.3450, 0.3517, 0.5284, 0.5045],
  12972. [0.6182, 0.4058, 0.8737, 0.4350, 0.3562, 0.3400, 0.5290, 0.5822],
  12973. [0.6151, 0.4058, 0.7068, 0.2680, 0.3400, 0.4083, 0.5775, 0.5733],
  12974. [0.0000, 0.0000, 0.7268, 0.2333, 0.4125, 0.1933, 0.5113, 0.5383],
  12975. [0.0000, 0.0000, 0.7515, 0.2708, 0.3988, 0.2267, 0.5163, 0.5567],
  12976. [0.6193, 0.4078, 0.7287, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  12977. [0.6189, 0.4049, 0.8888, 0.4417, 0.4212, 0.5200, 0.5987, 0.5633]]],
  12978. device='cuda:0')
  12979. loss_train_step before backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  12980. loss_train_step after backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  12981. loss_train: 0.025358807033626363
  12982. step: 18
  12983. running loss: 0.0014088226129792423
  12984. Train Steps: 18/90 Loss: 0.0014 torch.Size([8, 600, 800])
  12985. torch.Size([8, 8])
  12986. tensor([[0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  12987. [0.6151, 0.4085, 0.8750, 0.4367, 0.3887, 0.4367, 0.5066, 0.5846],
  12988. [0.6179, 0.4008, 0.7505, 0.2678, 0.4368, 0.1891, 0.5831, 0.5263],
  12989. [0.6082, 0.4024, 0.8738, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  12990. [0.6221, 0.4107, 0.7788, 0.3033, 0.3950, 0.2817, 0.6075, 0.5517],
  12991. [0.6212, 0.4159, 0.8675, 0.5783, 0.4088, 0.4317, 0.5613, 0.5917],
  12992. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  12993. [0.6299, 0.4008, 0.8450, 0.5350, 0.4213, 0.5000, 0.6350, 0.5100]],
  12994. device='cuda:0', dtype=torch.float64)
  12995. predictions are: tensor([[0.5624, 0.3717, 0.8425, 0.3025, 0.4063, 0.4040, 0.7155, 0.5545],
  12996. [0.5104, 0.3487, 0.8494, 0.4525, 0.3910, 0.4377, 0.5365, 0.5834],
  12997. [0.5058, 0.3339, 0.7332, 0.2560, 0.4338, 0.1844, 0.6126, 0.5182],
  12998. [0.5532, 0.3823, 0.8483, 0.4023, 0.3636, 0.3770, 0.5509, 0.4996],
  12999. [0.6263, 0.4146, 0.7884, 0.2855, 0.4145, 0.2659, 0.6401, 0.5257],
  13000. [0.5362, 0.3845, 0.8102, 0.5580, 0.4090, 0.4167, 0.5855, 0.5826],
  13001. [0.5352, 0.3609, 0.8476, 0.4723, 0.3517, 0.3063, 0.5778, 0.5627],
  13002. [0.4725, 0.3160, 0.8095, 0.5373, 0.4074, 0.4968, 0.6818, 0.5003]],
  13003. device='cuda:0', grad_fn=<AddmmBackward>)
  13004. landmarks are: tensor([[[0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  13005. [0.6151, 0.4085, 0.8750, 0.4367, 0.3887, 0.4367, 0.5066, 0.5846],
  13006. [0.6179, 0.4008, 0.7505, 0.2678, 0.4368, 0.1891, 0.5831, 0.5263],
  13007. [0.6082, 0.4024, 0.8737, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  13008. [0.6221, 0.4107, 0.7788, 0.3033, 0.3950, 0.2817, 0.6075, 0.5517],
  13009. [0.6212, 0.4159, 0.8675, 0.5783, 0.4087, 0.4317, 0.5612, 0.5917],
  13010. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  13011. [0.6299, 0.4008, 0.8450, 0.5350, 0.4212, 0.5000, 0.6350, 0.5100]]],
  13012. device='cuda:0')
  13013. loss_train_step before backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  13014. loss_train_step after backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  13015. loss_train: 0.027124431653646752
  13016. step: 19
  13017. running loss: 0.001427601665981408
  13018. Train Steps: 19/90 Loss: 0.0014 torch.Size([8, 600, 800])
  13019. torch.Size([8, 8])
  13020. tensor([[0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  13021. [0.6199, 0.4102, 0.8950, 0.4417, 0.4012, 0.5367, 0.6112, 0.5967],
  13022. [0.6165, 0.4106, 0.7575, 0.1733, 0.3838, 0.2650, 0.5680, 0.5116],
  13023. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  13024. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864],
  13025. [0.6308, 0.3990, 0.8688, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  13026. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  13027. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467]],
  13028. device='cuda:0', dtype=torch.float64)
  13029. predictions are: tensor([[0.5778, 0.3690, 0.8085, 0.2754, 0.4514, 0.2019, 0.6591, 0.5103],
  13030. [0.4937, 0.3214, 0.8699, 0.5001, 0.4092, 0.5391, 0.6217, 0.6006],
  13031. [0.5210, 0.3367, 0.7454, 0.2287, 0.3928, 0.2534, 0.5894, 0.5159],
  13032. [0.5524, 0.3582, 0.8768, 0.5135, 0.3693, 0.3642, 0.6268, 0.5321],
  13033. [0.5612, 0.3727, 0.8623, 0.5504, 0.3960, 0.4826, 0.7284, 0.5798],
  13034. [0.5088, 0.3190, 0.8487, 0.5750, 0.4073, 0.5123, 0.6379, 0.5277],
  13035. [0.5967, 0.3906, 0.8153, 0.3364, 0.4398, 0.2178, 0.6045, 0.5534],
  13036. [0.5869, 0.3812, 0.8239, 0.3602, 0.3586, 0.4270, 0.5885, 0.5591]],
  13037. device='cuda:0', grad_fn=<AddmmBackward>)
  13038. landmarks are: tensor([[[0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  13039. [0.6199, 0.4102, 0.8950, 0.4417, 0.4013, 0.5367, 0.6112, 0.5967],
  13040. [0.6165, 0.4106, 0.7575, 0.1733, 0.3837, 0.2650, 0.5680, 0.5116],
  13041. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  13042. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864],
  13043. [0.6308, 0.3990, 0.8687, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  13044. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  13045. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467]]],
  13046. device='cuda:0')
  13047. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  13048. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  13049. loss_train: 0.028719637513859197
  13050. step: 20
  13051. running loss: 0.0014359818756929598
  13052.  
  13053. Train Steps: 20/90 Loss: 0.0014 torch.Size([8, 600, 800])
  13054. torch.Size([8, 8])
  13055. tensor([[0.6151, 0.4125, 0.8738, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  13056. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  13057. [0.6124, 0.4075, 0.7696, 0.4153, 0.3475, 0.3767, 0.5157, 0.5427],
  13058. [0.6125, 0.3974, 0.7725, 0.2517, 0.3538, 0.3317, 0.5887, 0.5500],
  13059. [ nan, nan, 0.7240, 0.2722, 0.3900, 0.2567, 0.5168, 0.5933],
  13060. [0.6148, 0.4076, 0.8666, 0.4820, 0.4138, 0.5067, 0.5250, 0.5767],
  13061. [0.6276, 0.4002, 0.8800, 0.5533, 0.3575, 0.4400, 0.6132, 0.4672],
  13062. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116]],
  13063. device='cuda:0', dtype=torch.float64)
  13064. predictions are: tensor([[ 0.6311, 0.3910, 0.8631, 0.4934, 0.3649, 0.3741, 0.5482, 0.5658],
  13065. [ 0.6389, 0.4029, 0.8508, 0.3148, 0.4316, 0.2565, 0.6621, 0.4937],
  13066. [ 0.6481, 0.4064, 0.8025, 0.4326, 0.3589, 0.3769, 0.5521, 0.5612],
  13067. [ 0.6320, 0.3742, 0.7733, 0.2867, 0.3728, 0.3405, 0.6450, 0.5739],
  13068. [ 0.0239, -0.0033, 0.7480, 0.3124, 0.3933, 0.2245, 0.5602, 0.5838],
  13069. [ 0.6112, 0.3836, 0.8664, 0.5335, 0.4437, 0.5030, 0.5833, 0.5835],
  13070. [ 0.6534, 0.3981, 0.8733, 0.5679, 0.3811, 0.4323, 0.6578, 0.5283],
  13071. [ 0.6353, 0.4044, 0.8751, 0.4401, 0.4185, 0.5854, 0.6248, 0.5464]],
  13072. device='cuda:0', grad_fn=<AddmmBackward>)
  13073. landmarks are: tensor([[[0.6151, 0.4125, 0.8737, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  13074. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  13075. [0.6124, 0.4075, 0.7696, 0.4153, 0.3475, 0.3767, 0.5157, 0.5427],
  13076. [0.6125, 0.3974, 0.7725, 0.2517, 0.3537, 0.3317, 0.5888, 0.5500],
  13077. [0.0000, 0.0000, 0.7240, 0.2722, 0.3900, 0.2567, 0.5168, 0.5933],
  13078. [0.6148, 0.4076, 0.8666, 0.4820, 0.4137, 0.5067, 0.5250, 0.5767],
  13079. [0.6276, 0.4002, 0.8800, 0.5533, 0.3575, 0.4400, 0.6132, 0.4672],
  13080. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116]]],
  13081. device='cuda:0')
  13082. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  13083. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  13084. loss_train: 0.029402486368780956
  13085. step: 21
  13086. running loss: 0.0014001183985133789
  13087. Train Steps: 21/90 Loss: 0.0014 torch.Size([8, 600, 800])
  13088. torch.Size([8, 8])
  13089. tensor([[0.6264, 0.4248, 0.8938, 0.4183, 0.3875, 0.4100, 0.5400, 0.5350],
  13090. [0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038],
  13091. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  13092. [0.6113, 0.4088, 0.6859, 0.2208, 0.4363, 0.1700, 0.5188, 0.5533],
  13093. [0.6163, 0.4006, 0.8788, 0.4683, 0.3663, 0.4883, 0.5887, 0.5017],
  13094. [ nan, nan, 0.8363, 0.3317, 0.3563, 0.3367, 0.5329, 0.5142],
  13095. [0.6204, 0.4110, 0.7913, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367],
  13096. [0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500]],
  13097. device='cuda:0', dtype=torch.float64)
  13098. predictions are: tensor([[0.7024, 0.4579, 0.8919, 0.4660, 0.3621, 0.3956, 0.5566, 0.5544],
  13099. [0.6047, 0.3683, 0.8764, 0.5155, 0.4330, 0.5081, 0.5417, 0.5083],
  13100. [0.5885, 0.3489, 0.9078, 0.5210, 0.4192, 0.5181, 0.6419, 0.5389],
  13101. [0.2829, 0.1731, 0.7203, 0.2559, 0.4322, 0.1594, 0.5465, 0.5538],
  13102. [0.6801, 0.4293, 0.8756, 0.5013, 0.3782, 0.4941, 0.5780, 0.5519],
  13103. [0.3500, 0.2140, 0.8230, 0.3471, 0.3318, 0.3327, 0.5382, 0.5294],
  13104. [0.6547, 0.4065, 0.8151, 0.2838, 0.4189, 0.2514, 0.6232, 0.5550],
  13105. [0.5903, 0.3585, 0.7902, 0.3010, 0.3526, 0.3983, 0.6259, 0.5612]],
  13106. device='cuda:0', grad_fn=<AddmmBackward>)
  13107. landmarks are: tensor([[[0.6264, 0.4248, 0.8938, 0.4183, 0.3875, 0.4100, 0.5400, 0.5350],
  13108. [0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038],
  13109. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  13110. [0.6113, 0.4088, 0.6859, 0.2208, 0.4363, 0.1700, 0.5188, 0.5533],
  13111. [0.6163, 0.4006, 0.8788, 0.4683, 0.3663, 0.4883, 0.5888, 0.5017],
  13112. [0.0000, 0.0000, 0.8363, 0.3317, 0.3562, 0.3367, 0.5329, 0.5142],
  13113. [0.6204, 0.4110, 0.7912, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367],
  13114. [0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500]]],
  13115. device='cuda:0')
  13116. loss_train_step before backward: tensor(0.0058, device='cuda:0', grad_fn=<MseLossBackward>)
  13117. loss_train_step after backward: tensor(0.0058, device='cuda:0', grad_fn=<MseLossBackward>)
  13118. loss_train: 0.035240815341239795
  13119. step: 22
  13120. running loss: 0.001601855242783627
  13121. Train Steps: 22/90 Loss: 0.0016 torch.Size([8, 600, 800])
  13122. torch.Size([8, 8])
  13123. tensor([[0.6266, 0.4067, 0.8588, 0.2867, 0.4300, 0.2850, 0.6325, 0.5267],
  13124. [0.6124, 0.4075, 0.7696, 0.4153, 0.3475, 0.3767, 0.5157, 0.5427],
  13125. [0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050],
  13126. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  13127. [0.6113, 0.4088, 0.6859, 0.2208, 0.4363, 0.1700, 0.5188, 0.5533],
  13128. [0.6299, 0.4008, 0.8450, 0.5350, 0.4213, 0.5000, 0.6350, 0.5100],
  13129. [0.6284, 0.4093, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  13130. [0.6154, 0.4112, 0.7037, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600]],
  13131. device='cuda:0', dtype=torch.float64)
  13132. predictions are: tensor([[0.5260, 0.3243, 0.8976, 0.3190, 0.4337, 0.2847, 0.6278, 0.5195],
  13133. [0.5957, 0.3750, 0.8220, 0.4138, 0.3370, 0.3987, 0.4945, 0.5374],
  13134. [0.5451, 0.3495, 0.9044, 0.4324, 0.3525, 0.5072, 0.5708, 0.5915],
  13135. [0.6046, 0.3757, 0.7889, 0.2466, 0.4441, 0.2022, 0.5529, 0.5444],
  13136. [0.3285, 0.2059, 0.7271, 0.2460, 0.4224, 0.1879, 0.5087, 0.5453],
  13137. [0.5804, 0.3538, 0.8625, 0.5487, 0.4006, 0.5067, 0.6161, 0.5008],
  13138. [0.5873, 0.3687, 0.9221, 0.4838, 0.3646, 0.3840, 0.5822, 0.4999],
  13139. [0.5823, 0.3752, 0.7353, 0.2549, 0.4104, 0.2001, 0.5275, 0.5490]],
  13140. device='cuda:0', grad_fn=<AddmmBackward>)
  13141. landmarks are: tensor([[[0.6266, 0.4067, 0.8587, 0.2867, 0.4300, 0.2850, 0.6325, 0.5267],
  13142. [0.6124, 0.4075, 0.7696, 0.4153, 0.3475, 0.3767, 0.5157, 0.5427],
  13143. [0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050],
  13144. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  13145. [0.6113, 0.4088, 0.6859, 0.2208, 0.4363, 0.1700, 0.5188, 0.5533],
  13146. [0.6299, 0.4008, 0.8450, 0.5350, 0.4212, 0.5000, 0.6350, 0.5100],
  13147. [0.6284, 0.4092, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  13148. [0.6154, 0.4112, 0.7038, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600]]],
  13149. device='cuda:0')
  13150. loss_train_step before backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  13151. loss_train_step after backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  13152. loss_train: 0.03808972876868211
  13153. step: 23
  13154. running loss: 0.0016560751638557438
  13155. Train Steps: 23/90 Loss: 0.0017 torch.Size([8, 600, 800])
  13156. torch.Size([8, 8])
  13157. tensor([[0.6202, 0.4066, 0.8398, 0.2648, 0.3925, 0.2627, 0.5845, 0.5124],
  13158. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  13159. [ nan, nan, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733],
  13160. [0.6296, 0.4008, 0.9150, 0.4317, 0.4263, 0.3050, 0.7256, 0.5413],
  13161. [0.6109, 0.4041, 0.6975, 0.3167, 0.3513, 0.3383, 0.5153, 0.5319],
  13162. [0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  13163. [0.6185, 0.4079, 0.8838, 0.4617, 0.4838, 0.5650, 0.6175, 0.5850],
  13164. [0.6151, 0.4058, 0.7068, 0.2680, 0.3400, 0.4083, 0.5775, 0.5733]],
  13165. device='cuda:0', dtype=torch.float64)
  13166. predictions are: tensor([[0.5628, 0.3788, 0.8062, 0.2812, 0.3902, 0.2384, 0.5447, 0.4986],
  13167. [0.5909, 0.3884, 0.8611, 0.2857, 0.3847, 0.2486, 0.5559, 0.5137],
  13168. [0.0569, 0.0567, 0.7019, 0.2182, 0.4347, 0.1737, 0.5344, 0.5591],
  13169. [0.6344, 0.4047, 0.9095, 0.4098, 0.4022, 0.2848, 0.6425, 0.5178],
  13170. [0.5850, 0.3834, 0.7315, 0.3019, 0.3443, 0.3362, 0.4449, 0.4935],
  13171. [0.6786, 0.4350, 0.8980, 0.3853, 0.4181, 0.3716, 0.6707, 0.5098],
  13172. [0.5914, 0.3821, 0.9013, 0.4644, 0.4535, 0.5465, 0.5600, 0.5473],
  13173. [0.6159, 0.4107, 0.7133, 0.2632, 0.3203, 0.3963, 0.5143, 0.5451]],
  13174. device='cuda:0', grad_fn=<AddmmBackward>)
  13175. landmarks are: tensor([[[0.6202, 0.4066, 0.8398, 0.2648, 0.3925, 0.2627, 0.5845, 0.5124],
  13176. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  13177. [0.0000, 0.0000, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733],
  13178. [0.6296, 0.4008, 0.9150, 0.4317, 0.4263, 0.3050, 0.7256, 0.5413],
  13179. [0.6109, 0.4041, 0.6975, 0.3167, 0.3512, 0.3383, 0.5153, 0.5319],
  13180. [0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  13181. [0.6184, 0.4079, 0.8838, 0.4617, 0.4837, 0.5650, 0.6175, 0.5850],
  13182. [0.6151, 0.4058, 0.7068, 0.2680, 0.3400, 0.4083, 0.5775, 0.5733]]],
  13183. device='cuda:0')
  13184. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  13185. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  13186. loss_train: 0.039125452836742625
  13187. step: 24
  13188. running loss: 0.0016302272015309427
  13189.  
  13190. Train Steps: 24/90 Loss: 0.0016 torch.Size([8, 600, 800])
  13191. torch.Size([8, 8])
  13192. tensor([[0.6185, 0.4129, 0.8900, 0.4567, 0.3937, 0.5417, 0.5734, 0.5110],
  13193. [0.6193, 0.4079, 0.7288, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  13194. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467],
  13195. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  13196. [0.6286, 0.4055, 0.9000, 0.4717, 0.3763, 0.4683, 0.7018, 0.5494],
  13197. [0.6187, 0.4104, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  13198. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  13199. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524]],
  13200. device='cuda:0', dtype=torch.float64)
  13201. predictions are: tensor([[0.5659, 0.3794, 0.9019, 0.4457, 0.3871, 0.5255, 0.5347, 0.5126],
  13202. [0.6188, 0.4026, 0.7115, 0.2296, 0.4164, 0.2545, 0.5583, 0.6091],
  13203. [0.6343, 0.3958, 0.8512, 0.3105, 0.3461, 0.4123, 0.5670, 0.5198],
  13204. [0.6368, 0.4226, 0.8449, 0.2465, 0.4428, 0.2691, 0.6653, 0.5504],
  13205. [0.5647, 0.3627, 0.8970, 0.4503, 0.3598, 0.4549, 0.6685, 0.5267],
  13206. [0.6035, 0.3871, 0.7022, 0.1994, 0.3821, 0.2455, 0.5232, 0.5336],
  13207. [0.5298, 0.3530, 0.8426, 0.3095, 0.3614, 0.3037, 0.5191, 0.5351],
  13208. [0.5681, 0.3602, 0.8807, 0.4949, 0.3830, 0.4460, 0.4697, 0.5351]],
  13209. device='cuda:0', grad_fn=<AddmmBackward>)
  13210. landmarks are: tensor([[[0.6186, 0.4129, 0.8900, 0.4567, 0.3938, 0.5417, 0.5734, 0.5110],
  13211. [0.6193, 0.4078, 0.7287, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  13212. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467],
  13213. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  13214. [0.6286, 0.4055, 0.9000, 0.4717, 0.3762, 0.4683, 0.7018, 0.5494],
  13215. [0.6187, 0.4103, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  13216. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  13217. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524]]],
  13218. device='cuda:0')
  13219. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  13220. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  13221. loss_train: 0.03991389213479124
  13222. step: 25
  13223. running loss: 0.0015965556853916496
  13224. Train Steps: 25/90 Loss: 0.0016 torch.Size([8, 600, 800])
  13225. torch.Size([8, 8])
  13226. tensor([[0.6230, 0.4113, 0.7213, 0.1983, 0.4325, 0.2367, 0.6262, 0.5400],
  13227. [ nan, nan, 0.9088, 0.3783, 0.4562, 0.2617, 0.6741, 0.5575],
  13228. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  13229. [0.6193, 0.4079, 0.7288, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  13230. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  13231. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  13232. [0.6218, 0.4185, 0.7338, 0.2650, 0.4625, 0.1950, 0.5687, 0.5800],
  13233. [0.6289, 0.4032, 0.8419, 0.5446, 0.4075, 0.5017, 0.6312, 0.5117]],
  13234. device='cuda:0', dtype=torch.float64)
  13235. predictions are: tensor([[0.5831, 0.3913, 0.7295, 0.1769, 0.4231, 0.2429, 0.5962, 0.5176],
  13236. [0.3479, 0.2480, 0.9087, 0.3481, 0.4302, 0.2771, 0.6268, 0.5512],
  13237. [0.5968, 0.4228, 0.8091, 0.2909, 0.3452, 0.4028, 0.5369, 0.5207],
  13238. [0.6655, 0.4478, 0.7199, 0.2050, 0.4058, 0.2662, 0.5594, 0.6024],
  13239. [0.6753, 0.4789, 0.8746, 0.4532, 0.3920, 0.5765, 0.6450, 0.5538],
  13240. [0.6214, 0.4176, 0.8670, 0.5141, 0.3903, 0.4475, 0.5306, 0.5421],
  13241. [0.5894, 0.4003, 0.7401, 0.1952, 0.4385, 0.2045, 0.5250, 0.5758],
  13242. [0.6596, 0.4278, 0.8682, 0.4764, 0.3848, 0.4938, 0.6082, 0.5017]],
  13243. device='cuda:0', grad_fn=<AddmmBackward>)
  13244. landmarks are: tensor([[[0.6230, 0.4113, 0.7212, 0.1983, 0.4325, 0.2367, 0.6263, 0.5400],
  13245. [0.0000, 0.0000, 0.9087, 0.3783, 0.4563, 0.2617, 0.6741, 0.5575],
  13246. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  13247. [0.6193, 0.4078, 0.7287, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  13248. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  13249. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  13250. [0.6218, 0.4185, 0.7337, 0.2650, 0.4625, 0.1950, 0.5688, 0.5800],
  13251. [0.6289, 0.4031, 0.8419, 0.5446, 0.4075, 0.5017, 0.6313, 0.5117]]],
  13252. device='cuda:0')
  13253. loss_train_step before backward: tensor(0.0038, device='cuda:0', grad_fn=<MseLossBackward>)
  13254. loss_train_step after backward: tensor(0.0038, device='cuda:0', grad_fn=<MseLossBackward>)
  13255. loss_train: 0.04367428258410655
  13256. step: 26
  13257. running loss: 0.0016797800993887135
  13258. Train Steps: 26/90 Loss: 0.0017 torch.Size([8, 600, 800])
  13259. torch.Size([8, 8])
  13260. tensor([[0.6128, 0.4022, 0.8738, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  13261. [0.6026, 0.3979, 0.8550, 0.4233, 0.3613, 0.5233, 0.5582, 0.4967],
  13262. [0.6177, 0.4086, 0.8738, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700],
  13263. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  13264. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6063, 0.6083],
  13265. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  13266. [0.6083, 0.3957, 0.8638, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  13267. [0.6273, 0.4110, 0.8900, 0.3817, 0.4188, 0.2167, 0.5858, 0.4835]],
  13268. device='cuda:0', dtype=torch.float64)
  13269. predictions are: tensor([[0.6200, 0.4533, 0.8285, 0.4511, 0.4742, 0.4791, 0.5561, 0.5512],
  13270. [0.5848, 0.3958, 0.8567, 0.3728, 0.3393, 0.5062, 0.5970, 0.5217],
  13271. [0.6453, 0.4443, 0.8480, 0.3489, 0.3609, 0.5387, 0.6867, 0.5467],
  13272. [0.5991, 0.4077, 0.8601, 0.4618, 0.3996, 0.4926, 0.6376, 0.5499],
  13273. [0.6198, 0.4574, 0.8238, 0.3098, 0.3500, 0.4181, 0.6127, 0.6146],
  13274. [0.5995, 0.4142, 0.7769, 0.2683, 0.4217, 0.1999, 0.5966, 0.5539],
  13275. [0.6304, 0.4276, 0.8377, 0.4263, 0.4319, 0.5067, 0.5758, 0.5189],
  13276. [0.6700, 0.4491, 0.8450, 0.3254, 0.4215, 0.2076, 0.6212, 0.5116]],
  13277. device='cuda:0', grad_fn=<AddmmBackward>)
  13278. landmarks are: tensor([[[0.6128, 0.4022, 0.8737, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  13279. [0.6026, 0.3979, 0.8550, 0.4233, 0.3613, 0.5233, 0.5582, 0.4967],
  13280. [0.6177, 0.4085, 0.8737, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700],
  13281. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  13282. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6062, 0.6083],
  13283. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  13284. [0.6083, 0.3957, 0.8637, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  13285. [0.6273, 0.4110, 0.8900, 0.3817, 0.4187, 0.2167, 0.5858, 0.4835]]],
  13286. device='cuda:0')
  13287. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  13288. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  13289. loss_train: 0.04473406620672904
  13290. step: 27
  13291. running loss: 0.0016568172669158903
  13292. Train Steps: 27/90 Loss: 0.0017 torch.Size([8, 600, 800])
  13293. torch.Size([8, 8])
  13294. tensor([[0.6299, 0.4303, 0.7963, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  13295. [0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274],
  13296. [0.6193, 0.4050, 0.7313, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  13297. [0.6284, 0.4093, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  13298. [0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  13299. [0.6154, 0.4112, 0.7037, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  13300. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  13301. [0.6346, 0.4165, 0.9138, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471]],
  13302. device='cuda:0', dtype=torch.float64)
  13303. predictions are: tensor([[0.6745, 0.4781, 0.7742, 0.3606, 0.4799, 0.2929, 0.5665, 0.6555],
  13304. [0.5573, 0.3864, 0.8266, 0.2973, 0.4794, 0.2549, 0.6550, 0.5509],
  13305. [0.6332, 0.4213, 0.7079, 0.2312, 0.4146, 0.2592, 0.5788, 0.5784],
  13306. [0.5261, 0.3617, 0.8744, 0.4450, 0.3764, 0.4234, 0.6246, 0.5245],
  13307. [0.6037, 0.4098, 0.8713, 0.3536, 0.3686, 0.4002, 0.6372, 0.5027],
  13308. [0.7105, 0.4971, 0.6866, 0.2331, 0.4318, 0.2278, 0.5594, 0.5842],
  13309. [0.5790, 0.3899, 0.8336, 0.3894, 0.3639, 0.5031, 0.5720, 0.5713],
  13310. [0.5605, 0.3826, 0.8994, 0.3863, 0.4046, 0.4768, 0.7165, 0.5493]],
  13311. device='cuda:0', grad_fn=<AddmmBackward>)
  13312. landmarks are: tensor([[[0.6299, 0.4303, 0.7962, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  13313. [0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274],
  13314. [0.6193, 0.4050, 0.7312, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  13315. [0.6284, 0.4092, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  13316. [0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  13317. [0.6154, 0.4112, 0.7038, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  13318. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  13319. [0.6346, 0.4165, 0.9137, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471]]],
  13320. device='cuda:0')
  13321. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  13322. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  13323. loss_train: 0.045994794316357
  13324. step: 28
  13325. running loss: 0.0016426712255841786
  13326.  
  13327. Train Steps: 28/90 Loss: 0.0016 torch.Size([8, 600, 800])
  13328. torch.Size([8, 8])
  13329. tensor([[0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  13330. [0.6129, 0.4063, 0.8738, 0.5250, 0.4313, 0.4733, 0.5230, 0.5874],
  13331. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  13332. [0.6160, 0.4093, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  13333. [0.6286, 0.3977, 0.9038, 0.4733, 0.3900, 0.4150, 0.7074, 0.5320],
  13334. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5513, 0.5750],
  13335. [0.6293, 0.4024, 0.8750, 0.5000, 0.4012, 0.5733, 0.7121, 0.5633],
  13336. [ nan, nan, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633]],
  13337. device='cuda:0', dtype=torch.float64)
  13338. predictions are: tensor([[0.6088, 0.4196, 0.8702, 0.4265, 0.4201, 0.4764, 0.5855, 0.5290],
  13339. [0.7093, 0.4789, 0.8781, 0.5137, 0.4596, 0.4663, 0.5868, 0.5960],
  13340. [0.7140, 0.4928, 0.8827, 0.4662, 0.4619, 0.4862, 0.6021, 0.5137],
  13341. [0.7183, 0.4751, 0.8284, 0.4473, 0.3919, 0.4522, 0.5970, 0.5669],
  13342. [0.6877, 0.4502, 0.8986, 0.4377, 0.3943, 0.4305, 0.7234, 0.5148],
  13343. [0.7816, 0.5401, 0.6711, 0.2598, 0.4490, 0.1916, 0.5840, 0.5908],
  13344. [0.7330, 0.4897, 0.8921, 0.4890, 0.4254, 0.5871, 0.7601, 0.5764],
  13345. [0.1528, 0.1124, 0.7438, 0.2600, 0.3938, 0.2913, 0.5318, 0.5738]],
  13346. device='cuda:0', grad_fn=<AddmmBackward>)
  13347. landmarks are: tensor([[[0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  13348. [0.6130, 0.4063, 0.8737, 0.5250, 0.4313, 0.4733, 0.5230, 0.5874],
  13349. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  13350. [0.6160, 0.4092, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  13351. [0.6286, 0.3977, 0.9038, 0.4733, 0.3900, 0.4150, 0.7074, 0.5320],
  13352. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5512, 0.5750],
  13353. [0.6293, 0.4024, 0.8750, 0.5000, 0.4013, 0.5733, 0.7121, 0.5633],
  13354. [0.0000, 0.0000, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633]]],
  13355. device='cuda:0')
  13356. loss_train_step before backward: tensor(0.0030, device='cuda:0', grad_fn=<MseLossBackward>)
  13357. loss_train_step after backward: tensor(0.0030, device='cuda:0', grad_fn=<MseLossBackward>)
  13358. loss_train: 0.048961690947180614
  13359. step: 29
  13360. running loss: 0.001688334170592435
  13361. Train Steps: 29/90 Loss: 0.0017 torch.Size([8, 600, 800])
  13362. torch.Size([8, 8])
  13363. tensor([[0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  13364. [0.6122, 0.3993, 0.8738, 0.4667, 0.4517, 0.4879, 0.5155, 0.4927],
  13365. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092],
  13366. [0.6274, 0.4099, 0.8625, 0.3233, 0.4400, 0.1983, 0.5876, 0.4869],
  13367. [0.6263, 0.4065, 0.9038, 0.4317, 0.3588, 0.4550, 0.6325, 0.5250],
  13368. [0.6261, 0.3987, 0.9045, 0.4208, 0.3600, 0.4633, 0.6570, 0.5162],
  13369. [ nan, nan, 0.7512, 0.2117, 0.4288, 0.2000, 0.5600, 0.5617],
  13370. [0.6109, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117]],
  13371. device='cuda:0', dtype=torch.float64)
  13372. predictions are: tensor([[0.6794, 0.4479, 0.8627, 0.3964, 0.3654, 0.3532, 0.6276, 0.5969],
  13373. [0.6532, 0.4360, 0.8675, 0.4889, 0.4654, 0.5083, 0.5451, 0.5391],
  13374. [0.6788, 0.4359, 0.8759, 0.5463, 0.4182, 0.5275, 0.6290, 0.5163],
  13375. [0.6798, 0.4425, 0.8446, 0.3496, 0.4765, 0.2144, 0.6339, 0.5350],
  13376. [0.6577, 0.4384, 0.8928, 0.4528, 0.4095, 0.4527, 0.6980, 0.5613],
  13377. [0.6829, 0.4486, 0.8871, 0.4562, 0.3815, 0.4578, 0.7193, 0.5286],
  13378. [0.3316, 0.2288, 0.7307, 0.2535, 0.4495, 0.2235, 0.5810, 0.5954],
  13379. [0.6351, 0.4448, 0.8648, 0.4567, 0.4051, 0.4139, 0.5898, 0.5463]],
  13380. device='cuda:0', grad_fn=<AddmmBackward>)
  13381. landmarks are: tensor([[[0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  13382. [0.6122, 0.3993, 0.8737, 0.4667, 0.4517, 0.4879, 0.5155, 0.4927],
  13383. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092],
  13384. [0.6274, 0.4099, 0.8625, 0.3233, 0.4400, 0.1983, 0.5876, 0.4869],
  13385. [0.6263, 0.4065, 0.9038, 0.4317, 0.3587, 0.4550, 0.6325, 0.5250],
  13386. [0.6261, 0.3987, 0.9045, 0.4208, 0.3600, 0.4633, 0.6570, 0.5162],
  13387. [0.0000, 0.0000, 0.7513, 0.2117, 0.4288, 0.2000, 0.5600, 0.5617],
  13388. [0.6108, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117]]],
  13389. device='cuda:0')
  13390. loss_train_step before backward: tensor(0.0036, device='cuda:0', grad_fn=<MseLossBackward>)
  13391. loss_train_step after backward: tensor(0.0036, device='cuda:0', grad_fn=<MseLossBackward>)
  13392. loss_train: 0.05260280644870363
  13393. step: 30
  13394. running loss: 0.0017534268816234543
  13395. Train Steps: 30/90 Loss: 0.0018 torch.Size([8, 600, 800])
  13396. torch.Size([8, 8])
  13397. tensor([[0.6200, 0.4098, 0.8237, 0.2917, 0.4012, 0.2967, 0.6000, 0.5683],
  13398. [0.6203, 0.4021, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  13399. [0.6219, 0.4097, 0.8738, 0.3400, 0.3563, 0.4117, 0.5975, 0.5683],
  13400. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  13401. [0.6275, 0.4157, 0.8337, 0.5800, 0.3763, 0.4200, 0.5547, 0.6125],
  13402. [0.6212, 0.4171, 0.7875, 0.3633, 0.3813, 0.2933, 0.5675, 0.5700],
  13403. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  13404. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082]],
  13405. device='cuda:0', dtype=torch.float64)
  13406. predictions are: tensor([[0.5934, 0.3907, 0.8409, 0.3052, 0.4328, 0.2824, 0.6257, 0.5610],
  13407. [0.5742, 0.3419, 0.8959, 0.5044, 0.3988, 0.3896, 0.5758, 0.5160],
  13408. [0.5488, 0.3516, 0.8657, 0.3616, 0.3794, 0.3953, 0.6189, 0.5567],
  13409. [0.6236, 0.3822, 0.8468, 0.5635, 0.4088, 0.4823, 0.6499, 0.5146],
  13410. [0.7318, 0.4671, 0.8511, 0.5613, 0.4100, 0.4089, 0.5688, 0.5884],
  13411. [0.6320, 0.3917, 0.8081, 0.3804, 0.4160, 0.2909, 0.5750, 0.5517],
  13412. [0.6117, 0.3748, 0.8802, 0.4653, 0.4100, 0.4083, 0.6179, 0.5411],
  13413. [0.6195, 0.3790, 0.8797, 0.5160, 0.4637, 0.5493, 0.6130, 0.5171]],
  13414. device='cuda:0', grad_fn=<AddmmBackward>)
  13415. landmarks are: tensor([[[0.6200, 0.4098, 0.8238, 0.2917, 0.4013, 0.2967, 0.6000, 0.5683],
  13416. [0.6203, 0.4020, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  13417. [0.6219, 0.4097, 0.8737, 0.3400, 0.3562, 0.4117, 0.5975, 0.5683],
  13418. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  13419. [0.6275, 0.4157, 0.8338, 0.5800, 0.3762, 0.4200, 0.5547, 0.6125],
  13420. [0.6212, 0.4171, 0.7875, 0.3633, 0.3812, 0.2933, 0.5675, 0.5700],
  13421. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  13422. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082]]],
  13423. device='cuda:0')
  13424. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  13425. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  13426. loss_train: 0.05339707116945647
  13427. step: 31
  13428. running loss: 0.0017224861667566602
  13429. Train Steps: 31/90 Loss: 0.0017 torch.Size([8, 600, 800])
  13430. torch.Size([8, 8])
  13431. tensor([[0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  13432. [0.6069, 0.3975, 0.8625, 0.5083, 0.4388, 0.5483, 0.5650, 0.4967],
  13433. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  13434. [0.6229, 0.4198, 0.7662, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783],
  13435. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5363, 0.5550],
  13436. [0.6165, 0.4106, 0.7575, 0.1733, 0.3838, 0.2650, 0.5680, 0.5116],
  13437. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461],
  13438. [0.6201, 0.4151, 0.8588, 0.5467, 0.3700, 0.3950, 0.5637, 0.5933]],
  13439. device='cuda:0', dtype=torch.float64)
  13440. predictions are: tensor([[0.4726, 0.2672, 0.9171, 0.4492, 0.4305, 0.3561, 0.7157, 0.5673],
  13441. [0.5934, 0.3507, 0.8860, 0.5450, 0.4450, 0.5328, 0.5489, 0.5182],
  13442. [0.6378, 0.3892, 0.8093, 0.2568, 0.4691, 0.1763, 0.6213, 0.4853],
  13443. [0.5002, 0.3126, 0.7706, 0.3071, 0.4582, 0.2126, 0.5596, 0.5627],
  13444. [0.6113, 0.3908, 0.7203, 0.3242, 0.4085, 0.2067, 0.5273, 0.5523],
  13445. [0.5958, 0.3624, 0.7685, 0.2393, 0.3788, 0.2428, 0.5749, 0.4980],
  13446. [0.5660, 0.3472, 0.8770, 0.5506, 0.4446, 0.4987, 0.5519, 0.5381],
  13447. [0.5380, 0.3342, 0.8785, 0.5946, 0.3659, 0.3793, 0.5535, 0.5536]],
  13448. device='cuda:0', grad_fn=<AddmmBackward>)
  13449. landmarks are: tensor([[[0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  13450. [0.6069, 0.3975, 0.8625, 0.5083, 0.4387, 0.5483, 0.5650, 0.4967],
  13451. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  13452. [0.6229, 0.4198, 0.7663, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783],
  13453. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5362, 0.5550],
  13454. [0.6165, 0.4106, 0.7575, 0.1733, 0.3837, 0.2650, 0.5680, 0.5116],
  13455. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461],
  13456. [0.6202, 0.4151, 0.8587, 0.5467, 0.3700, 0.3950, 0.5638, 0.5933]]],
  13457. device='cuda:0')
  13458. loss_train_step before backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  13459. loss_train_step after backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  13460. loss_train: 0.05541980228736065
  13461. step: 32
  13462. running loss: 0.0017318688214800204
  13463.  
  13464. Train Steps: 32/90 Loss: 0.0017 torch.Size([8, 600, 800])
  13465. torch.Size([8, 8])
  13466. tensor([[0.6087, 0.3951, 0.8387, 0.5833, 0.4188, 0.4933, 0.5146, 0.4830],
  13467. [0.6132, 0.4037, 0.6963, 0.2217, 0.4100, 0.1950, 0.5395, 0.5175],
  13468. [0.6229, 0.4066, 0.7612, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  13469. [0.6270, 0.4267, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  13470. [0.6198, 0.4115, 0.7762, 0.2717, 0.3713, 0.3200, 0.5837, 0.5683],
  13471. [0.6308, 0.3990, 0.8688, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  13472. [0.6250, 0.3993, 0.9138, 0.4333, 0.3763, 0.5217, 0.6995, 0.5320],
  13473. [ nan, nan, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621]],
  13474. device='cuda:0', dtype=torch.float64)
  13475. predictions are: tensor([[0.6566, 0.3715, 0.8816, 0.6038, 0.4144, 0.4639, 0.4896, 0.5030],
  13476. [0.5464, 0.3426, 0.7286, 0.2806, 0.3849, 0.1901, 0.4855, 0.4883],
  13477. [0.6351, 0.3949, 0.7865, 0.3325, 0.4041, 0.2448, 0.5552, 0.5239],
  13478. [0.6059, 0.3743, 0.7327, 0.3129, 0.4484, 0.1898, 0.5123, 0.5763],
  13479. [0.5316, 0.3136, 0.8101, 0.3168, 0.3599, 0.3169, 0.5467, 0.5450],
  13480. [0.7205, 0.4120, 0.9059, 0.5588, 0.3834, 0.5009, 0.5938, 0.4897],
  13481. [0.6610, 0.3933, 0.9364, 0.4873, 0.3515, 0.5126, 0.6947, 0.5138],
  13482. [0.0570, 0.0145, 0.8727, 0.3291, 0.5222, 0.1890, 0.6622, 0.5340]],
  13483. device='cuda:0', grad_fn=<AddmmBackward>)
  13484. landmarks are: tensor([[[0.6087, 0.3951, 0.8388, 0.5833, 0.4187, 0.4933, 0.5146, 0.4830],
  13485. [0.6132, 0.4037, 0.6963, 0.2217, 0.4100, 0.1950, 0.5395, 0.5175],
  13486. [0.6229, 0.4066, 0.7613, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  13487. [0.6270, 0.4266, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  13488. [0.6198, 0.4115, 0.7763, 0.2717, 0.3713, 0.3200, 0.5838, 0.5683],
  13489. [0.6308, 0.3990, 0.8687, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  13490. [0.6250, 0.3993, 0.9137, 0.4333, 0.3762, 0.5217, 0.6995, 0.5320],
  13491. [0.0000, 0.0000, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621]]],
  13492. device='cuda:0')
  13493. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  13494. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  13495. loss_train: 0.05678630419424735
  13496. step: 33
  13497. running loss: 0.0017207970967953743
  13498. Train Steps: 33/90 Loss: 0.0017 torch.Size([8, 600, 800])
  13499. torch.Size([8, 8])
  13500. tensor([[0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667],
  13501. [0.6286, 0.4078, 0.8063, 0.2267, 0.4788, 0.1533, 0.5953, 0.4913],
  13502. [0.6222, 0.3937, 0.8350, 0.5617, 0.4138, 0.4600, 0.5800, 0.5233],
  13503. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  13504. [0.6124, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  13505. [0.6127, 0.4084, 0.8700, 0.4467, 0.3987, 0.4317, 0.5013, 0.5471],
  13506. [0.6053, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  13507. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5413, 0.5717]],
  13508. device='cuda:0', dtype=torch.float64)
  13509. predictions are: tensor([[0.5069, 0.3218, 0.8555, 0.3668, 0.3512, 0.3650, 0.5693, 0.5543],
  13510. [0.4332, 0.2523, 0.8236, 0.2801, 0.4832, 0.1434, 0.5698, 0.5128],
  13511. [0.6045, 0.3508, 0.8627, 0.6034, 0.4019, 0.4417, 0.5702, 0.5253],
  13512. [0.5790, 0.3402, 0.8591, 0.3597, 0.3533, 0.3820, 0.6203, 0.5123],
  13513. [0.5639, 0.3524, 0.7444, 0.3332, 0.3590, 0.2914, 0.5039, 0.5421],
  13514. [0.6302, 0.3901, 0.8920, 0.4865, 0.3971, 0.4280, 0.4932, 0.5448],
  13515. [0.4977, 0.3001, 0.7139, 0.2390, 0.3956, 0.1843, 0.5345, 0.5176],
  13516. [0.6276, 0.3901, 0.8870, 0.5276, 0.4145, 0.4734, 0.5482, 0.5294]],
  13517. device='cuda:0', grad_fn=<AddmmBackward>)
  13518. landmarks are: tensor([[[0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667],
  13519. [0.6286, 0.4078, 0.8062, 0.2267, 0.4787, 0.1533, 0.5953, 0.4913],
  13520. [0.6222, 0.3937, 0.8350, 0.5617, 0.4137, 0.4600, 0.5800, 0.5233],
  13521. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  13522. [0.6123, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  13523. [0.6127, 0.4084, 0.8700, 0.4467, 0.3988, 0.4317, 0.5013, 0.5471],
  13524. [0.6054, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  13525. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5412, 0.5717]]],
  13526. device='cuda:0')
  13527. loss_train_step before backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  13528. loss_train_step after backward: tensor(0.0023, device='cuda:0', grad_fn=<MseLossBackward>)
  13529. loss_train: 0.05907099656178616
  13530. step: 34
  13531. running loss: 0.00173738225181724
  13532. Train Steps: 34/90 Loss: 0.0017 torch.Size([8, 600, 800])
  13533. torch.Size([8, 8])
  13534. tensor([[0.6213, 0.4131, 0.8438, 0.3550, 0.3513, 0.4400, 0.5716, 0.5123],
  13535. [0.6144, 0.4032, 0.8563, 0.3283, 0.3525, 0.4200, 0.5775, 0.5583],
  13536. [0.6224, 0.4179, 0.8700, 0.5683, 0.4037, 0.4683, 0.5650, 0.5600],
  13537. [0.6163, 0.4114, 0.7650, 0.2017, 0.3763, 0.2867, 0.5631, 0.5071],
  13538. [0.6271, 0.4040, 0.9138, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  13539. [ nan, nan, 0.7625, 0.2433, 0.3713, 0.2867, 0.5235, 0.5220],
  13540. [0.6275, 0.4048, 0.8488, 0.2883, 0.4463, 0.2033, 0.6321, 0.5155],
  13541. [0.6176, 0.3911, 0.8738, 0.4217, 0.3488, 0.4033, 0.6025, 0.4817]],
  13542. device='cuda:0', dtype=torch.float64)
  13543. predictions are: tensor([[0.5963, 0.3928, 0.8422, 0.3795, 0.3410, 0.4309, 0.5365, 0.5404],
  13544. [0.6860, 0.4379, 0.8387, 0.3597, 0.3536, 0.4246, 0.5370, 0.5621],
  13545. [0.6555, 0.4334, 0.8594, 0.6063, 0.3841, 0.4639, 0.5436, 0.5663],
  13546. [0.6021, 0.3857, 0.7330, 0.2327, 0.3687, 0.2826, 0.5496, 0.5033],
  13547. [0.5491, 0.3401, 0.9177, 0.3947, 0.4646, 0.2704, 0.7094, 0.5530],
  13548. [0.0126, 0.0070, 0.7575, 0.2523, 0.3890, 0.2771, 0.4934, 0.5518],
  13549. [0.7066, 0.4475, 0.8518, 0.2999, 0.4414, 0.2243, 0.6433, 0.5179],
  13550. [0.5654, 0.3630, 0.8516, 0.4159, 0.3413, 0.4084, 0.5547, 0.5023]],
  13551. device='cuda:0', grad_fn=<AddmmBackward>)
  13552. landmarks are: tensor([[[0.6213, 0.4131, 0.8438, 0.3550, 0.3512, 0.4400, 0.5716, 0.5123],
  13553. [0.6144, 0.4032, 0.8562, 0.3283, 0.3525, 0.4200, 0.5775, 0.5583],
  13554. [0.6224, 0.4179, 0.8700, 0.5683, 0.4038, 0.4683, 0.5650, 0.5600],
  13555. [0.6163, 0.4114, 0.7650, 0.2017, 0.3762, 0.2867, 0.5631, 0.5071],
  13556. [0.6271, 0.4040, 0.9137, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  13557. [0.0000, 0.0000, 0.7625, 0.2433, 0.3713, 0.2867, 0.5235, 0.5220],
  13558. [0.6275, 0.4048, 0.8487, 0.2883, 0.4462, 0.2033, 0.6321, 0.5155],
  13559. [0.6176, 0.3911, 0.8737, 0.4217, 0.3487, 0.4033, 0.6025, 0.4817]]],
  13560. device='cuda:0')
  13561. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  13562. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  13563. loss_train: 0.05983991172979586
  13564. step: 35
  13565. running loss: 0.001709711763708453
  13566. Train Steps: 35/90 Loss: 0.0017 torch.Size([8, 600, 800])
  13567. torch.Size([8, 8])
  13568. tensor([[0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  13569. [0.6266, 0.4067, 0.8588, 0.2867, 0.4300, 0.2850, 0.6325, 0.5267],
  13570. [0.6198, 0.3997, 0.8582, 0.5361, 0.4117, 0.5016, 0.5942, 0.5134],
  13571. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  13572. [0.6234, 0.4179, 0.7825, 0.3450, 0.3813, 0.2867, 0.5675, 0.5617],
  13573. [0.6148, 0.4076, 0.8666, 0.4820, 0.4138, 0.5067, 0.5250, 0.5767],
  13574. [0.6097, 0.4024, 0.8488, 0.3717, 0.3875, 0.5517, 0.5836, 0.5591],
  13575. [0.6128, 0.4022, 0.8738, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064]],
  13576. device='cuda:0', dtype=torch.float64)
  13577. predictions are: tensor([[0.5901, 0.4088, 0.8563, 0.3605, 0.3315, 0.3447, 0.5860, 0.4939],
  13578. [0.5359, 0.3621, 0.8485, 0.2784, 0.4185, 0.2434, 0.6223, 0.5242],
  13579. [0.5851, 0.3998, 0.8402, 0.5130, 0.3952, 0.5066, 0.5769, 0.5002],
  13580. [0.5745, 0.3787, 0.8197, 0.3058, 0.3349, 0.3723, 0.6011, 0.5366],
  13581. [0.5456, 0.3811, 0.7809, 0.3230, 0.3732, 0.2617, 0.5631, 0.5644],
  13582. [0.6271, 0.4279, 0.8482, 0.4558, 0.3857, 0.4892, 0.5296, 0.5692],
  13583. [0.5132, 0.3385, 0.8294, 0.3506, 0.3816, 0.5313, 0.5937, 0.5284],
  13584. [0.5584, 0.3972, 0.8556, 0.4872, 0.4582, 0.4736, 0.5132, 0.5291]],
  13585. device='cuda:0', grad_fn=<AddmmBackward>)
  13586. landmarks are: tensor([[[0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  13587. [0.6266, 0.4067, 0.8587, 0.2867, 0.4300, 0.2850, 0.6325, 0.5267],
  13588. [0.6198, 0.3997, 0.8582, 0.5361, 0.4117, 0.5016, 0.5942, 0.5134],
  13589. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  13590. [0.6234, 0.4179, 0.7825, 0.3450, 0.3812, 0.2867, 0.5675, 0.5617],
  13591. [0.6148, 0.4076, 0.8666, 0.4820, 0.4137, 0.5067, 0.5250, 0.5767],
  13592. [0.6097, 0.4024, 0.8487, 0.3717, 0.3875, 0.5517, 0.5836, 0.5591],
  13593. [0.6128, 0.4022, 0.8737, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064]]],
  13594. device='cuda:0')
  13595. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  13596. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  13597. loss_train: 0.060806872352259234
  13598. step: 36
  13599. running loss: 0.0016890797875627566
  13600.  
  13601. Train Steps: 36/90 Loss: 0.0017 torch.Size([8, 600, 800])
  13602. torch.Size([8, 8])
  13603. tensor([[0.6081, 0.3950, 0.8538, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954],
  13604. [0.6182, 0.3972, 0.8552, 0.5914, 0.3683, 0.4181, 0.5688, 0.5378],
  13605. [0.6207, 0.4081, 0.7662, 0.2067, 0.3962, 0.3200, 0.6312, 0.5300],
  13606. [0.6219, 0.3934, 0.8688, 0.5267, 0.4313, 0.4967, 0.5988, 0.4983],
  13607. [0.6097, 0.4000, 0.7325, 0.2667, 0.3450, 0.3517, 0.5284, 0.5045],
  13608. [0.6159, 0.4085, 0.6900, 0.2283, 0.4088, 0.1950, 0.5123, 0.5397],
  13609. [0.6127, 0.4066, 0.8550, 0.5567, 0.4662, 0.5141, 0.5070, 0.5412],
  13610. [ nan, nan, 0.7525, 0.2291, 0.3838, 0.3017, 0.6050, 0.5667]],
  13611. device='cuda:0', dtype=torch.float64)
  13612. predictions are: tensor([[0.6335, 0.4346, 0.8782, 0.4073, 0.3632, 0.4977, 0.5222, 0.5061],
  13613. [0.6177, 0.4202, 0.8657, 0.5371, 0.3564, 0.4533, 0.5620, 0.5361],
  13614. [0.6114, 0.4360, 0.7637, 0.1912, 0.3863, 0.3127, 0.6328, 0.5387],
  13615. [0.6132, 0.4126, 0.8703, 0.4958, 0.4261, 0.4992, 0.5843, 0.4942],
  13616. [0.5834, 0.4001, 0.7386, 0.2229, 0.3346, 0.3644, 0.5495, 0.5346],
  13617. [0.6371, 0.4433, 0.6882, 0.1988, 0.3955, 0.1833, 0.5116, 0.5594],
  13618. [0.6863, 0.4815, 0.8768, 0.5128, 0.4381, 0.4661, 0.5487, 0.5464],
  13619. [0.2251, 0.1772, 0.7638, 0.1963, 0.3851, 0.3008, 0.6064, 0.5866]],
  13620. device='cuda:0', grad_fn=<AddmmBackward>)
  13621. landmarks are: tensor([[[0.6081, 0.3950, 0.8537, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954],
  13622. [0.6182, 0.3972, 0.8552, 0.5914, 0.3683, 0.4181, 0.5688, 0.5378],
  13623. [0.6207, 0.4081, 0.7663, 0.2067, 0.3963, 0.3200, 0.6313, 0.5300],
  13624. [0.6219, 0.3934, 0.8687, 0.5267, 0.4313, 0.4967, 0.5987, 0.4983],
  13625. [0.6097, 0.4000, 0.7325, 0.2667, 0.3450, 0.3517, 0.5284, 0.5045],
  13626. [0.6159, 0.4085, 0.6900, 0.2283, 0.4087, 0.1950, 0.5123, 0.5397],
  13627. [0.6127, 0.4066, 0.8550, 0.5567, 0.4662, 0.5141, 0.5070, 0.5412],
  13628. [0.0000, 0.0000, 0.7525, 0.2291, 0.3837, 0.3017, 0.6050, 0.5667]]],
  13629. device='cuda:0')
  13630. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  13631. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  13632. loss_train: 0.06275377355632372
  13633. step: 37
  13634. running loss: 0.0016960479339546953
  13635. Train Steps: 37/90 Loss: 0.0017 torch.Size([8, 600, 800])
  13636. torch.Size([8, 8])
  13637. tensor([[0.6122, 0.4006, 0.8850, 0.4217, 0.4088, 0.5517, 0.6063, 0.5517],
  13638. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  13639. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  13640. [0.6182, 0.3930, 0.8841, 0.3892, 0.3556, 0.4967, 0.6222, 0.5279],
  13641. [0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  13642. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  13643. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  13644. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5837, 0.5500]],
  13645. device='cuda:0', dtype=torch.float64)
  13646. predictions are: tensor([[0.5923, 0.4131, 0.8545, 0.3751, 0.3961, 0.5505, 0.5932, 0.5513],
  13647. [0.6174, 0.4432, 0.7769, 0.2844, 0.3408, 0.4038, 0.5863, 0.5461],
  13648. [0.5197, 0.3747, 0.8393, 0.4754, 0.3550, 0.4011, 0.5515, 0.5350],
  13649. [0.5182, 0.3501, 0.8382, 0.3487, 0.3390, 0.4937, 0.6212, 0.5191],
  13650. [0.6206, 0.4425, 0.8216, 0.4711, 0.4638, 0.5038, 0.5184, 0.5643],
  13651. [0.6210, 0.4301, 0.7798, 0.1495, 0.4564, 0.1728, 0.6289, 0.4910],
  13652. [0.6128, 0.4325, 0.8455, 0.2361, 0.4737, 0.2206, 0.6600, 0.5483],
  13653. [0.6007, 0.4236, 0.8583, 0.4114, 0.3887, 0.5358, 0.5525, 0.5445]],
  13654. device='cuda:0', grad_fn=<AddmmBackward>)
  13655. landmarks are: tensor([[[0.6122, 0.4006, 0.8850, 0.4217, 0.4087, 0.5517, 0.6062, 0.5517],
  13656. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  13657. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  13658. [0.6182, 0.3930, 0.8841, 0.3892, 0.3556, 0.4967, 0.6222, 0.5279],
  13659. [0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  13660. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  13661. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  13662. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5838, 0.5500]]],
  13663. device='cuda:0')
  13664. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  13665. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  13666. loss_train: 0.06382370277424343
  13667. step: 38
  13668. running loss: 0.001679571125637985
  13669. Train Steps: 38/90 Loss: 0.0017 torch.Size([8, 600, 800])
  13670. torch.Size([8, 8])
  13671. tensor([[0.6193, 0.4034, 0.7757, 0.2347, 0.3733, 0.2919, 0.5930, 0.4926],
  13672. [0.6079, 0.3964, 0.7420, 0.2958, 0.3563, 0.2917, 0.5351, 0.4980],
  13673. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  13674. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  13675. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  13676. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  13677. [0.6230, 0.4113, 0.7213, 0.1983, 0.4325, 0.2367, 0.6262, 0.5400],
  13678. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400]],
  13679. device='cuda:0', dtype=torch.float64)
  13680. predictions are: tensor([[0.5740, 0.4027, 0.7804, 0.2006, 0.3989, 0.3092, 0.5924, 0.5124],
  13681. [0.5530, 0.3796, 0.7529, 0.2496, 0.3690, 0.3354, 0.5114, 0.5020],
  13682. [0.6239, 0.4282, 0.7437, 0.1662, 0.4529, 0.2339, 0.5796, 0.5355],
  13683. [0.6111, 0.4053, 0.8515, 0.5777, 0.4085, 0.5404, 0.5554, 0.4905],
  13684. [0.6401, 0.4081, 0.8852, 0.4734, 0.4462, 0.5911, 0.6910, 0.5573],
  13685. [0.5619, 0.4020, 0.8669, 0.4254, 0.3718, 0.4445, 0.5490, 0.5866],
  13686. [0.6114, 0.4139, 0.7461, 0.1707, 0.4561, 0.2525, 0.6348, 0.5516],
  13687. [0.5993, 0.4027, 0.7931, 0.1914, 0.4720, 0.2240, 0.6078, 0.5352]],
  13688. device='cuda:0', grad_fn=<AddmmBackward>)
  13689. landmarks are: tensor([[[0.6193, 0.4034, 0.7757, 0.2347, 0.3733, 0.2919, 0.5930, 0.4926],
  13690. [0.6079, 0.3964, 0.7420, 0.2958, 0.3562, 0.2917, 0.5351, 0.4980],
  13691. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  13692. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  13693. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  13694. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  13695. [0.6230, 0.4113, 0.7212, 0.1983, 0.4325, 0.2367, 0.6263, 0.5400],
  13696. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400]]],
  13697. device='cuda:0')
  13698. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  13699. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  13700. loss_train: 0.06450154227786697
  13701. step: 39
  13702. running loss: 0.0016538856994324864
  13703. Train Steps: 39/90 Loss: 0.0017 torch.Size([8, 600, 800])
  13704. torch.Size([8, 8])
  13705. tensor([[0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  13706. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5837, 0.5583],
  13707. [0.6250, 0.4110, 0.7238, 0.2067, 0.4263, 0.1883, 0.5625, 0.5633],
  13708. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  13709. [0.6262, 0.4163, 0.8850, 0.5183, 0.3763, 0.4150, 0.6025, 0.5500],
  13710. [0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  13711. [0.6109, 0.4015, 0.7668, 0.3639, 0.3513, 0.3667, 0.5200, 0.5641],
  13712. [0.6277, 0.4057, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823]],
  13713. device='cuda:0', dtype=torch.float64)
  13714. predictions are: tensor([[0.5910, 0.4063, 0.8675, 0.3079, 0.3950, 0.3282, 0.6319, 0.5391],
  13715. [0.6301, 0.4437, 0.8669, 0.4167, 0.4878, 0.5472, 0.5882, 0.5389],
  13716. [0.5174, 0.3574, 0.7003, 0.1974, 0.4357, 0.2005, 0.5694, 0.5409],
  13717. [0.5996, 0.4060, 0.8421, 0.4511, 0.3845, 0.5482, 0.5982, 0.5075],
  13718. [0.5995, 0.4152, 0.8363, 0.4647, 0.3705, 0.4227, 0.5858, 0.5332],
  13719. [0.6194, 0.4364, 0.8583, 0.3602, 0.3719, 0.3915, 0.6232, 0.4786],
  13720. [0.6075, 0.4301, 0.7486, 0.3395, 0.3588, 0.4001, 0.5355, 0.5402],
  13721. [0.6480, 0.4403, 0.8199, 0.2405, 0.4479, 0.2270, 0.6470, 0.4727]],
  13722. device='cuda:0', grad_fn=<AddmmBackward>)
  13723. landmarks are: tensor([[[0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  13724. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5838, 0.5583],
  13725. [0.6250, 0.4110, 0.7237, 0.2067, 0.4263, 0.1883, 0.5625, 0.5633],
  13726. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  13727. [0.6262, 0.4163, 0.8850, 0.5183, 0.3762, 0.4150, 0.6025, 0.5500],
  13728. [0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  13729. [0.6109, 0.4015, 0.7668, 0.3639, 0.3512, 0.3667, 0.5200, 0.5641],
  13730. [0.6277, 0.4056, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823]]],
  13731. device='cuda:0')
  13732. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  13733. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  13734. loss_train: 0.06523508534883149
  13735. step: 40
  13736. running loss: 0.0016308771337207872
  13737.  
  13738. Train Steps: 40/90 Loss: 0.0016 torch.Size([8, 600, 800])
  13739. torch.Size([8, 8])
  13740. tensor([[ nan, nan, 0.8888, 0.3100, 0.5262, 0.2817, 0.7145, 0.6003],
  13741. [0.6147, 0.4081, 0.8538, 0.3400, 0.3663, 0.3150, 0.5142, 0.4875],
  13742. [0.6126, 0.4039, 0.8237, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138],
  13743. [0.6293, 0.4097, 0.8800, 0.2517, 0.5262, 0.2600, 0.7430, 0.5378],
  13744. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  13745. [0.6255, 0.4017, 0.8688, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  13746. [0.6185, 0.4129, 0.8900, 0.4567, 0.3937, 0.5417, 0.5734, 0.5110],
  13747. [0.6212, 0.4171, 0.7875, 0.3633, 0.3813, 0.2933, 0.5675, 0.5700]],
  13748. device='cuda:0', dtype=torch.float64)
  13749. predictions are: tensor([[0.0891, 0.0631, 0.8773, 0.3103, 0.5095, 0.2728, 0.7551, 0.5479],
  13750. [0.6648, 0.4404, 0.8403, 0.3713, 0.3801, 0.3216, 0.5263, 0.5005],
  13751. [0.6272, 0.4338, 0.8008, 0.4137, 0.3686, 0.3707, 0.5899, 0.5850],
  13752. [0.7378, 0.4701, 0.8775, 0.2688, 0.5286, 0.2359, 0.7385, 0.5006],
  13753. [0.5938, 0.4020, 0.6731, 0.2113, 0.4046, 0.1986, 0.5346, 0.5199],
  13754. [0.6902, 0.4420, 0.8249, 0.3392, 0.3674, 0.3924, 0.6394, 0.4805],
  13755. [0.7369, 0.4858, 0.8802, 0.4684, 0.4083, 0.5596, 0.5782, 0.5072],
  13756. [0.6505, 0.4323, 0.7702, 0.3830, 0.3922, 0.3041, 0.5700, 0.5416]],
  13757. device='cuda:0', grad_fn=<AddmmBackward>)
  13758. landmarks are: tensor([[[0.0000, 0.0000, 0.8888, 0.3100, 0.5263, 0.2817, 0.7145, 0.6003],
  13759. [0.6147, 0.4081, 0.8537, 0.3400, 0.3663, 0.3150, 0.5142, 0.4875],
  13760. [0.6126, 0.4038, 0.8238, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138],
  13761. [0.6293, 0.4097, 0.8800, 0.2517, 0.5263, 0.2600, 0.7430, 0.5378],
  13762. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  13763. [0.6255, 0.4017, 0.8687, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  13764. [0.6186, 0.4129, 0.8900, 0.4567, 0.3938, 0.5417, 0.5734, 0.5110],
  13765. [0.6212, 0.4171, 0.7875, 0.3633, 0.3812, 0.2933, 0.5675, 0.5700]]],
  13766. device='cuda:0')
  13767. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  13768. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  13769. loss_train: 0.06645582363125868
  13770. step: 41
  13771. running loss: 0.0016208737471038702
  13772. Train Steps: 41/90 Loss: 0.0016 torch.Size([8, 600, 800])
  13773. torch.Size([8, 8])
  13774. tensor([[0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  13775. [ nan, nan, 0.6512, 0.1717, 0.4100, 0.1983, 0.5253, 0.5240],
  13776. [0.6178, 0.4059, 0.7525, 0.2250, 0.4313, 0.1783, 0.5404, 0.5240],
  13777. [0.6139, 0.4019, 0.7137, 0.2150, 0.4375, 0.1533, 0.5293, 0.5006],
  13778. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  13779. [0.6215, 0.4119, 0.7688, 0.2300, 0.4200, 0.2283, 0.5925, 0.5317],
  13780. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  13781. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309]],
  13782. device='cuda:0', dtype=torch.float64)
  13783. predictions are: tensor([[0.6891, 0.4599, 0.8777, 0.4752, 0.3979, 0.5489, 0.5949, 0.5166],
  13784. [0.1478, 0.0824, 0.7012, 0.2130, 0.4257, 0.1784, 0.5816, 0.5343],
  13785. [0.4423, 0.2698, 0.7500, 0.2366, 0.4303, 0.1242, 0.5547, 0.5396],
  13786. [0.6466, 0.4050, 0.7299, 0.2141, 0.4476, 0.1427, 0.5879, 0.5038],
  13787. [0.6940, 0.4596, 0.8240, 0.3588, 0.3734, 0.3761, 0.6182, 0.5333],
  13788. [0.6037, 0.3890, 0.7641, 0.2457, 0.4580, 0.2087, 0.6180, 0.5329],
  13789. [0.6408, 0.4135, 0.8916, 0.3551, 0.4042, 0.2893, 0.6089, 0.5080],
  13790. [0.6872, 0.4480, 0.9009, 0.5420, 0.4146, 0.3651, 0.6688, 0.5193]],
  13791. device='cuda:0', grad_fn=<AddmmBackward>)
  13792. landmarks are: tensor([[[0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  13793. [0.0000, 0.0000, 0.6513, 0.1717, 0.4100, 0.1983, 0.5253, 0.5240],
  13794. [0.6178, 0.4059, 0.7525, 0.2250, 0.4313, 0.1783, 0.5404, 0.5240],
  13795. [0.6139, 0.4019, 0.7138, 0.2150, 0.4375, 0.1533, 0.5293, 0.5006],
  13796. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  13797. [0.6215, 0.4119, 0.7688, 0.2300, 0.4200, 0.2283, 0.5925, 0.5317],
  13798. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  13799. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309]]],
  13800. device='cuda:0')
  13801. loss_train_step before backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  13802. loss_train_step after backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  13803. loss_train: 0.06857384138857014
  13804. step: 42
  13805. running loss: 0.00163271050925167
  13806. Train Steps: 42/90 Loss: 0.0016 torch.Size([8, 600, 800])
  13807. torch.Size([8, 8])
  13808. tensor([[ nan, nan, 0.8300, 0.3150, 0.3588, 0.3383, 0.5208, 0.5194],
  13809. [0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6088, 0.5583],
  13810. [0.6126, 0.3954, 0.8538, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  13811. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  13812. [0.6200, 0.3978, 0.8900, 0.4550, 0.3775, 0.5200, 0.6150, 0.5367],
  13813. [0.6332, 0.4165, 0.9100, 0.3350, 0.4188, 0.3683, 0.7438, 0.5528],
  13814. [0.6122, 0.4006, 0.8850, 0.4217, 0.4088, 0.5517, 0.6063, 0.5517],
  13815. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864]],
  13816. device='cuda:0', dtype=torch.float64)
  13817. predictions are: tensor([[0.1296, 0.0703, 0.8199, 0.3338, 0.3646, 0.2900, 0.5425, 0.4966],
  13818. [0.7671, 0.4832, 0.7103, 0.2382, 0.3894, 0.2149, 0.5864, 0.5264],
  13819. [0.7036, 0.4180, 0.8707, 0.5230, 0.4307, 0.4502, 0.5363, 0.5218],
  13820. [0.6885, 0.4491, 0.8544, 0.5484, 0.4120, 0.5351, 0.6645, 0.5570],
  13821. [0.6785, 0.4147, 0.8790, 0.4961, 0.3750, 0.4876, 0.5926, 0.5207],
  13822. [0.6171, 0.3889, 0.9387, 0.3866, 0.4237, 0.3242, 0.7216, 0.5220],
  13823. [0.6880, 0.4228, 0.9020, 0.4466, 0.4167, 0.5017, 0.5754, 0.5190],
  13824. [0.7329, 0.4590, 0.8893, 0.5391, 0.3890, 0.4513, 0.6981, 0.5399]],
  13825. device='cuda:0', grad_fn=<AddmmBackward>)
  13826. landmarks are: tensor([[[0.0000, 0.0000, 0.8300, 0.3150, 0.3587, 0.3383, 0.5208, 0.5194],
  13827. [0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6087, 0.5583],
  13828. [0.6126, 0.3954, 0.8537, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  13829. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  13830. [0.6199, 0.3978, 0.8900, 0.4550, 0.3775, 0.5200, 0.6150, 0.5367],
  13831. [0.6332, 0.4165, 0.9100, 0.3350, 0.4187, 0.3683, 0.7438, 0.5528],
  13832. [0.6122, 0.4006, 0.8850, 0.4217, 0.4087, 0.5517, 0.6062, 0.5517],
  13833. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864]]],
  13834. device='cuda:0')
  13835. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  13836. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  13837. loss_train: 0.07046938224812038
  13838. step: 43
  13839. running loss: 0.0016388228429795439
  13840. Train Steps: 43/90 Loss: 0.0016 torch.Size([8, 600, 800])
  13841. torch.Size([8, 8])
  13842. tensor([[0.6339, 0.4118, 0.7988, 0.5800, 0.3912, 0.4583, 0.7343, 0.5760],
  13843. [0.6126, 0.3954, 0.8538, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  13844. [0.6202, 0.4066, 0.8746, 0.3376, 0.3717, 0.3090, 0.5842, 0.5165],
  13845. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  13846. [0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  13847. [0.6262, 0.4052, 0.8888, 0.4700, 0.3675, 0.5117, 0.6350, 0.5233],
  13848. [0.6353, 0.4128, 0.8488, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  13849. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103]],
  13850. device='cuda:0', dtype=torch.float64)
  13851. predictions are: tensor([[0.6533, 0.3930, 0.8091, 0.5610, 0.3700, 0.4263, 0.6663, 0.5426],
  13852. [0.6025, 0.3455, 0.8612, 0.5094, 0.4215, 0.4670, 0.5519, 0.5381],
  13853. [0.6022, 0.3682, 0.8667, 0.3334, 0.3617, 0.2666, 0.5935, 0.5086],
  13854. [0.5780, 0.3655, 0.7793, 0.3759, 0.3514, 0.4040, 0.5146, 0.5343],
  13855. [0.6462, 0.4082, 0.9039, 0.5446, 0.3658, 0.4411, 0.5685, 0.5704],
  13856. [0.6143, 0.3738, 0.9155, 0.4762, 0.3818, 0.5059, 0.6533, 0.5145],
  13857. [0.5468, 0.3345, 0.8641, 0.2554, 0.5273, 0.1434, 0.6819, 0.5426],
  13858. [0.5857, 0.3659, 0.8191, 0.3233, 0.3502, 0.3641, 0.5801, 0.5197]],
  13859. device='cuda:0', grad_fn=<AddmmBackward>)
  13860. landmarks are: tensor([[[0.6339, 0.4118, 0.7987, 0.5800, 0.3913, 0.4583, 0.7343, 0.5760],
  13861. [0.6126, 0.3954, 0.8537, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  13862. [0.6202, 0.4066, 0.8746, 0.3376, 0.3717, 0.3090, 0.5842, 0.5165],
  13863. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  13864. [0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  13865. [0.6262, 0.4052, 0.8888, 0.4700, 0.3675, 0.5117, 0.6350, 0.5233],
  13866. [0.6353, 0.4128, 0.8487, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  13867. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103]]],
  13868. device='cuda:0')
  13869. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  13870. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  13871. loss_train: 0.07114345880108885
  13872. step: 44
  13873. running loss: 0.0016168967909338376
  13874.  
  13875. Train Steps: 44/90 Loss: 0.0016 torch.Size([8, 600, 800])
  13876. torch.Size([8, 8])
  13877. tensor([[0.6296, 0.4076, 0.8400, 0.5583, 0.3700, 0.4367, 0.6876, 0.5494],
  13878. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  13879. [0.6208, 0.4082, 0.8538, 0.3067, 0.3588, 0.3717, 0.6112, 0.5517],
  13880. [0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950],
  13881. [ nan, nan, 0.8525, 0.2217, 0.5413, 0.2367, 0.7367, 0.5482],
  13882. [0.6231, 0.3973, 0.8650, 0.3950, 0.3625, 0.3183, 0.5837, 0.5167],
  13883. [0.6219, 0.3934, 0.8688, 0.5267, 0.4313, 0.4967, 0.5988, 0.4983],
  13884. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167]],
  13885. device='cuda:0', dtype=torch.float64)
  13886. predictions are: tensor([[0.6648, 0.4051, 0.8547, 0.5253, 0.3593, 0.4408, 0.6620, 0.5442],
  13887. [0.5646, 0.3528, 0.8634, 0.4902, 0.3912, 0.5007, 0.6026, 0.5536],
  13888. [0.6554, 0.4185, 0.8312, 0.3112, 0.3282, 0.3703, 0.5874, 0.5649],
  13889. [0.6001, 0.3667, 0.8595, 0.5331, 0.3922, 0.5166, 0.5689, 0.5086],
  13890. [0.2252, 0.1352, 0.8452, 0.2125, 0.5094, 0.2397, 0.7010, 0.5618],
  13891. [0.6177, 0.3958, 0.8650, 0.4102, 0.3560, 0.3034, 0.5887, 0.5341],
  13892. [0.6059, 0.3636, 0.8642, 0.5232, 0.4138, 0.4863, 0.5702, 0.5012],
  13893. [0.6755, 0.4330, 0.7479, 0.3293, 0.4549, 0.1847, 0.5319, 0.6135]],
  13894. device='cuda:0', grad_fn=<AddmmBackward>)
  13895. landmarks are: tensor([[[0.6296, 0.4076, 0.8400, 0.5583, 0.3700, 0.4367, 0.6876, 0.5494],
  13896. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  13897. [0.6208, 0.4082, 0.8537, 0.3067, 0.3587, 0.3717, 0.6112, 0.5517],
  13898. [0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950],
  13899. [0.0000, 0.0000, 0.8525, 0.2217, 0.5412, 0.2367, 0.7367, 0.5482],
  13900. [0.6231, 0.3973, 0.8650, 0.3950, 0.3625, 0.3183, 0.5838, 0.5167],
  13901. [0.6219, 0.3934, 0.8687, 0.5267, 0.4313, 0.4967, 0.5987, 0.4983],
  13902. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167]]],
  13903. device='cuda:0')
  13904. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  13905. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  13906. loss_train: 0.07271968291024677
  13907. step: 45
  13908. running loss: 0.0016159929535610395
  13909. Train Steps: 45/90 Loss: 0.0016 torch.Size([8, 600, 800])
  13910. torch.Size([8, 8])
  13911. tensor([[0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  13912. [0.6068, 0.3963, 0.8650, 0.4317, 0.4037, 0.5083, 0.5253, 0.4999],
  13913. [0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  13914. [0.6305, 0.3983, 0.8950, 0.4833, 0.3688, 0.4683, 0.6375, 0.5117],
  13915. [0.6241, 0.4143, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  13916. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  13917. [0.6353, 0.4128, 0.9138, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  13918. [0.6109, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117]],
  13919. device='cuda:0', dtype=torch.float64)
  13920. predictions are: tensor([[0.6014, 0.3906, 0.8760, 0.5141, 0.3662, 0.3807, 0.5987, 0.5159],
  13921. [0.5682, 0.3432, 0.8353, 0.4166, 0.3854, 0.5007, 0.5066, 0.5207],
  13922. [0.5977, 0.3875, 0.8380, 0.5157, 0.3754, 0.5035, 0.6486, 0.5565],
  13923. [0.6293, 0.3856, 0.8906, 0.4898, 0.3633, 0.4632, 0.6219, 0.5406],
  13924. [0.6094, 0.3830, 0.8745, 0.4607, 0.4074, 0.5207, 0.6207, 0.5795],
  13925. [0.5853, 0.3684, 0.8890, 0.4049, 0.4215, 0.3655, 0.7029, 0.5958],
  13926. [0.4055, 0.2576, 0.8998, 0.3556, 0.4626, 0.3115, 0.7230, 0.6119],
  13927. [0.6224, 0.4179, 0.8621, 0.4533, 0.3715, 0.4162, 0.5211, 0.5285]],
  13928. device='cuda:0', grad_fn=<AddmmBackward>)
  13929. landmarks are: tensor([[[0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  13930. [0.6068, 0.3963, 0.8650, 0.4317, 0.4038, 0.5083, 0.5253, 0.4999],
  13931. [0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  13932. [0.6305, 0.3983, 0.8950, 0.4833, 0.3688, 0.4683, 0.6375, 0.5117],
  13933. [0.6241, 0.4142, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  13934. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  13935. [0.6353, 0.4128, 0.9137, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  13936. [0.6108, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117]]],
  13937. device='cuda:0')
  13938. loss_train_step before backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  13939. loss_train_step after backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  13940. loss_train: 0.07447020555264316
  13941. step: 46
  13942. running loss: 0.0016189175120139819
  13943. Train Steps: 46/90 Loss: 0.0016 torch.Size([8, 600, 800])
  13944. torch.Size([8, 8])
  13945. tensor([[0.6267, 0.4080, 0.8438, 0.2633, 0.4763, 0.1800, 0.6259, 0.5240],
  13946. [0.6118, 0.4052, 0.8463, 0.3917, 0.3538, 0.3450, 0.5053, 0.5593],
  13947. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  13948. [0.6102, 0.3999, 0.8750, 0.5133, 0.3825, 0.4750, 0.5637, 0.5083],
  13949. [0.6231, 0.3973, 0.8650, 0.3950, 0.3625, 0.3183, 0.5837, 0.5167],
  13950. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  13951. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  13952. [0.6204, 0.4049, 0.7975, 0.2700, 0.3937, 0.2567, 0.5700, 0.5183]],
  13953. device='cuda:0', dtype=torch.float64)
  13954. predictions are: tensor([[0.6041, 0.3938, 0.8301, 0.2768, 0.4639, 0.2062, 0.6434, 0.5514],
  13955. [0.6029, 0.4004, 0.8234, 0.3782, 0.3560, 0.3713, 0.5147, 0.5802],
  13956. [0.5090, 0.3451, 0.8887, 0.4574, 0.4401, 0.5328, 0.6144, 0.5899],
  13957. [0.5273, 0.3276, 0.8808, 0.5208, 0.3927, 0.5246, 0.5674, 0.5295],
  13958. [0.5351, 0.3431, 0.8607, 0.4114, 0.3778, 0.3320, 0.6319, 0.5436],
  13959. [0.5631, 0.3568, 0.8728, 0.2954, 0.5005, 0.2429, 0.7542, 0.5450],
  13960. [0.5440, 0.3390, 0.8046, 0.2910, 0.4050, 0.2686, 0.6255, 0.5356],
  13961. [0.4561, 0.2983, 0.7701, 0.2767, 0.3897, 0.2866, 0.5682, 0.5719]],
  13962. device='cuda:0', grad_fn=<AddmmBackward>)
  13963. landmarks are: tensor([[[0.6267, 0.4080, 0.8438, 0.2633, 0.4762, 0.1800, 0.6259, 0.5240],
  13964. [0.6118, 0.4052, 0.8462, 0.3917, 0.3537, 0.3450, 0.5053, 0.5593],
  13965. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  13966. [0.6102, 0.3999, 0.8750, 0.5133, 0.3825, 0.4750, 0.5638, 0.5083],
  13967. [0.6231, 0.3973, 0.8650, 0.3950, 0.3625, 0.3183, 0.5838, 0.5167],
  13968. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  13969. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  13970. [0.6204, 0.4049, 0.7975, 0.2700, 0.3938, 0.2567, 0.5700, 0.5183]]],
  13971. device='cuda:0')
  13972. loss_train_step before backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  13973. loss_train_step after backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  13974. loss_train: 0.07631988901994191
  13975. step: 47
  13976. running loss: 0.0016238274259562108
  13977. Train Steps: 47/90 Loss: 0.0016 torch.Size([8, 600, 800])
  13978. torch.Size([8, 8])
  13979. tensor([[0.6164, 0.4076, 0.8838, 0.4117, 0.3713, 0.5550, 0.6238, 0.5350],
  13980. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864],
  13981. [0.6137, 0.4038, 0.8563, 0.4050, 0.3813, 0.2550, 0.5106, 0.4954],
  13982. [0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950],
  13983. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  13984. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  13985. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  13986. [0.6239, 0.4174, 0.8425, 0.5733, 0.4825, 0.4500, 0.5625, 0.5933]],
  13987. device='cuda:0', dtype=torch.float64)
  13988. predictions are: tensor([[0.5418, 0.3592, 0.8929, 0.3954, 0.3914, 0.5638, 0.6728, 0.5593],
  13989. [0.5866, 0.3958, 0.8892, 0.5073, 0.3876, 0.5094, 0.7513, 0.5850],
  13990. [0.6361, 0.4119, 0.8607, 0.3998, 0.4074, 0.2834, 0.5328, 0.5083],
  13991. [0.5712, 0.3631, 0.8827, 0.5297, 0.4291, 0.5318, 0.6194, 0.4938],
  13992. [0.5730, 0.3708, 0.8171, 0.2721, 0.4153, 0.2532, 0.6215, 0.5359],
  13993. [0.5721, 0.3738, 0.7804, 0.2966, 0.3465, 0.3352, 0.5659, 0.5552],
  13994. [0.5397, 0.3652, 0.8544, 0.3131, 0.3652, 0.3929, 0.6385, 0.5626],
  13995. [0.5920, 0.4128, 0.8771, 0.5720, 0.4985, 0.4259, 0.6170, 0.6088]],
  13996. device='cuda:0', grad_fn=<AddmmBackward>)
  13997. landmarks are: tensor([[[0.6164, 0.4076, 0.8838, 0.4117, 0.3713, 0.5550, 0.6237, 0.5350],
  13998. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864],
  13999. [0.6137, 0.4038, 0.8562, 0.4050, 0.3812, 0.2550, 0.5106, 0.4954],
  14000. [0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950],
  14001. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  14002. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  14003. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  14004. [0.6239, 0.4174, 0.8425, 0.5733, 0.4825, 0.4500, 0.5625, 0.5933]]],
  14005. device='cuda:0')
  14006. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  14007. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  14008. loss_train: 0.0770722002198454
  14009. step: 48
  14010. running loss: 0.0016056708379134459
  14011.  
  14012. Train Steps: 48/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14013. torch.Size([8, 8])
  14014. tensor([[0.6102, 0.4020, 0.8638, 0.3717, 0.3625, 0.5017, 0.6038, 0.5500],
  14015. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  14016. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5413, 0.5717],
  14017. [0.6095, 0.3970, 0.8688, 0.4767, 0.4860, 0.4879, 0.5191, 0.4940],
  14018. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  14019. [ nan, nan, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733],
  14020. [0.6262, 0.4163, 0.8850, 0.5183, 0.3763, 0.4150, 0.6025, 0.5500],
  14021. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208]],
  14022. device='cuda:0', dtype=torch.float64)
  14023. predictions are: tensor([[0.6116, 0.4067, 0.8828, 0.3820, 0.3839, 0.5019, 0.6557, 0.5338],
  14024. [0.7061, 0.4629, 0.8807, 0.5306, 0.3652, 0.3922, 0.5974, 0.5166],
  14025. [0.6063, 0.4012, 0.8859, 0.4928, 0.4343, 0.4867, 0.5892, 0.5541],
  14026. [0.5990, 0.3772, 0.9000, 0.4734, 0.4776, 0.4876, 0.5473, 0.5143],
  14027. [0.6460, 0.4354, 0.8652, 0.4529, 0.3676, 0.4091, 0.5995, 0.5648],
  14028. [0.0335, 0.0334, 0.7170, 0.1800, 0.4606, 0.2172, 0.5830, 0.5723],
  14029. [0.6761, 0.4458, 0.8849, 0.4941, 0.3624, 0.4209, 0.6187, 0.5309],
  14030. [0.6916, 0.4472, 0.8236, 0.2844, 0.4038, 0.2530, 0.6177, 0.5026]],
  14031. device='cuda:0', grad_fn=<AddmmBackward>)
  14032. landmarks are: tensor([[[0.6102, 0.4020, 0.8637, 0.3717, 0.3625, 0.5017, 0.6037, 0.5500],
  14033. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  14034. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5412, 0.5717],
  14035. [0.6095, 0.3970, 0.8687, 0.4767, 0.4860, 0.4879, 0.5191, 0.4940],
  14036. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  14037. [0.0000, 0.0000, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733],
  14038. [0.6262, 0.4163, 0.8850, 0.5183, 0.3762, 0.4150, 0.6025, 0.5500],
  14039. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208]]],
  14040. device='cuda:0')
  14041. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  14042. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  14043. loss_train: 0.07790248477249406
  14044. step: 49
  14045. running loss: 0.001589846628010083
  14046. Train Steps: 49/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14047. torch.Size([8, 8])
  14048. tensor([[0.6038, 0.3946, 0.8413, 0.4883, 0.3563, 0.4550, 0.5266, 0.4693],
  14049. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  14050. [0.6117, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103],
  14051. [0.6203, 0.4073, 0.8189, 0.2398, 0.4400, 0.2054, 0.5929, 0.5501],
  14052. [0.6260, 0.4214, 0.8538, 0.5500, 0.3663, 0.3767, 0.5587, 0.5983],
  14053. [ nan, nan, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  14054. [0.6168, 0.4029, 0.8523, 0.3417, 0.3588, 0.5000, 0.6125, 0.5400],
  14055. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882]],
  14056. device='cuda:0', dtype=torch.float64)
  14057. predictions are: tensor([[ 0.5667, 0.3969, 0.8627, 0.5026, 0.3836, 0.4889, 0.5366, 0.5161],
  14058. [ 0.7085, 0.4804, 0.8648, 0.3856, 0.3830, 0.3395, 0.6228, 0.5103],
  14059. [ 0.6474, 0.4425, 0.6780, 0.2258, 0.3905, 0.2618, 0.5197, 0.4995],
  14060. [ 0.7389, 0.5162, 0.8219, 0.2639, 0.4505, 0.2242, 0.6105, 0.5333],
  14061. [ 0.7038, 0.4682, 0.8615, 0.5522, 0.3831, 0.4068, 0.6023, 0.5929],
  14062. [-0.0587, -0.0297, 0.9008, 0.3119, 0.5381, 0.2447, 0.6911, 0.5574],
  14063. [ 0.6270, 0.4166, 0.8636, 0.3521, 0.3723, 0.5205, 0.6568, 0.5517],
  14064. [ 0.6500, 0.4548, 0.8787, 0.3977, 0.3747, 0.4232, 0.5579, 0.4989]],
  14065. device='cuda:0', grad_fn=<AddmmBackward>)
  14066. landmarks are: tensor([[[0.6038, 0.3946, 0.8413, 0.4883, 0.3562, 0.4550, 0.5266, 0.4693],
  14067. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  14068. [0.6116, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103],
  14069. [0.6203, 0.4073, 0.8189, 0.2398, 0.4400, 0.2054, 0.5929, 0.5501],
  14070. [0.6260, 0.4214, 0.8537, 0.5500, 0.3663, 0.3767, 0.5587, 0.5983],
  14071. [0.0000, 0.0000, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  14072. [0.6168, 0.4029, 0.8523, 0.3417, 0.3587, 0.5000, 0.6125, 0.5400],
  14073. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882]]],
  14074. device='cuda:0')
  14075. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  14076. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  14077. loss_train: 0.07921916476334445
  14078. step: 50
  14079. running loss: 0.001584383295266889
  14080. Train Steps: 50/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14081. torch.Size([8, 8])
  14082. tensor([[0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  14083. [0.6265, 0.4091, 0.8950, 0.3533, 0.3600, 0.3967, 0.6295, 0.4901],
  14084. [0.6087, 0.3951, 0.8387, 0.5833, 0.4188, 0.4933, 0.5146, 0.4830],
  14085. [0.6118, 0.4052, 0.8463, 0.3917, 0.3538, 0.3450, 0.5053, 0.5593],
  14086. [0.6250, 0.4146, 0.8838, 0.3933, 0.3588, 0.4283, 0.6162, 0.5367],
  14087. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  14088. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  14089. [0.6218, 0.4098, 0.7238, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350]],
  14090. device='cuda:0', dtype=torch.float64)
  14091. predictions are: tensor([[0.5915, 0.4134, 0.7551, 0.3165, 0.4903, 0.1958, 0.5103, 0.5853],
  14092. [0.5764, 0.3818, 0.9127, 0.3342, 0.3573, 0.3764, 0.6141, 0.4978],
  14093. [0.5345, 0.3572, 0.8517, 0.5715, 0.4077, 0.4864, 0.5188, 0.4861],
  14094. [0.5781, 0.4059, 0.8441, 0.3646, 0.3541, 0.3400, 0.4695, 0.5328],
  14095. [0.5979, 0.4039, 0.8465, 0.3703, 0.3558, 0.4157, 0.5717, 0.5015],
  14096. [0.5912, 0.4007, 0.9070, 0.3955, 0.4255, 0.3726, 0.6829, 0.5573],
  14097. [0.5950, 0.4042, 0.8472, 0.5555, 0.3779, 0.4802, 0.6432, 0.5175],
  14098. [0.6436, 0.4484, 0.7460, 0.2079, 0.4379, 0.2331, 0.5944, 0.5265]],
  14099. device='cuda:0', grad_fn=<AddmmBackward>)
  14100. landmarks are: tensor([[[0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  14101. [0.6265, 0.4091, 0.8950, 0.3533, 0.3600, 0.3967, 0.6295, 0.4901],
  14102. [0.6087, 0.3951, 0.8388, 0.5833, 0.4187, 0.4933, 0.5146, 0.4830],
  14103. [0.6118, 0.4052, 0.8462, 0.3917, 0.3537, 0.3450, 0.5053, 0.5593],
  14104. [0.6250, 0.4146, 0.8838, 0.3933, 0.3587, 0.4283, 0.6162, 0.5367],
  14105. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  14106. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  14107. [0.6218, 0.4098, 0.7237, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350]]],
  14108. device='cuda:0')
  14109. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  14110. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  14111. loss_train: 0.07985461122007109
  14112. step: 51
  14113. running loss: 0.0015657766905896292
  14114. Train Steps: 51/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14115. torch.Size([8, 8])
  14116. tensor([[0.6268, 0.4061, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  14117. [0.6230, 0.4113, 0.7213, 0.1983, 0.4325, 0.2367, 0.6262, 0.5400],
  14118. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  14119. [0.6128, 0.4115, 0.7934, 0.3778, 0.3450, 0.4033, 0.5337, 0.5456],
  14120. [0.6215, 0.4119, 0.7688, 0.2300, 0.4200, 0.2283, 0.5925, 0.5317],
  14121. [0.6218, 0.4098, 0.7238, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  14122. [0.6154, 0.4112, 0.7037, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  14123. [0.6268, 0.4029, 0.8500, 0.2683, 0.3937, 0.3500, 0.6860, 0.5297]],
  14124. device='cuda:0', dtype=torch.float64)
  14125. predictions are: tensor([[0.5507, 0.3677, 0.8739, 0.2727, 0.4537, 0.1950, 0.5911, 0.5079],
  14126. [0.6034, 0.4039, 0.7401, 0.2273, 0.4371, 0.2180, 0.5913, 0.5303],
  14127. [0.5627, 0.3646, 0.9055, 0.5243, 0.4412, 0.4729, 0.5209, 0.4831],
  14128. [0.6226, 0.3989, 0.8272, 0.4100, 0.3243, 0.4034, 0.5167, 0.5258],
  14129. [0.5484, 0.3707, 0.7738, 0.2498, 0.4351, 0.2388, 0.5541, 0.5149],
  14130. [0.6328, 0.4251, 0.7578, 0.2479, 0.4367, 0.2240, 0.5967, 0.5330],
  14131. [0.5407, 0.3771, 0.7163, 0.2564, 0.4186, 0.1785, 0.4994, 0.5358],
  14132. [0.5917, 0.3917, 0.8435, 0.3103, 0.3848, 0.3313, 0.6699, 0.5156]],
  14133. device='cuda:0', grad_fn=<AddmmBackward>)
  14134. landmarks are: tensor([[[0.6268, 0.4060, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  14135. [0.6230, 0.4113, 0.7212, 0.1983, 0.4325, 0.2367, 0.6263, 0.5400],
  14136. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  14137. [0.6128, 0.4115, 0.7934, 0.3778, 0.3450, 0.4033, 0.5337, 0.5456],
  14138. [0.6215, 0.4119, 0.7688, 0.2300, 0.4200, 0.2283, 0.5925, 0.5317],
  14139. [0.6218, 0.4098, 0.7237, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  14140. [0.6154, 0.4112, 0.7038, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  14141. [0.6268, 0.4029, 0.8500, 0.2683, 0.3938, 0.3500, 0.6860, 0.5297]]],
  14142. device='cuda:0')
  14143. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  14144. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  14145. loss_train: 0.08078289640252478
  14146. step: 52
  14147. running loss: 0.001553517238510092
  14148.  
  14149. Train Steps: 52/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14150. torch.Size([8, 8])
  14151. tensor([[ nan, nan, 0.7612, 0.3250, 0.4037, 0.2533, 0.5438, 0.5767],
  14152. [0.6111, 0.3995, 0.8788, 0.4567, 0.3813, 0.4833, 0.5450, 0.5700],
  14153. [ nan, nan, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600],
  14154. [0.6202, 0.4053, 0.8638, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  14155. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  14156. [0.6268, 0.4102, 0.8938, 0.3667, 0.4025, 0.2833, 0.6275, 0.5183],
  14157. [0.6218, 0.4185, 0.7338, 0.2650, 0.4625, 0.1950, 0.5687, 0.5800],
  14158. [0.6273, 0.4100, 0.7137, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633]],
  14159. device='cuda:0', dtype=torch.float64)
  14160. predictions are: tensor([[0.1549, 0.0988, 0.7728, 0.2984, 0.3965, 0.2483, 0.5293, 0.5679],
  14161. [0.7679, 0.4968, 0.8719, 0.4534, 0.3593, 0.4609, 0.5512, 0.5227],
  14162. [0.2748, 0.1950, 0.7612, 0.2568, 0.3874, 0.2569, 0.5394, 0.5318],
  14163. [0.7197, 0.4644, 0.8651, 0.5402, 0.4529, 0.4879, 0.6104, 0.5005],
  14164. [0.6871, 0.4434, 0.8564, 0.4625, 0.3680, 0.4583, 0.5583, 0.5248],
  14165. [0.7213, 0.4817, 0.8850, 0.3534, 0.3950, 0.2741, 0.6116, 0.5200],
  14166. [0.7321, 0.4945, 0.7385, 0.2308, 0.4484, 0.1679, 0.5341, 0.5670],
  14167. [0.6125, 0.3907, 0.7119, 0.1921, 0.3792, 0.2419, 0.5768, 0.5231]],
  14168. device='cuda:0', grad_fn=<AddmmBackward>)
  14169. landmarks are: tensor([[[0.0000, 0.0000, 0.7613, 0.3250, 0.4038, 0.2533, 0.5437, 0.5767],
  14170. [0.6111, 0.3995, 0.8788, 0.4567, 0.3812, 0.4833, 0.5450, 0.5700],
  14171. [0.0000, 0.0000, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600],
  14172. [0.6202, 0.4053, 0.8637, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  14173. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  14174. [0.6268, 0.4102, 0.8938, 0.3667, 0.4025, 0.2833, 0.6275, 0.5183],
  14175. [0.6218, 0.4185, 0.7337, 0.2650, 0.4625, 0.1950, 0.5688, 0.5800],
  14176. [0.6273, 0.4099, 0.7138, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633]]],
  14177. device='cuda:0')
  14178. loss_train_step before backward: tensor(0.0040, device='cuda:0', grad_fn=<MseLossBackward>)
  14179. loss_train_step after backward: tensor(0.0040, device='cuda:0', grad_fn=<MseLossBackward>)
  14180. loss_train: 0.08473318166215904
  14181. step: 53
  14182. running loss: 0.00159873927664451
  14183. Train Steps: 53/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14184. torch.Size([8, 8])
  14185. tensor([[0.6131, 0.4064, 0.8638, 0.5200, 0.4788, 0.4783, 0.5258, 0.5867],
  14186. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  14187. [0.6233, 0.4091, 0.8100, 0.2950, 0.3563, 0.3883, 0.6013, 0.5200],
  14188. [0.6099, 0.4030, 0.8638, 0.5117, 0.4983, 0.4965, 0.5086, 0.5388],
  14189. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  14190. [0.6250, 0.4110, 0.7238, 0.2067, 0.4263, 0.1883, 0.5625, 0.5633],
  14191. [0.6304, 0.4024, 0.8925, 0.4800, 0.3937, 0.4817, 0.7485, 0.5297],
  14192. [ nan, nan, 0.9088, 0.3783, 0.4562, 0.2617, 0.6741, 0.5575]],
  14193. device='cuda:0', dtype=torch.float64)
  14194. predictions are: tensor([[0.6802, 0.4430, 0.8318, 0.5072, 0.4398, 0.4642, 0.5124, 0.5521],
  14195. [0.6704, 0.4308, 0.8398, 0.3959, 0.3240, 0.3879, 0.5255, 0.5485],
  14196. [0.6728, 0.4352, 0.7903, 0.2926, 0.3243, 0.3784, 0.5523, 0.5330],
  14197. [0.6356, 0.4171, 0.8449, 0.5169, 0.4777, 0.4851, 0.4889, 0.5001],
  14198. [0.7161, 0.4596, 0.8432, 0.2972, 0.3887, 0.3969, 0.6283, 0.5522],
  14199. [0.6475, 0.4336, 0.6810, 0.2289, 0.3940, 0.1639, 0.4837, 0.5542],
  14200. [0.7142, 0.4602, 0.8448, 0.4690, 0.3557, 0.4662, 0.6634, 0.5211],
  14201. [0.1883, 0.1220, 0.8808, 0.3820, 0.4354, 0.2764, 0.6137, 0.5517]],
  14202. device='cuda:0', grad_fn=<AddmmBackward>)
  14203. landmarks are: tensor([[[0.6132, 0.4063, 0.8637, 0.5200, 0.4787, 0.4783, 0.5258, 0.5867],
  14204. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  14205. [0.6233, 0.4091, 0.8100, 0.2950, 0.3562, 0.3883, 0.6012, 0.5200],
  14206. [0.6098, 0.4030, 0.8637, 0.5117, 0.4983, 0.4965, 0.5086, 0.5388],
  14207. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  14208. [0.6250, 0.4110, 0.7237, 0.2067, 0.4263, 0.1883, 0.5625, 0.5633],
  14209. [0.6304, 0.4024, 0.8925, 0.4800, 0.3938, 0.4817, 0.7485, 0.5297],
  14210. [0.0000, 0.0000, 0.9087, 0.3783, 0.4563, 0.2617, 0.6741, 0.5575]]],
  14211. device='cuda:0')
  14212. loss_train_step before backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  14213. loss_train_step after backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  14214. loss_train: 0.0869114036031533
  14215. step: 54
  14216. running loss: 0.0016094704370954315
  14217. Train Steps: 54/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14218. torch.Size([8, 8])
  14219. tensor([[0.6182, 0.4099, 0.7812, 0.3000, 0.3937, 0.2367, 0.5325, 0.5750],
  14220. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  14221. [0.6200, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  14222. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  14223. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  14224. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  14225. [0.6127, 0.4066, 0.8550, 0.5567, 0.4662, 0.5141, 0.5070, 0.5412],
  14226. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466]],
  14227. device='cuda:0', dtype=torch.float64)
  14228. predictions are: tensor([[0.4948, 0.3271, 0.7504, 0.2717, 0.3895, 0.2462, 0.5128, 0.5725],
  14229. [0.6211, 0.3864, 0.8843, 0.3831, 0.4266, 0.3624, 0.7101, 0.5913],
  14230. [0.5479, 0.3498, 0.8209, 0.2813, 0.4246, 0.2453, 0.5822, 0.5351],
  14231. [0.5730, 0.3568, 0.8808, 0.3202, 0.3858, 0.3009, 0.5986, 0.5536],
  14232. [0.6147, 0.3810, 0.8437, 0.3561, 0.3642, 0.3641, 0.5878, 0.5957],
  14233. [0.5775, 0.3750, 0.6776, 0.2280, 0.4179, 0.2321, 0.5617, 0.5697],
  14234. [0.5753, 0.3730, 0.8436, 0.5263, 0.4640, 0.4712, 0.5426, 0.5266],
  14235. [0.5807, 0.3767, 0.7651, 0.3495, 0.3590, 0.4238, 0.5293, 0.5445]],
  14236. device='cuda:0', grad_fn=<AddmmBackward>)
  14237. landmarks are: tensor([[[0.6182, 0.4099, 0.7812, 0.3000, 0.3938, 0.2367, 0.5325, 0.5750],
  14238. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  14239. [0.6199, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  14240. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  14241. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  14242. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  14243. [0.6127, 0.4066, 0.8550, 0.5567, 0.4662, 0.5141, 0.5070, 0.5412],
  14244. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466]]],
  14245. device='cuda:0')
  14246. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  14247. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  14248. loss_train: 0.08789730866556056
  14249. step: 55
  14250. running loss: 0.0015981328848283737
  14251. Train Steps: 55/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14252. torch.Size([8, 8])
  14253. tensor([[0.6115, 0.4005, 0.8838, 0.3867, 0.3763, 0.4700, 0.5800, 0.5550],
  14254. [0.6068, 0.3963, 0.8650, 0.4317, 0.4037, 0.5083, 0.5253, 0.4999],
  14255. [ nan, nan, 0.9088, 0.3783, 0.4562, 0.2617, 0.6741, 0.5575],
  14256. [0.6270, 0.4267, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  14257. [0.6275, 0.4081, 0.8063, 0.2017, 0.4825, 0.1583, 0.6156, 0.4869],
  14258. [0.6203, 0.4096, 0.8862, 0.4267, 0.3538, 0.4117, 0.6025, 0.5650],
  14259. [0.6188, 0.4099, 0.7400, 0.2433, 0.3962, 0.2750, 0.6162, 0.5467],
  14260. [0.6127, 0.4118, 0.8650, 0.5083, 0.4088, 0.5367, 0.5300, 0.5456]],
  14261. device='cuda:0', dtype=torch.float64)
  14262. predictions are: tensor([[0.6512, 0.4229, 0.8618, 0.3790, 0.3667, 0.4765, 0.5779, 0.5789],
  14263. [0.6838, 0.4149, 0.8431, 0.4274, 0.3979, 0.5157, 0.5459, 0.5217],
  14264. [0.1031, 0.0614, 0.8903, 0.3869, 0.4564, 0.2801, 0.6815, 0.5916],
  14265. [0.5853, 0.3887, 0.6898, 0.2776, 0.4559, 0.1826, 0.5637, 0.6347],
  14266. [0.6206, 0.3911, 0.7994, 0.2031, 0.4864, 0.1657, 0.6258, 0.5273],
  14267. [0.6177, 0.4136, 0.8885, 0.4072, 0.3660, 0.4113, 0.6229, 0.6092],
  14268. [0.6280, 0.4210, 0.7120, 0.2187, 0.3850, 0.2793, 0.6336, 0.5902],
  14269. [0.6331, 0.4122, 0.8470, 0.4942, 0.4314, 0.5270, 0.5640, 0.5644]],
  14270. device='cuda:0', grad_fn=<AddmmBackward>)
  14271. landmarks are: tensor([[[0.6115, 0.4005, 0.8838, 0.3867, 0.3762, 0.4700, 0.5800, 0.5550],
  14272. [0.6068, 0.3963, 0.8650, 0.4317, 0.4038, 0.5083, 0.5253, 0.4999],
  14273. [0.0000, 0.0000, 0.9087, 0.3783, 0.4563, 0.2617, 0.6741, 0.5575],
  14274. [0.6270, 0.4266, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  14275. [0.6275, 0.4081, 0.8062, 0.2017, 0.4825, 0.1583, 0.6156, 0.4869],
  14276. [0.6203, 0.4096, 0.8863, 0.4267, 0.3537, 0.4117, 0.6025, 0.5650],
  14277. [0.6188, 0.4099, 0.7400, 0.2433, 0.3963, 0.2750, 0.6162, 0.5467],
  14278. [0.6127, 0.4118, 0.8650, 0.5083, 0.4087, 0.5367, 0.5300, 0.5456]]],
  14279. device='cuda:0')
  14280. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  14281. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  14282. loss_train: 0.08864424467901699
  14283. step: 56
  14284. running loss: 0.001582932940696732
  14285.  
  14286. Train Steps: 56/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14287. torch.Size([8, 8])
  14288. tensor([[0.6346, 0.4144, 0.9088, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  14289. [ nan, nan, 0.8938, 0.2850, 0.4662, 0.3117, 0.7406, 0.5528],
  14290. [0.6229, 0.4066, 0.8513, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  14291. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317],
  14292. [0.6200, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  14293. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  14294. [0.6132, 0.4066, 0.7259, 0.2402, 0.3588, 0.3300, 0.6000, 0.5600],
  14295. [0.6192, 0.4128, 0.8513, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633]],
  14296. device='cuda:0', dtype=torch.float64)
  14297. predictions are: tensor([[0.6746, 0.4382, 0.8947, 0.4301, 0.4024, 0.4315, 0.6823, 0.5842],
  14298. [0.0795, 0.0425, 0.8813, 0.2713, 0.4756, 0.3016, 0.7231, 0.5805],
  14299. [0.6175, 0.3949, 0.8288, 0.5494, 0.4417, 0.4728, 0.5768, 0.5437],
  14300. [0.5646, 0.3568, 0.8741, 0.5262, 0.3934, 0.3711, 0.5676, 0.5532],
  14301. [0.6259, 0.4015, 0.7931, 0.2748, 0.3971, 0.2923, 0.5981, 0.5459],
  14302. [0.6373, 0.4096, 0.8234, 0.4940, 0.4075, 0.5483, 0.6636, 0.5948],
  14303. [0.6988, 0.4403, 0.7122, 0.2193, 0.3589, 0.3156, 0.5807, 0.5981],
  14304. [0.6013, 0.3702, 0.8606, 0.5430, 0.4338, 0.4950, 0.5684, 0.5725]],
  14305. device='cuda:0', grad_fn=<AddmmBackward>)
  14306. landmarks are: tensor([[[0.6346, 0.4144, 0.9087, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  14307. [0.0000, 0.0000, 0.8938, 0.2850, 0.4663, 0.3117, 0.7406, 0.5528],
  14308. [0.6229, 0.4066, 0.8512, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  14309. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317],
  14310. [0.6201, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  14311. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  14312. [0.6132, 0.4066, 0.7259, 0.2402, 0.3587, 0.3300, 0.6000, 0.5600],
  14313. [0.6192, 0.4128, 0.8512, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633]]],
  14314. device='cuda:0')
  14315. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  14316. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  14317. loss_train: 0.08935375869623385
  14318. step: 57
  14319. running loss: 0.0015676098016883131
  14320. Train Steps: 57/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14321. torch.Size([8, 8])
  14322. tensor([[0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133],
  14323. [0.6263, 0.4233, 0.7924, 0.4626, 0.3788, 0.2883, 0.5573, 0.6047],
  14324. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  14325. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  14326. [0.6109, 0.4041, 0.6975, 0.3167, 0.3513, 0.3383, 0.5153, 0.5319],
  14327. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  14328. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  14329. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767]],
  14330. device='cuda:0', dtype=torch.float64)
  14331. predictions are: tensor([[0.5957, 0.3914, 0.8953, 0.5222, 0.4134, 0.4844, 0.6682, 0.5365],
  14332. [0.5112, 0.3648, 0.8072, 0.4491, 0.4238, 0.3225, 0.6067, 0.6391],
  14333. [0.4843, 0.3409, 0.7270, 0.2297, 0.4249, 0.2574, 0.5738, 0.5921],
  14334. [0.5497, 0.3573, 0.8853, 0.4831, 0.4449, 0.5554, 0.6904, 0.5349],
  14335. [0.5592, 0.3758, 0.7483, 0.2906, 0.3913, 0.3537, 0.5571, 0.5424],
  14336. [0.5409, 0.3602, 0.9018, 0.5374, 0.4245, 0.4396, 0.6098, 0.5927],
  14337. [0.5360, 0.3595, 0.8909, 0.2613, 0.4279, 0.2904, 0.6356, 0.5659],
  14338. [0.4802, 0.3276, 0.8787, 0.4517, 0.3916, 0.3253, 0.5922, 0.6089]],
  14339. device='cuda:0', grad_fn=<AddmmBackward>)
  14340. landmarks are: tensor([[[0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133],
  14341. [0.6263, 0.4232, 0.7924, 0.4626, 0.3787, 0.2883, 0.5573, 0.6047],
  14342. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  14343. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  14344. [0.6109, 0.4041, 0.6975, 0.3167, 0.3512, 0.3383, 0.5153, 0.5319],
  14345. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  14346. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  14347. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767]]],
  14348. device='cuda:0')
  14349. loss_train_step before backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  14350. loss_train_step after backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  14351. loss_train: 0.09153115292428993
  14352. step: 58
  14353. running loss: 0.001578123326280861
  14354. Train Steps: 58/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14355. torch.Size([8, 8])
  14356. tensor([[0.6112, 0.4029, 0.8638, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  14357. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  14358. [0.6125, 0.3999, 0.8750, 0.4883, 0.4750, 0.4700, 0.5533, 0.5617],
  14359. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  14360. [ nan, nan, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  14361. [ nan, nan, 0.7515, 0.2708, 0.3987, 0.2267, 0.5162, 0.5567],
  14362. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  14363. [0.6151, 0.4125, 0.8738, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483]],
  14364. device='cuda:0', dtype=torch.float64)
  14365. predictions are: tensor([[0.6248, 0.4224, 0.8997, 0.4786, 0.4800, 0.4956, 0.6026, 0.5643],
  14366. [0.7256, 0.4915, 0.8637, 0.5552, 0.3857, 0.5023, 0.7066, 0.5893],
  14367. [0.6146, 0.4201, 0.8895, 0.4822, 0.4621, 0.5014, 0.5588, 0.5500],
  14368. [0.7021, 0.4626, 0.8826, 0.5517, 0.3648, 0.4587, 0.6539, 0.4981],
  14369. [0.0649, 0.0642, 0.9014, 0.3397, 0.4924, 0.2533, 0.7295, 0.5833],
  14370. [0.0735, 0.0730, 0.7442, 0.2641, 0.3869, 0.2435, 0.5210, 0.5732],
  14371. [0.6739, 0.4641, 0.8749, 0.5194, 0.4548, 0.5321, 0.6159, 0.5258],
  14372. [0.6096, 0.4239, 0.8694, 0.4396, 0.3547, 0.3754, 0.5269, 0.5667]],
  14373. device='cuda:0', grad_fn=<AddmmBackward>)
  14374. landmarks are: tensor([[[0.6112, 0.4029, 0.8637, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  14375. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  14376. [0.6125, 0.3999, 0.8750, 0.4883, 0.4750, 0.4700, 0.5533, 0.5617],
  14377. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  14378. [0.0000, 0.0000, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  14379. [0.0000, 0.0000, 0.7515, 0.2708, 0.3988, 0.2267, 0.5163, 0.5567],
  14380. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  14381. [0.6151, 0.4125, 0.8737, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483]]],
  14382. device='cuda:0')
  14383. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  14384. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  14385. loss_train: 0.09251674797269516
  14386. step: 59
  14387. running loss: 0.0015680804741134773
  14388. Train Steps: 59/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14389. torch.Size([8, 8])
  14390. tensor([[0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038],
  14391. [0.6126, 0.4067, 0.8638, 0.5383, 0.4188, 0.4850, 0.5016, 0.5392],
  14392. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  14393. [0.6200, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  14394. [0.6300, 0.4133, 0.8538, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  14395. [0.6260, 0.4153, 0.9000, 0.4533, 0.4025, 0.2633, 0.6223, 0.4967],
  14396. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683],
  14397. [0.6199, 0.4093, 0.7913, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700]],
  14398. device='cuda:0', dtype=torch.float64)
  14399. predictions are: tensor([[0.5696, 0.3680, 0.8549, 0.4859, 0.4327, 0.5179, 0.5572, 0.4864],
  14400. [0.5494, 0.3816, 0.8490, 0.5808, 0.4114, 0.5042, 0.5461, 0.5231],
  14401. [0.5412, 0.3755, 0.8835, 0.4247, 0.3414, 0.3799, 0.6158, 0.5338],
  14402. [0.5183, 0.3418, 0.8871, 0.4990, 0.3579, 0.4325, 0.6339, 0.5204],
  14403. [0.4538, 0.3013, 0.8366, 0.2473, 0.5292, 0.2839, 0.7284, 0.5587],
  14404. [0.5399, 0.3747, 0.9038, 0.4655, 0.3930, 0.3018, 0.6449, 0.5132],
  14405. [0.5060, 0.3411, 0.8525, 0.5727, 0.3688, 0.3681, 0.5648, 0.5554],
  14406. [0.5151, 0.3688, 0.7626, 0.2425, 0.4081, 0.2603, 0.6358, 0.5804]],
  14407. device='cuda:0', grad_fn=<AddmmBackward>)
  14408. landmarks are: tensor([[[0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038],
  14409. [0.6126, 0.4067, 0.8637, 0.5383, 0.4187, 0.4850, 0.5016, 0.5392],
  14410. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  14411. [0.6201, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  14412. [0.6300, 0.4133, 0.8537, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  14413. [0.6260, 0.4153, 0.9000, 0.4533, 0.4025, 0.2633, 0.6223, 0.4967],
  14414. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683],
  14415. [0.6198, 0.4093, 0.7912, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700]]],
  14416. device='cuda:0')
  14417. loss_train_step before backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  14418. loss_train_step after backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  14419. loss_train: 0.09467396742547862
  14420. step: 60
  14421. running loss: 0.0015778994570913103
  14422.  
  14423. Train Steps: 60/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14424. torch.Size([8, 8])
  14425. tensor([[0.6277, 0.4057, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  14426. [0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166],
  14427. [0.6202, 0.4079, 0.8025, 0.2500, 0.3763, 0.3217, 0.6125, 0.5533],
  14428. [0.6311, 0.3998, 0.7975, 0.5767, 0.3838, 0.4850, 0.7327, 0.5343],
  14429. [0.6185, 0.4129, 0.8900, 0.4567, 0.3937, 0.5417, 0.5734, 0.5110],
  14430. [0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500],
  14431. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  14432. [0.6149, 0.4054, 0.6713, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695]],
  14433. device='cuda:0', dtype=torch.float64)
  14434. predictions are: tensor([[0.5558, 0.3568, 0.8592, 0.3219, 0.4439, 0.2026, 0.6285, 0.4938],
  14435. [0.5836, 0.3509, 0.9035, 0.5453, 0.4080, 0.5315, 0.6395, 0.5006],
  14436. [0.6229, 0.4163, 0.8235, 0.3249, 0.3960, 0.3336, 0.6160, 0.5354],
  14437. [0.5874, 0.3863, 0.8409, 0.5990, 0.4015, 0.4852, 0.6877, 0.5168],
  14438. [0.4982, 0.3207, 0.9290, 0.5280, 0.4096, 0.5368, 0.5549, 0.4946],
  14439. [0.5610, 0.3600, 0.8111, 0.3305, 0.3712, 0.4111, 0.6245, 0.5420],
  14440. [0.5940, 0.3892, 0.8602, 0.3784, 0.3739, 0.3779, 0.6145, 0.5378],
  14441. [0.5522, 0.3593, 0.7109, 0.2939, 0.4385, 0.2026, 0.5390, 0.5488]],
  14442. device='cuda:0', grad_fn=<AddmmBackward>)
  14443. landmarks are: tensor([[[0.6277, 0.4056, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  14444. [0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166],
  14445. [0.6202, 0.4079, 0.8025, 0.2500, 0.3762, 0.3217, 0.6125, 0.5533],
  14446. [0.6311, 0.3998, 0.7975, 0.5767, 0.3837, 0.4850, 0.7327, 0.5343],
  14447. [0.6186, 0.4129, 0.8900, 0.4567, 0.3938, 0.5417, 0.5734, 0.5110],
  14448. [0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500],
  14449. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  14450. [0.6149, 0.4054, 0.6712, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695]]],
  14451. device='cuda:0')
  14452. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  14453. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  14454. loss_train: 0.09618941575172357
  14455. step: 61
  14456. running loss: 0.0015768756680610422
  14457. Train Steps: 61/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14458. torch.Size([8, 8])
  14459. tensor([[0.6178, 0.4059, 0.7525, 0.2250, 0.4313, 0.1783, 0.5404, 0.5240],
  14460. [0.6189, 0.4029, 0.8375, 0.5767, 0.4745, 0.4829, 0.5551, 0.5598],
  14461. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5463, 0.5800],
  14462. [0.6255, 0.4017, 0.8688, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  14463. [0.6262, 0.4163, 0.8850, 0.5183, 0.3763, 0.4150, 0.6025, 0.5500],
  14464. [0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5837, 0.5317],
  14465. [ nan, nan, 0.7240, 0.2722, 0.3900, 0.2567, 0.5168, 0.5933],
  14466. [0.6213, 0.4131, 0.8438, 0.3550, 0.3513, 0.4400, 0.5716, 0.5123]],
  14467. device='cuda:0', dtype=torch.float64)
  14468. predictions are: tensor([[0.6293, 0.3999, 0.7578, 0.2511, 0.4518, 0.1490, 0.5578, 0.5300],
  14469. [0.6185, 0.4012, 0.8560, 0.5950, 0.4820, 0.4643, 0.5926, 0.5478],
  14470. [0.5366, 0.3654, 0.7644, 0.3100, 0.3725, 0.2572, 0.5464, 0.5521],
  14471. [0.5998, 0.3699, 0.8737, 0.3380, 0.3826, 0.3640, 0.6650, 0.4941],
  14472. [0.5691, 0.3774, 0.9011, 0.5399, 0.3759, 0.4019, 0.6141, 0.5238],
  14473. [0.6344, 0.4208, 0.8663, 0.3617, 0.3727, 0.3084, 0.6055, 0.5220],
  14474. [0.0288, 0.0073, 0.7313, 0.2803, 0.4008, 0.2289, 0.5316, 0.5480],
  14475. [0.6353, 0.4239, 0.8864, 0.3825, 0.3525, 0.4215, 0.5744, 0.4790]],
  14476. device='cuda:0', grad_fn=<AddmmBackward>)
  14477. landmarks are: tensor([[[0.6178, 0.4059, 0.7525, 0.2250, 0.4313, 0.1783, 0.5404, 0.5240],
  14478. [0.6189, 0.4029, 0.8375, 0.5767, 0.4745, 0.4829, 0.5551, 0.5598],
  14479. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5462, 0.5800],
  14480. [0.6255, 0.4017, 0.8687, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  14481. [0.6262, 0.4163, 0.8850, 0.5183, 0.3762, 0.4150, 0.6025, 0.5500],
  14482. [0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5838, 0.5317],
  14483. [0.0000, 0.0000, 0.7240, 0.2722, 0.3900, 0.2567, 0.5168, 0.5933],
  14484. [0.6213, 0.4131, 0.8438, 0.3550, 0.3512, 0.4400, 0.5716, 0.5123]]],
  14485. device='cuda:0')
  14486. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  14487. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  14488. loss_train: 0.09672151939594187
  14489. step: 62
  14490. running loss: 0.0015600245063861593
  14491. Train Steps: 62/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14492. torch.Size([8, 8])
  14493. tensor([[0.6149, 0.4054, 0.6713, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  14494. [0.6068, 0.3963, 0.8650, 0.4317, 0.4037, 0.5083, 0.5253, 0.4999],
  14495. [0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  14496. [0.6277, 0.4083, 0.8350, 0.2717, 0.4562, 0.1800, 0.5918, 0.4878],
  14497. [0.6117, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103],
  14498. [0.6207, 0.4081, 0.7662, 0.2067, 0.3962, 0.3200, 0.6312, 0.5300],
  14499. [0.6213, 0.4001, 0.7712, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183],
  14500. [0.6125, 0.4076, 0.8488, 0.3883, 0.3700, 0.3683, 0.5026, 0.5505]],
  14501. device='cuda:0', dtype=torch.float64)
  14502. predictions are: tensor([[0.5845, 0.3731, 0.7125, 0.2705, 0.4301, 0.1965, 0.5510, 0.5530],
  14503. [0.5448, 0.3305, 0.8915, 0.4691, 0.3918, 0.5389, 0.5213, 0.4772],
  14504. [0.5425, 0.3449, 0.9177, 0.5062, 0.3573, 0.4249, 0.5814, 0.4759],
  14505. [0.6598, 0.4127, 0.8657, 0.3145, 0.4771, 0.2235, 0.6076, 0.4885],
  14506. [0.5708, 0.3631, 0.7040, 0.2629, 0.3926, 0.2395, 0.5489, 0.5093],
  14507. [0.5866, 0.3729, 0.7770, 0.2500, 0.3852, 0.3128, 0.6369, 0.5207],
  14508. [0.6762, 0.4158, 0.7889, 0.2567, 0.4456, 0.1642, 0.6077, 0.5199],
  14509. [0.5105, 0.3045, 0.8920, 0.4341, 0.3555, 0.3648, 0.4901, 0.5239]],
  14510. device='cuda:0', grad_fn=<AddmmBackward>)
  14511. landmarks are: tensor([[[0.6149, 0.4054, 0.6712, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  14512. [0.6068, 0.3963, 0.8650, 0.4317, 0.4038, 0.5083, 0.5253, 0.4999],
  14513. [0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  14514. [0.6277, 0.4083, 0.8350, 0.2717, 0.4563, 0.1800, 0.5918, 0.4878],
  14515. [0.6116, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103],
  14516. [0.6207, 0.4081, 0.7663, 0.2067, 0.3963, 0.3200, 0.6313, 0.5300],
  14517. [0.6213, 0.4001, 0.7713, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183],
  14518. [0.6125, 0.4076, 0.8487, 0.3883, 0.3700, 0.3683, 0.5026, 0.5505]]],
  14519. device='cuda:0')
  14520. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  14521. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  14522. loss_train: 0.0980854112713132
  14523. step: 63
  14524. running loss: 0.0015569112900208446
  14525. Train Steps: 63/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14526. torch.Size([8, 8])
  14527. tensor([[0.6136, 0.4085, 0.6688, 0.2317, 0.3862, 0.2367, 0.5517, 0.5783],
  14528. [0.6222, 0.3937, 0.8350, 0.5617, 0.4138, 0.4600, 0.5800, 0.5233],
  14529. [0.6272, 0.4045, 0.8538, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  14530. [0.6082, 0.4024, 0.8738, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  14531. [0.6183, 0.4076, 0.8838, 0.4517, 0.3813, 0.4483, 0.5775, 0.5633],
  14532. [0.6202, 0.4066, 0.8746, 0.3376, 0.3717, 0.3090, 0.5842, 0.5165],
  14533. [0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332],
  14534. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320]],
  14535. device='cuda:0', dtype=torch.float64)
  14536. predictions are: tensor([[0.6443, 0.4123, 0.6926, 0.2116, 0.3916, 0.2383, 0.5444, 0.5459],
  14537. [0.5970, 0.3596, 0.8289, 0.5472, 0.4022, 0.4737, 0.5319, 0.5218],
  14538. [0.6078, 0.3907, 0.8406, 0.5781, 0.3599, 0.4388, 0.5628, 0.4617],
  14539. [0.5568, 0.3553, 0.8705, 0.3905, 0.3469, 0.3844, 0.4892, 0.4773],
  14540. [0.7234, 0.4420, 0.8791, 0.4286, 0.3780, 0.4776, 0.5646, 0.5362],
  14541. [0.6287, 0.4008, 0.8808, 0.3129, 0.3524, 0.3089, 0.5696, 0.4900],
  14542. [0.7104, 0.4353, 0.9023, 0.4677, 0.3627, 0.4362, 0.6397, 0.5030],
  14543. [0.6305, 0.4165, 0.8075, 0.1914, 0.4149, 0.2623, 0.6327, 0.5142]],
  14544. device='cuda:0', grad_fn=<AddmmBackward>)
  14545. landmarks are: tensor([[[0.6136, 0.4085, 0.6687, 0.2317, 0.3862, 0.2367, 0.5517, 0.5783],
  14546. [0.6222, 0.3937, 0.8350, 0.5617, 0.4137, 0.4600, 0.5800, 0.5233],
  14547. [0.6271, 0.4045, 0.8537, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  14548. [0.6082, 0.4024, 0.8737, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  14549. [0.6183, 0.4076, 0.8838, 0.4517, 0.3812, 0.4483, 0.5775, 0.5633],
  14550. [0.6202, 0.4066, 0.8746, 0.3376, 0.3717, 0.3090, 0.5842, 0.5165],
  14551. [0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332],
  14552. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320]]],
  14553. device='cuda:0')
  14554. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  14555. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  14556. loss_train: 0.09885636498802342
  14557. step: 64
  14558. running loss: 0.0015446307029378659
  14559.  
  14560. Train Steps: 64/90 Loss: 0.0015 torch.Size([8, 600, 800])
  14561. torch.Size([8, 8])
  14562. tensor([[0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  14563. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  14564. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  14565. [ nan, nan, 0.9088, 0.3783, 0.4562, 0.2617, 0.6741, 0.5575],
  14566. [0.6224, 0.4179, 0.8700, 0.5683, 0.4037, 0.4683, 0.5650, 0.5600],
  14567. [ nan, nan, 0.8525, 0.2217, 0.5413, 0.2367, 0.7367, 0.5482],
  14568. [0.6250, 0.4131, 0.8688, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  14569. [0.6148, 0.4053, 0.8750, 0.4550, 0.4850, 0.5218, 0.5863, 0.5567]],
  14570. device='cuda:0', dtype=torch.float64)
  14571. predictions are: tensor([[0.7994, 0.5367, 0.7996, 0.5162, 0.3479, 0.4840, 0.5989, 0.5018],
  14572. [0.7304, 0.4841, 0.8247, 0.4373, 0.4467, 0.4928, 0.4858, 0.4879],
  14573. [0.6837, 0.4597, 0.7335, 0.2450, 0.3503, 0.2686, 0.5322, 0.5364],
  14574. [0.0720, 0.0430, 0.8790, 0.3516, 0.4122, 0.2363, 0.6346, 0.5353],
  14575. [0.7726, 0.5192, 0.8251, 0.5377, 0.3656, 0.4561, 0.5352, 0.5271],
  14576. [0.1233, 0.0741, 0.8074, 0.1840, 0.4907, 0.2171, 0.6621, 0.5080],
  14577. [0.6856, 0.4354, 0.8384, 0.2577, 0.4017, 0.2215, 0.5753, 0.5097],
  14578. [0.8259, 0.5425, 0.8471, 0.4142, 0.4311, 0.4802, 0.5386, 0.5119]],
  14579. device='cuda:0', grad_fn=<AddmmBackward>)
  14580. landmarks are: tensor([[[0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  14581. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  14582. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  14583. [0.0000, 0.0000, 0.9087, 0.3783, 0.4563, 0.2617, 0.6741, 0.5575],
  14584. [0.6224, 0.4179, 0.8700, 0.5683, 0.4038, 0.4683, 0.5650, 0.5600],
  14585. [0.0000, 0.0000, 0.8525, 0.2217, 0.5412, 0.2367, 0.7367, 0.5482],
  14586. [0.6250, 0.4131, 0.8687, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  14587. [0.6148, 0.4053, 0.8750, 0.4550, 0.4850, 0.5218, 0.5863, 0.5567]]],
  14588. device='cuda:0')
  14589. loss_train_step before backward: tensor(0.0043, device='cuda:0', grad_fn=<MseLossBackward>)
  14590. loss_train_step after backward: tensor(0.0043, device='cuda:0', grad_fn=<MseLossBackward>)
  14591. loss_train: 0.10314330333494581
  14592. step: 65
  14593. running loss: 0.0015868200513068586
  14594. Train Steps: 65/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14595. torch.Size([8, 8])
  14596. tensor([[0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  14597. [0.6350, 0.4043, 0.8738, 0.5650, 0.3850, 0.4750, 0.6401, 0.4950],
  14598. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  14599. [0.6276, 0.4120, 0.8738, 0.3133, 0.4225, 0.2217, 0.6203, 0.4892],
  14600. [0.6182, 0.4099, 0.7812, 0.3000, 0.3937, 0.2367, 0.5325, 0.5750],
  14601. [0.6224, 0.4061, 0.8988, 0.4300, 0.3838, 0.4750, 0.6112, 0.5483],
  14602. [0.6147, 0.4026, 0.6600, 0.2467, 0.4088, 0.2150, 0.5489, 0.5773],
  14603. [0.6122, 0.4006, 0.8850, 0.4217, 0.4088, 0.5517, 0.6063, 0.5517]],
  14604. device='cuda:0', dtype=torch.float64)
  14605. predictions are: tensor([[0.6405, 0.4135, 0.8524, 0.4461, 0.4034, 0.5525, 0.5545, 0.5209],
  14606. [0.6956, 0.4378, 0.8569, 0.5324, 0.3790, 0.4938, 0.6071, 0.5266],
  14607. [0.5868, 0.3892, 0.8500, 0.4279, 0.3564, 0.3593, 0.5179, 0.5958],
  14608. [0.6459, 0.4150, 0.8801, 0.3083, 0.4352, 0.2120, 0.6126, 0.5217],
  14609. [0.6575, 0.4345, 0.7819, 0.2809, 0.3977, 0.2372, 0.5261, 0.5703],
  14610. [0.6123, 0.4102, 0.8703, 0.3776, 0.3593, 0.4997, 0.5979, 0.5484],
  14611. [0.6556, 0.4435, 0.7015, 0.2308, 0.4147, 0.2139, 0.5435, 0.6041],
  14612. [0.5834, 0.3792, 0.8637, 0.3678, 0.4143, 0.5444, 0.5634, 0.5441]],
  14613. device='cuda:0', grad_fn=<AddmmBackward>)
  14614. landmarks are: tensor([[[0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  14615. [0.6350, 0.4043, 0.8737, 0.5650, 0.3850, 0.4750, 0.6401, 0.4950],
  14616. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  14617. [0.6276, 0.4120, 0.8737, 0.3133, 0.4225, 0.2217, 0.6203, 0.4892],
  14618. [0.6182, 0.4099, 0.7812, 0.3000, 0.3938, 0.2367, 0.5325, 0.5750],
  14619. [0.6224, 0.4061, 0.8988, 0.4300, 0.3837, 0.4750, 0.6112, 0.5483],
  14620. [0.6147, 0.4026, 0.6600, 0.2467, 0.4087, 0.2150, 0.5489, 0.5773],
  14621. [0.6122, 0.4006, 0.8850, 0.4217, 0.4087, 0.5517, 0.6062, 0.5517]]],
  14622. device='cuda:0')
  14623. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  14624. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  14625. loss_train: 0.1037804797233548
  14626. step: 66
  14627. running loss: 0.0015724315109599213
  14628. Train Steps: 66/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14629. torch.Size([8, 8])
  14630. tensor([[0.6276, 0.4095, 0.8237, 0.2250, 0.4662, 0.1783, 0.6171, 0.4869],
  14631. [0.6255, 0.4017, 0.8688, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  14632. [0.6254, 0.3993, 0.8988, 0.4767, 0.3987, 0.5517, 0.6955, 0.5285],
  14633. [0.6111, 0.3995, 0.8788, 0.4567, 0.3813, 0.4833, 0.5450, 0.5700],
  14634. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  14635. [0.6148, 0.4053, 0.8750, 0.4550, 0.4850, 0.5218, 0.5863, 0.5567],
  14636. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  14637. [ nan, nan, 0.7625, 0.2433, 0.3713, 0.2867, 0.5235, 0.5220]],
  14638. device='cuda:0', dtype=torch.float64)
  14639. predictions are: tensor([[0.6785, 0.4481, 0.8139, 0.2020, 0.4749, 0.1850, 0.6138, 0.5308],
  14640. [0.6331, 0.4025, 0.8234, 0.2862, 0.3667, 0.3415, 0.6191, 0.5370],
  14641. [0.6745, 0.4569, 0.8635, 0.4377, 0.4150, 0.5758, 0.6779, 0.5648],
  14642. [0.7044, 0.4675, 0.8349, 0.4179, 0.3905, 0.4767, 0.5015, 0.5859],
  14643. [0.6103, 0.4076, 0.8960, 0.4278, 0.4193, 0.3025, 0.6647, 0.5860],
  14644. [0.7442, 0.4936, 0.8570, 0.4359, 0.4789, 0.4910, 0.5747, 0.5813],
  14645. [0.7321, 0.5025, 0.8436, 0.4162, 0.3645, 0.3410, 0.6136, 0.5306],
  14646. [0.0770, 0.0437, 0.7523, 0.2256, 0.3834, 0.2551, 0.5500, 0.5512]],
  14647. device='cuda:0', grad_fn=<AddmmBackward>)
  14648. landmarks are: tensor([[[0.6276, 0.4095, 0.8238, 0.2250, 0.4663, 0.1783, 0.6171, 0.4869],
  14649. [0.6255, 0.4017, 0.8687, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  14650. [0.6254, 0.3993, 0.8988, 0.4767, 0.3988, 0.5517, 0.6955, 0.5285],
  14651. [0.6111, 0.3995, 0.8788, 0.4567, 0.3812, 0.4833, 0.5450, 0.5700],
  14652. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  14653. [0.6148, 0.4053, 0.8750, 0.4550, 0.4850, 0.5218, 0.5863, 0.5567],
  14654. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  14655. [0.0000, 0.0000, 0.7625, 0.2433, 0.3713, 0.2867, 0.5235, 0.5220]]],
  14656. device='cuda:0')
  14657. loss_train_step before backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  14658. loss_train_step after backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  14659. loss_train: 0.10556276931311004
  14660. step: 67
  14661. running loss: 0.0015755637210911947
  14662. Train Steps: 67/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14663. torch.Size([8, 8])
  14664. tensor([[0.6068, 0.3963, 0.8650, 0.4317, 0.4037, 0.5083, 0.5253, 0.4999],
  14665. [0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  14666. [0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  14667. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  14668. [0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5837, 0.5317],
  14669. [0.6216, 0.4167, 0.8588, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667],
  14670. [0.6273, 0.4105, 0.8988, 0.4517, 0.3912, 0.2550, 0.5894, 0.4811],
  14671. [0.6128, 0.4022, 0.8738, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064]],
  14672. device='cuda:0', dtype=torch.float64)
  14673. predictions are: tensor([[0.5519, 0.3711, 0.8373, 0.3983, 0.3957, 0.4941, 0.5430, 0.5392],
  14674. [0.5763, 0.3926, 0.8314, 0.3637, 0.3484, 0.4117, 0.5856, 0.5564],
  14675. [0.5165, 0.3468, 0.8993, 0.3403, 0.4354, 0.3505, 0.7177, 0.5684],
  14676. [0.6442, 0.4390, 0.8608, 0.4409, 0.3806, 0.4275, 0.5905, 0.5781],
  14677. [0.6688, 0.4501, 0.8287, 0.3181, 0.3599, 0.2964, 0.6153, 0.5782],
  14678. [0.6583, 0.4522, 0.8290, 0.5357, 0.4189, 0.5248, 0.6321, 0.5842],
  14679. [0.6497, 0.4241, 0.8771, 0.4061, 0.4136, 0.2432, 0.6155, 0.5340],
  14680. [0.6057, 0.4289, 0.8389, 0.4843, 0.4893, 0.4952, 0.5470, 0.5456]],
  14681. device='cuda:0', grad_fn=<AddmmBackward>)
  14682. landmarks are: tensor([[[0.6068, 0.3963, 0.8650, 0.4317, 0.4038, 0.5083, 0.5253, 0.4999],
  14683. [0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  14684. [0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  14685. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  14686. [0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5838, 0.5317],
  14687. [0.6216, 0.4167, 0.8587, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667],
  14688. [0.6273, 0.4105, 0.8988, 0.4517, 0.3913, 0.2550, 0.5894, 0.4811],
  14689. [0.6128, 0.4022, 0.8737, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064]]],
  14690. device='cuda:0')
  14691. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  14692. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  14693. loss_train: 0.10667031232151203
  14694. step: 68
  14695. running loss: 0.0015686810635516474
  14696.  
  14697. Train Steps: 68/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14698. torch.Size([8, 8])
  14699. tensor([[0.6185, 0.4129, 0.8900, 0.4567, 0.3937, 0.5417, 0.5734, 0.5110],
  14700. [0.6098, 0.3991, 0.8638, 0.4717, 0.4263, 0.4967, 0.5212, 0.5650],
  14701. [0.6332, 0.4118, 0.9238, 0.4267, 0.4012, 0.4733, 0.7525, 0.5436],
  14702. [0.6182, 0.4058, 0.8738, 0.4350, 0.3563, 0.3400, 0.5290, 0.5822],
  14703. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  14704. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461],
  14705. [0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  14706. [0.6310, 0.4017, 0.8563, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006]],
  14707. device='cuda:0', dtype=torch.float64)
  14708. predictions are: tensor([[0.6513, 0.4373, 0.8968, 0.4452, 0.4122, 0.5273, 0.6020, 0.5217],
  14709. [0.6055, 0.3957, 0.8620, 0.4681, 0.4406, 0.4793, 0.5524, 0.5462],
  14710. [0.5849, 0.3926, 0.9103, 0.4227, 0.4073, 0.4678, 0.7246, 0.5526],
  14711. [0.6474, 0.4221, 0.8637, 0.4193, 0.3624, 0.3218, 0.5677, 0.5930],
  14712. [0.5914, 0.3841, 0.8379, 0.3149, 0.4315, 0.1825, 0.6207, 0.5486],
  14713. [0.5509, 0.3660, 0.8650, 0.5164, 0.4588, 0.5093, 0.5628, 0.5499],
  14714. [0.5631, 0.3833, 0.8470, 0.3880, 0.3613, 0.4727, 0.5943, 0.5614],
  14715. [0.6208, 0.4106, 0.8653, 0.5627, 0.3887, 0.4721, 0.6646, 0.5145]],
  14716. device='cuda:0', grad_fn=<AddmmBackward>)
  14717. landmarks are: tensor([[[0.6186, 0.4129, 0.8900, 0.4567, 0.3938, 0.5417, 0.5734, 0.5110],
  14718. [0.6098, 0.3991, 0.8637, 0.4717, 0.4263, 0.4967, 0.5213, 0.5650],
  14719. [0.6332, 0.4118, 0.9237, 0.4267, 0.4013, 0.4733, 0.7525, 0.5436],
  14720. [0.6182, 0.4058, 0.8737, 0.4350, 0.3562, 0.3400, 0.5290, 0.5822],
  14721. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  14722. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461],
  14723. [0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  14724. [0.6310, 0.4017, 0.8562, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006]]],
  14725. device='cuda:0')
  14726. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  14727. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  14728. loss_train: 0.10714027981157415
  14729. step: 69
  14730. running loss: 0.001552757678428611
  14731. Train Steps: 69/90 Loss: 0.0016 torch.Size([8, 600, 800])
  14732. torch.Size([8, 8])
  14733. tensor([[0.6197, 0.4118, 0.8688, 0.5517, 0.4037, 0.5233, 0.5875, 0.5600],
  14734. [0.6168, 0.4029, 0.8523, 0.3417, 0.3588, 0.5000, 0.6125, 0.5400],
  14735. [0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6262, 0.5367],
  14736. [0.6261, 0.4066, 0.8325, 0.2150, 0.4763, 0.2667, 0.7002, 0.5633],
  14737. [ nan, nan, 0.7425, 0.2117, 0.3937, 0.2433, 0.5438, 0.5567],
  14738. [0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  14739. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  14740. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167]],
  14741. device='cuda:0', dtype=torch.float64)
  14742. predictions are: tensor([[0.6797, 0.4425, 0.8824, 0.5833, 0.4246, 0.5367, 0.5965, 0.5615],
  14743. [0.5917, 0.3821, 0.8896, 0.3713, 0.3764, 0.4707, 0.6431, 0.5447],
  14744. [0.6304, 0.4132, 0.9070, 0.4863, 0.4543, 0.5783, 0.6085, 0.5383],
  14745. [0.6103, 0.3874, 0.8507, 0.2457, 0.4849, 0.2317, 0.7062, 0.5364],
  14746. [0.1297, 0.0855, 0.7537, 0.2486, 0.4253, 0.2378, 0.5901, 0.5524],
  14747. [0.6390, 0.4320, 0.9092, 0.5563, 0.3869, 0.4732, 0.5894, 0.5641],
  14748. [0.6987, 0.4476, 0.8577, 0.6106, 0.3919, 0.5049, 0.6735, 0.5354],
  14749. [0.6582, 0.4399, 0.8386, 0.3634, 0.4016, 0.2942, 0.5853, 0.5336]],
  14750. device='cuda:0', grad_fn=<AddmmBackward>)
  14751. landmarks are: tensor([[[0.6197, 0.4118, 0.8687, 0.5517, 0.4038, 0.5233, 0.5875, 0.5600],
  14752. [0.6168, 0.4029, 0.8523, 0.3417, 0.3587, 0.5000, 0.6125, 0.5400],
  14753. [0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6263, 0.5367],
  14754. [0.6261, 0.4066, 0.8325, 0.2150, 0.4762, 0.2667, 0.7002, 0.5633],
  14755. [0.0000, 0.0000, 0.7425, 0.2117, 0.3938, 0.2433, 0.5437, 0.5567],
  14756. [0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  14757. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  14758. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167]]],
  14759. device='cuda:0')
  14760. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  14761. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  14762. loss_train: 0.10814244553330354
  14763. step: 70
  14764. running loss: 0.0015448920790471935
  14765. Train Steps: 70/90 Loss: 0.0015 torch.Size([8, 600, 800])
  14766. torch.Size([8, 8])
  14767. tensor([[0.6068, 0.3963, 0.8650, 0.4317, 0.4037, 0.5083, 0.5253, 0.4999],
  14768. [0.6098, 0.3991, 0.8638, 0.4717, 0.4263, 0.4967, 0.5212, 0.5650],
  14769. [0.6198, 0.3997, 0.8582, 0.5361, 0.4117, 0.5016, 0.5942, 0.5134],
  14770. [0.6300, 0.4133, 0.8538, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  14771. [0.6257, 0.4034, 0.8287, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  14772. [0.6276, 0.4095, 0.8237, 0.2250, 0.4662, 0.1783, 0.6171, 0.4869],
  14773. [0.6200, 0.3961, 0.8461, 0.5497, 0.4142, 0.4577, 0.5892, 0.5402],
  14774. [0.6193, 0.4050, 0.7313, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656]],
  14775. device='cuda:0', dtype=torch.float64)
  14776. predictions are: tensor([[0.5551, 0.3648, 0.8834, 0.4773, 0.3928, 0.5299, 0.5418, 0.5109],
  14777. [0.5776, 0.3809, 0.8719, 0.5290, 0.4246, 0.5201, 0.5414, 0.5491],
  14778. [0.5532, 0.3692, 0.8817, 0.5963, 0.4139, 0.5325, 0.6002, 0.5089],
  14779. [0.5847, 0.3793, 0.8621, 0.2714, 0.5321, 0.2797, 0.7289, 0.5481],
  14780. [0.5375, 0.3378, 0.8474, 0.2894, 0.3923, 0.2840, 0.6445, 0.4855],
  14781. [0.5882, 0.3816, 0.8420, 0.2702, 0.4611, 0.2218, 0.6174, 0.4954],
  14782. [0.5419, 0.3502, 0.8819, 0.5863, 0.3925, 0.4763, 0.6229, 0.5348],
  14783. [0.5298, 0.3446, 0.7433, 0.2817, 0.4003, 0.2351, 0.5706, 0.5667]],
  14784. device='cuda:0', grad_fn=<AddmmBackward>)
  14785. landmarks are: tensor([[[0.6068, 0.3963, 0.8650, 0.4317, 0.4038, 0.5083, 0.5253, 0.4999],
  14786. [0.6098, 0.3991, 0.8637, 0.4717, 0.4263, 0.4967, 0.5213, 0.5650],
  14787. [0.6198, 0.3997, 0.8582, 0.5361, 0.4117, 0.5016, 0.5942, 0.5134],
  14788. [0.6300, 0.4133, 0.8537, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  14789. [0.6257, 0.4034, 0.8288, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  14790. [0.6276, 0.4095, 0.8238, 0.2250, 0.4663, 0.1783, 0.6171, 0.4869],
  14791. [0.6200, 0.3961, 0.8461, 0.5497, 0.4142, 0.4577, 0.5892, 0.5402],
  14792. [0.6193, 0.4050, 0.7312, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656]]],
  14793. device='cuda:0')
  14794. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  14795. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  14796. loss_train: 0.10943879742990248
  14797. step: 71
  14798. running loss: 0.001541391513097218
  14799. Train Steps: 71/90 Loss: 0.0015 torch.Size([8, 600, 800])
  14800. torch.Size([8, 8])
  14801. tensor([[ nan, nan, 0.7097, 0.2346, 0.4250, 0.1850, 0.5175, 0.5583],
  14802. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  14803. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  14804. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  14805. [ nan, nan, 0.7425, 0.2117, 0.3937, 0.2433, 0.5438, 0.5567],
  14806. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  14807. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  14808. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383]],
  14809. device='cuda:0', dtype=torch.float64)
  14810. predictions are: tensor([[0.1746, 0.1223, 0.7293, 0.2208, 0.4302, 0.2071, 0.5374, 0.5330],
  14811. [0.6888, 0.4421, 0.9003, 0.5702, 0.3797, 0.4544, 0.6007, 0.5144],
  14812. [0.6518, 0.4239, 0.9269, 0.5022, 0.4492, 0.5962, 0.6580, 0.5112],
  14813. [0.6560, 0.4232, 0.8784, 0.5940, 0.3866, 0.4556, 0.6057, 0.5096],
  14814. [0.0818, 0.0514, 0.7669, 0.2419, 0.4034, 0.2672, 0.5743, 0.5333],
  14815. [0.6324, 0.4228, 0.7939, 0.2588, 0.4446, 0.2049, 0.5882, 0.5233],
  14816. [0.7643, 0.4976, 0.8812, 0.6012, 0.3931, 0.5022, 0.5749, 0.5795],
  14817. [0.5979, 0.3965, 0.9247, 0.4614, 0.4477, 0.5889, 0.6334, 0.5222]],
  14818. device='cuda:0', grad_fn=<AddmmBackward>)
  14819. landmarks are: tensor([[[0.0000, 0.0000, 0.7097, 0.2346, 0.4250, 0.1850, 0.5175, 0.5583],
  14820. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  14821. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  14822. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  14823. [0.0000, 0.0000, 0.7425, 0.2117, 0.3938, 0.2433, 0.5437, 0.5567],
  14824. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  14825. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  14826. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383]]],
  14827. device='cuda:0')
  14828. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  14829. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  14830. loss_train: 0.11130989887169562
  14831. step: 72
  14832. running loss: 0.0015459708176624393
  14833.  
  14834. Train Steps: 72/90 Loss: 0.0015 torch.Size([8, 600, 800])
  14835. torch.Size([8, 8])
  14836. tensor([[0.6114, 0.4018, 0.7213, 0.1967, 0.3763, 0.2700, 0.5875, 0.5533],
  14837. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5787, 0.5117],
  14838. [0.6057, 0.4011, 0.8750, 0.4267, 0.4400, 0.5800, 0.5845, 0.5585],
  14839. [0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297],
  14840. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  14841. [0.6185, 0.4079, 0.8838, 0.4617, 0.4838, 0.5650, 0.6175, 0.5850],
  14842. [0.6250, 0.4236, 0.8638, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  14843. [0.6078, 0.4033, 0.8019, 0.3055, 0.3450, 0.4200, 0.6025, 0.5550]],
  14844. device='cuda:0', dtype=torch.float64)
  14845. predictions are: tensor([[0.5244, 0.3230, 0.7322, 0.2232, 0.3741, 0.2648, 0.5930, 0.5196],
  14846. [0.5423, 0.3285, 0.7587, 0.2411, 0.4187, 0.2186, 0.5792, 0.4869],
  14847. [0.5157, 0.3403, 0.8917, 0.4692, 0.4561, 0.5478, 0.5834, 0.5112],
  14848. [0.5739, 0.3679, 0.8928, 0.5662, 0.3678, 0.4318, 0.6299, 0.5162],
  14849. [0.5439, 0.3567, 0.7502, 0.2889, 0.3702, 0.2684, 0.5522, 0.4780],
  14850. [0.5751, 0.3509, 0.9098, 0.4848, 0.4810, 0.5481, 0.6122, 0.5371],
  14851. [0.5244, 0.3461, 0.8728, 0.4333, 0.3907, 0.3227, 0.5516, 0.5536],
  14852. [0.5542, 0.3530, 0.8331, 0.3390, 0.3716, 0.4300, 0.6096, 0.5287]],
  14853. device='cuda:0', grad_fn=<AddmmBackward>)
  14854. landmarks are: tensor([[[0.6114, 0.4018, 0.7212, 0.1967, 0.3762, 0.2700, 0.5875, 0.5533],
  14855. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5788, 0.5117],
  14856. [0.6057, 0.4011, 0.8750, 0.4267, 0.4400, 0.5800, 0.5845, 0.5585],
  14857. [0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297],
  14858. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  14859. [0.6184, 0.4079, 0.8838, 0.4617, 0.4837, 0.5650, 0.6175, 0.5850],
  14860. [0.6250, 0.4236, 0.8637, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  14861. [0.6078, 0.4033, 0.8019, 0.3055, 0.3450, 0.4200, 0.6025, 0.5550]]],
  14862. device='cuda:0')
  14863. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  14864. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  14865. loss_train: 0.11289354818291031
  14866. step: 73
  14867. running loss: 0.0015464869614097302
  14868. Train Steps: 73/90 Loss: 0.0015 torch.Size([8, 600, 800])
  14869. torch.Size([8, 8])
  14870. tensor([[0.6099, 0.4030, 0.8638, 0.5117, 0.4983, 0.4965, 0.5086, 0.5388],
  14871. [0.6300, 0.4133, 0.8538, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  14872. [0.6178, 0.4012, 0.8900, 0.4495, 0.3891, 0.5617, 0.5972, 0.4879],
  14873. [0.6198, 0.4101, 0.8838, 0.5283, 0.3763, 0.5267, 0.5913, 0.5567],
  14874. [0.6325, 0.4165, 0.9000, 0.4617, 0.3813, 0.4900, 0.7485, 0.5447],
  14875. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  14876. [0.6179, 0.4008, 0.7505, 0.2678, 0.4368, 0.1891, 0.5831, 0.5263],
  14877. [0.6275, 0.4048, 0.8488, 0.2883, 0.4463, 0.2033, 0.6321, 0.5155]],
  14878. device='cuda:0', dtype=torch.float64)
  14879. predictions are: tensor([[0.5338, 0.3522, 0.8780, 0.5271, 0.4792, 0.5190, 0.4882, 0.5306],
  14880. [0.5412, 0.3423, 0.8516, 0.2336, 0.5212, 0.2795, 0.6972, 0.5319],
  14881. [0.5630, 0.3671, 0.8744, 0.4487, 0.3698, 0.5784, 0.5697, 0.4878],
  14882. [0.5473, 0.3473, 0.8602, 0.5516, 0.3654, 0.5584, 0.5371, 0.5472],
  14883. [0.5457, 0.3635, 0.9200, 0.4802, 0.3689, 0.5329, 0.6613, 0.5378],
  14884. [0.5589, 0.3510, 0.7413, 0.3505, 0.4733, 0.2244, 0.5237, 0.6045],
  14885. [0.5656, 0.3641, 0.7410, 0.2551, 0.4121, 0.2023, 0.5469, 0.5126],
  14886. [0.5246, 0.3155, 0.8481, 0.2755, 0.4280, 0.2449, 0.6312, 0.5021]],
  14887. device='cuda:0', grad_fn=<AddmmBackward>)
  14888. landmarks are: tensor([[[0.6098, 0.4030, 0.8637, 0.5117, 0.4983, 0.4965, 0.5086, 0.5388],
  14889. [0.6300, 0.4133, 0.8537, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  14890. [0.6178, 0.4012, 0.8900, 0.4495, 0.3891, 0.5617, 0.5972, 0.4879],
  14891. [0.6198, 0.4101, 0.8838, 0.5283, 0.3762, 0.5267, 0.5913, 0.5567],
  14892. [0.6325, 0.4165, 0.9000, 0.4617, 0.3812, 0.4900, 0.7485, 0.5447],
  14893. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  14894. [0.6179, 0.4008, 0.7505, 0.2678, 0.4368, 0.1891, 0.5831, 0.5263],
  14895. [0.6275, 0.4048, 0.8487, 0.2883, 0.4462, 0.2033, 0.6321, 0.5155]]],
  14896. device='cuda:0')
  14897. loss_train_step before backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  14898. loss_train_step after backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  14899. loss_train: 0.11465108217089437
  14900. step: 74
  14901. running loss: 0.0015493389482553292
  14902. Train Steps: 74/90 Loss: 0.0015 torch.Size([8, 600, 800])
  14903. torch.Size([8, 8])
  14904. tensor([[0.6275, 0.4111, 0.8463, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  14905. [0.6329, 0.4055, 0.9050, 0.4783, 0.3613, 0.3917, 0.6464, 0.5019],
  14906. [0.6250, 0.4054, 0.8770, 0.4723, 0.4662, 0.5367, 0.6162, 0.5433],
  14907. [0.6164, 0.4076, 0.8838, 0.4117, 0.3713, 0.5550, 0.6238, 0.5350],
  14908. [0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  14909. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  14910. [0.6200, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  14911. [0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364]],
  14912. device='cuda:0', dtype=torch.float64)
  14913. predictions are: tensor([[0.5727, 0.3657, 0.8181, 0.2616, 0.4693, 0.2311, 0.5641, 0.5257],
  14914. [0.6030, 0.3700, 0.8953, 0.4802, 0.3780, 0.4110, 0.6093, 0.5120],
  14915. [0.5460, 0.3730, 0.8862, 0.4525, 0.4609, 0.5342, 0.5684, 0.5456],
  14916. [0.6147, 0.4029, 0.8721, 0.3928, 0.3868, 0.5654, 0.5933, 0.5465],
  14917. [0.5614, 0.3649, 0.8514, 0.5074, 0.3873, 0.5299, 0.6509, 0.5383],
  14918. [0.6059, 0.3993, 0.8745, 0.4442, 0.3633, 0.3604, 0.5813, 0.5028],
  14919. [0.5465, 0.3564, 0.8698, 0.4750, 0.3883, 0.4464, 0.5409, 0.5462],
  14920. [0.5784, 0.3736, 0.7716, 0.2147, 0.4488, 0.2392, 0.6080, 0.5363]],
  14921. device='cuda:0', grad_fn=<AddmmBackward>)
  14922. landmarks are: tensor([[[0.6275, 0.4111, 0.8462, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  14923. [0.6329, 0.4055, 0.9050, 0.4783, 0.3613, 0.3917, 0.6464, 0.5019],
  14924. [0.6250, 0.4054, 0.8770, 0.4723, 0.4663, 0.5367, 0.6162, 0.5433],
  14925. [0.6164, 0.4076, 0.8838, 0.4117, 0.3713, 0.5550, 0.6237, 0.5350],
  14926. [0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  14927. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  14928. [0.6201, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  14929. [0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364]]],
  14930. device='cuda:0')
  14931. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  14932. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  14933. loss_train: 0.1157409083971288
  14934. step: 75
  14935. running loss: 0.0015432121119617174
  14936. Train Steps: 75/90 Loss: 0.0015 torch.Size([8, 600, 800])
  14937. torch.Size([8, 8])
  14938. tensor([[0.6111, 0.3995, 0.8788, 0.4567, 0.3813, 0.4833, 0.5450, 0.5700],
  14939. [ nan, nan, 0.7625, 0.2433, 0.3713, 0.2867, 0.5235, 0.5220],
  14940. [0.6166, 0.4008, 0.8563, 0.5667, 0.4388, 0.4933, 0.5575, 0.5567],
  14941. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  14942. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  14943. [0.6198, 0.4130, 0.8762, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  14944. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  14945. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116]],
  14946. device='cuda:0', dtype=torch.float64)
  14947. predictions are: tensor([[ 0.7047, 0.4521, 0.8542, 0.4457, 0.3852, 0.4534, 0.5695, 0.5806],
  14948. [-0.0310, -0.0268, 0.7352, 0.2170, 0.3854, 0.2452, 0.5475, 0.5456],
  14949. [ 0.6074, 0.3916, 0.8318, 0.5185, 0.4381, 0.4620, 0.5896, 0.5820],
  14950. [ 0.6495, 0.4064, 0.7906, 0.2347, 0.4436, 0.2241, 0.6480, 0.5400],
  14951. [ 0.6545, 0.4307, 0.8227, 0.5311, 0.4055, 0.4264, 0.5485, 0.5173],
  14952. [ 0.6268, 0.4137, 0.8465, 0.3939, 0.3636, 0.4709, 0.5830, 0.5404],
  14953. [ 0.6327, 0.4152, 0.8263, 0.2706, 0.4183, 0.2436, 0.6445, 0.5048],
  14954. [ 0.6758, 0.4513, 0.8386, 0.3769, 0.4227, 0.5609, 0.6045, 0.5502]],
  14955. device='cuda:0', grad_fn=<AddmmBackward>)
  14956. landmarks are: tensor([[[0.6111, 0.3995, 0.8788, 0.4567, 0.3812, 0.4833, 0.5450, 0.5700],
  14957. [0.0000, 0.0000, 0.7625, 0.2433, 0.3713, 0.2867, 0.5235, 0.5220],
  14958. [0.6166, 0.4008, 0.8562, 0.5667, 0.4387, 0.4933, 0.5575, 0.5567],
  14959. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  14960. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  14961. [0.6198, 0.4130, 0.8763, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  14962. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  14963. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116]]],
  14964. device='cuda:0')
  14965. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  14966. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  14967. loss_train: 0.11660776424105279
  14968. step: 76
  14969. running loss: 0.0015343126873822736
  14970.  
  14971. Train Steps: 76/90 Loss: 0.0015 torch.Size([8, 600, 800])
  14972. torch.Size([8, 8])
  14973. tensor([[0.6148, 0.4076, 0.8666, 0.4820, 0.4138, 0.5067, 0.5250, 0.5767],
  14974. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  14975. [0.6361, 0.4076, 0.8862, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297],
  14976. [0.6265, 0.4251, 0.7113, 0.3550, 0.4375, 0.2117, 0.5587, 0.6118],
  14977. [0.6147, 0.4026, 0.6600, 0.2467, 0.4088, 0.2150, 0.5489, 0.5773],
  14978. [0.6214, 0.4112, 0.7838, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  14979. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  14980. [0.6272, 0.4071, 0.8738, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742]],
  14981. device='cuda:0', dtype=torch.float64)
  14982. predictions are: tensor([[0.5444, 0.3557, 0.8830, 0.4480, 0.4301, 0.5141, 0.5388, 0.5729],
  14983. [0.6076, 0.3915, 0.9206, 0.4102, 0.4174, 0.3018, 0.6659, 0.5342],
  14984. [0.6002, 0.3981, 0.8804, 0.5211, 0.3667, 0.4781, 0.6635, 0.5433],
  14985. [0.6297, 0.4335, 0.7276, 0.2955, 0.4258, 0.2230, 0.5619, 0.6147],
  14986. [0.5596, 0.3659, 0.6780, 0.1973, 0.4080, 0.2274, 0.5393, 0.5786],
  14987. [0.5413, 0.3425, 0.7714, 0.1700, 0.3657, 0.2916, 0.5894, 0.5097],
  14988. [0.6796, 0.4358, 0.8973, 0.3880, 0.3720, 0.3797, 0.6493, 0.5082],
  14989. [0.6840, 0.4333, 0.8701, 0.5319, 0.3848, 0.3931, 0.6216, 0.4820]],
  14990. device='cuda:0', grad_fn=<AddmmBackward>)
  14991. landmarks are: tensor([[[0.6148, 0.4076, 0.8666, 0.4820, 0.4137, 0.5067, 0.5250, 0.5767],
  14992. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  14993. [0.6361, 0.4076, 0.8863, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297],
  14994. [0.6265, 0.4251, 0.7113, 0.3550, 0.4375, 0.2117, 0.5587, 0.6118],
  14995. [0.6147, 0.4026, 0.6600, 0.2467, 0.4087, 0.2150, 0.5489, 0.5773],
  14996. [0.6214, 0.4112, 0.7837, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  14997. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  14998. [0.6272, 0.4071, 0.8737, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742]]],
  14999. device='cuda:0')
  15000. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  15001. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  15002. loss_train: 0.11740422362345271
  15003. step: 77
  15004. running loss: 0.0015247301769279573
  15005. Train Steps: 77/90 Loss: 0.0015 torch.Size([8, 600, 800])
  15006. torch.Size([8, 8])
  15007. tensor([[0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  15008. [0.6200, 0.3993, 0.8519, 0.4923, 0.3962, 0.4717, 0.6013, 0.5433],
  15009. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  15010. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167],
  15011. [0.6140, 0.4070, 0.8700, 0.5000, 0.4612, 0.4900, 0.5260, 0.5852],
  15012. [ nan, nan, 0.7850, 0.2700, 0.4288, 0.1717, 0.5199, 0.4999],
  15013. [0.6143, 0.4034, 0.8800, 0.4833, 0.4512, 0.5367, 0.5289, 0.5097],
  15014. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5963, 0.6217]],
  15015. device='cuda:0', dtype=torch.float64)
  15016. predictions are: tensor([[0.6762, 0.4645, 0.8378, 0.5068, 0.3814, 0.3885, 0.6824, 0.5250],
  15017. [0.6535, 0.4385, 0.8755, 0.4775, 0.3767, 0.4595, 0.6373, 0.5371],
  15018. [0.6861, 0.4507, 0.8526, 0.5125, 0.3286, 0.3462, 0.5766, 0.5065],
  15019. [0.6160, 0.4223, 0.7629, 0.2434, 0.3948, 0.2627, 0.6210, 0.6043],
  15020. [0.6554, 0.4457, 0.8855, 0.4622, 0.4515, 0.4659, 0.5474, 0.5615],
  15021. [0.1650, 0.1163, 0.7443, 0.1991, 0.4005, 0.1677, 0.5614, 0.4940],
  15022. [0.6359, 0.4271, 0.8692, 0.4393, 0.4413, 0.5079, 0.5420, 0.4931],
  15023. [0.7311, 0.4831, 0.6914, 0.1935, 0.4226, 0.2248, 0.6182, 0.5892]],
  15024. device='cuda:0', grad_fn=<AddmmBackward>)
  15025. landmarks are: tensor([[[0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  15026. [0.6200, 0.3993, 0.8519, 0.4923, 0.3963, 0.4717, 0.6012, 0.5433],
  15027. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  15028. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167],
  15029. [0.6140, 0.4070, 0.8700, 0.5000, 0.4613, 0.4900, 0.5260, 0.5852],
  15030. [0.0000, 0.0000, 0.7850, 0.2700, 0.4288, 0.1717, 0.5199, 0.4999],
  15031. [0.6143, 0.4034, 0.8800, 0.4833, 0.4512, 0.5367, 0.5289, 0.5097],
  15032. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5962, 0.6217]]],
  15033. device='cuda:0')
  15034. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  15035. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  15036. loss_train: 0.11927059848676436
  15037. step: 78
  15038. running loss: 0.0015291102370097993
  15039. Train Steps: 78/90 Loss: 0.0015 torch.Size([8, 600, 800])
  15040. torch.Size([8, 8])
  15041. tensor([[0.6321, 0.4048, 0.8738, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  15042. [0.6087, 0.3951, 0.8387, 0.5833, 0.4188, 0.4933, 0.5146, 0.4830],
  15043. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5413, 0.5717],
  15044. [0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367],
  15045. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461],
  15046. [0.6259, 0.4156, 0.8812, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960],
  15047. [0.6271, 0.4040, 0.9138, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  15048. [0.6127, 0.4115, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495]],
  15049. device='cuda:0', dtype=torch.float64)
  15050. predictions are: tensor([[0.6987, 0.4584, 0.8383, 0.5464, 0.3512, 0.4217, 0.6277, 0.4938],
  15051. [0.5863, 0.3896, 0.8132, 0.5523, 0.3801, 0.4898, 0.5403, 0.5217],
  15052. [0.5919, 0.4097, 0.8451, 0.4679, 0.4123, 0.4491, 0.5641, 0.5819],
  15053. [0.6159, 0.4131, 0.8378, 0.2951, 0.3678, 0.2500, 0.6546, 0.5304],
  15054. [0.5996, 0.4130, 0.8316, 0.5114, 0.4123, 0.5076, 0.5484, 0.5542],
  15055. [0.5672, 0.3895, 0.8641, 0.2796, 0.4345, 0.1819, 0.6205, 0.5191],
  15056. [0.6335, 0.4170, 0.9235, 0.3661, 0.4421, 0.2307, 0.7317, 0.5447],
  15057. [0.5672, 0.3924, 0.7009, 0.2516, 0.3476, 0.2907, 0.5554, 0.5630]],
  15058. device='cuda:0', grad_fn=<AddmmBackward>)
  15059. landmarks are: tensor([[[0.6321, 0.4048, 0.8737, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  15060. [0.6087, 0.3951, 0.8388, 0.5833, 0.4187, 0.4933, 0.5146, 0.4830],
  15061. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5412, 0.5717],
  15062. [0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367],
  15063. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461],
  15064. [0.6259, 0.4156, 0.8813, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960],
  15065. [0.6271, 0.4040, 0.9137, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  15066. [0.6127, 0.4114, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495]]],
  15067. device='cuda:0')
  15068. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  15069. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  15070. loss_train: 0.11994522050372325
  15071. step: 79
  15072. running loss: 0.0015182939304268766
  15073. Train Steps: 79/90 Loss: 0.0015 torch.Size([8, 600, 800])
  15074. torch.Size([8, 8])
  15075. tensor([[ nan, nan, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  15076. [0.6167, 0.4048, 0.6831, 0.3639, 0.3763, 0.3017, 0.5700, 0.5883],
  15077. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  15078. [0.6299, 0.4008, 0.8450, 0.5350, 0.4213, 0.5000, 0.6350, 0.5100],
  15079. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650],
  15080. [0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133],
  15081. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  15082. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385]],
  15083. device='cuda:0', dtype=torch.float64)
  15084. predictions are: tensor([[0.1057, 0.0799, 0.6776, 0.2266, 0.4075, 0.1826, 0.5185, 0.5593],
  15085. [0.5822, 0.4002, 0.7446, 0.3650, 0.3553, 0.3069, 0.5613, 0.5871],
  15086. [0.7067, 0.4896, 0.7654, 0.2641, 0.4527, 0.1474, 0.6178, 0.5330],
  15087. [0.6936, 0.4618, 0.8660, 0.5623, 0.4117, 0.5137, 0.6381, 0.5277],
  15088. [0.6334, 0.4676, 0.8516, 0.4369, 0.3563, 0.3323, 0.5630, 0.5715],
  15089. [0.7058, 0.4797, 0.8861, 0.5675, 0.3770, 0.4973, 0.6390, 0.5419],
  15090. [0.6569, 0.4481, 0.7611, 0.2014, 0.4641, 0.1767, 0.6098, 0.5115],
  15091. [0.6362, 0.4427, 0.8369, 0.2680, 0.4433, 0.2233, 0.7059, 0.5670]],
  15092. device='cuda:0', grad_fn=<AddmmBackward>)
  15093. landmarks are: tensor([[[0.0000, 0.0000, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  15094. [0.6167, 0.4048, 0.6831, 0.3639, 0.3762, 0.3017, 0.5700, 0.5883],
  15095. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  15096. [0.6299, 0.4008, 0.8450, 0.5350, 0.4212, 0.5000, 0.6350, 0.5100],
  15097. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650],
  15098. [0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133],
  15099. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  15100. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385]]],
  15101. device='cuda:0')
  15102. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  15103. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  15104. loss_train: 0.121143347438192
  15105. step: 80
  15106. running loss: 0.0015142918429774
  15107.  
  15108. Train Steps: 80/90 Loss: 0.0015 torch.Size([8, 600, 800])
  15109. torch.Size([8, 8])
  15110. tensor([[0.6201, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  15111. [0.6111, 0.3995, 0.8788, 0.4567, 0.3813, 0.4833, 0.5450, 0.5700],
  15112. [0.6198, 0.4130, 0.8762, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  15113. [0.6185, 0.4080, 0.8625, 0.3483, 0.3788, 0.2650, 0.5320, 0.5272],
  15114. [0.6193, 0.4108, 0.7438, 0.2700, 0.3650, 0.3683, 0.6238, 0.5717],
  15115. [0.6197, 0.4091, 0.8800, 0.4783, 0.3538, 0.4767, 0.5950, 0.5550],
  15116. [0.6185, 0.4129, 0.8900, 0.4567, 0.3937, 0.5417, 0.5734, 0.5110],
  15117. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235]],
  15118. device='cuda:0', dtype=torch.float64)
  15119. predictions are: tensor([[0.5598, 0.3661, 0.7617, 0.2352, 0.4312, 0.1759, 0.5951, 0.5085],
  15120. [0.6472, 0.4201, 0.8735, 0.5111, 0.3916, 0.4521, 0.5697, 0.5639],
  15121. [0.6271, 0.4169, 0.8698, 0.4614, 0.3704, 0.4657, 0.5780, 0.5201],
  15122. [0.5978, 0.3887, 0.8416, 0.4152, 0.4066, 0.2550, 0.5161, 0.5368],
  15123. [0.6568, 0.4365, 0.7498, 0.3367, 0.3714, 0.3466, 0.6398, 0.5593],
  15124. [0.6814, 0.4443, 0.8768, 0.5257, 0.3883, 0.4427, 0.6180, 0.5738],
  15125. [0.5511, 0.3807, 0.8901, 0.5058, 0.4081, 0.5096, 0.5988, 0.5190],
  15126. [0.7312, 0.4653, 0.8871, 0.5383, 0.4363, 0.5173, 0.6679, 0.5432]],
  15127. device='cuda:0', grad_fn=<AddmmBackward>)
  15128. landmarks are: tensor([[[0.6202, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  15129. [0.6111, 0.3995, 0.8788, 0.4567, 0.3812, 0.4833, 0.5450, 0.5700],
  15130. [0.6198, 0.4130, 0.8763, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  15131. [0.6186, 0.4080, 0.8625, 0.3483, 0.3787, 0.2650, 0.5320, 0.5272],
  15132. [0.6193, 0.4108, 0.7437, 0.2700, 0.3650, 0.3683, 0.6237, 0.5717],
  15133. [0.6197, 0.4091, 0.8800, 0.4783, 0.3537, 0.4767, 0.5950, 0.5550],
  15134. [0.6186, 0.4129, 0.8900, 0.4567, 0.3938, 0.5417, 0.5734, 0.5110],
  15135. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235]]],
  15136. device='cuda:0')
  15137. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  15138. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  15139. loss_train: 0.12236118371947668
  15140. step: 81
  15141. running loss: 0.001510631897771317
  15142. Train Steps: 81/90 Loss: 0.0015 torch.Size([8, 600, 800])
  15143. torch.Size([8, 8])
  15144. tensor([[0.6145, 0.4007, 0.8775, 0.4533, 0.4562, 0.5533, 0.6088, 0.5533],
  15145. [0.6275, 0.4024, 0.8500, 0.5383, 0.3912, 0.4883, 0.6288, 0.5100],
  15146. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  15147. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385],
  15148. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  15149. [0.6200, 0.4059, 0.8700, 0.4900, 0.4163, 0.5000, 0.6162, 0.5467],
  15150. [0.6364, 0.4165, 0.9088, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  15151. [0.6198, 0.4101, 0.8838, 0.5283, 0.3763, 0.5267, 0.5913, 0.5567]],
  15152. device='cuda:0', dtype=torch.float64)
  15153. predictions are: tensor([[0.6552, 0.4201, 0.8652, 0.4655, 0.4552, 0.5254, 0.5912, 0.5429],
  15154. [0.6301, 0.3977, 0.8483, 0.5500, 0.4006, 0.4668, 0.6007, 0.4875],
  15155. [0.6143, 0.4100, 0.8121, 0.5916, 0.3956, 0.4262, 0.5624, 0.5386],
  15156. [0.6091, 0.4006, 0.8325, 0.2715, 0.4438, 0.2067, 0.6716, 0.5364],
  15157. [0.6086, 0.3938, 0.8833, 0.4870, 0.4127, 0.5155, 0.5829, 0.5079],
  15158. [0.5988, 0.3928, 0.8847, 0.5002, 0.4260, 0.4873, 0.5860, 0.5524],
  15159. [0.6430, 0.4311, 0.9103, 0.4571, 0.4090, 0.2978, 0.6228, 0.5287],
  15160. [0.6248, 0.3931, 0.8567, 0.5376, 0.3810, 0.5050, 0.5612, 0.5507]],
  15161. device='cuda:0', grad_fn=<AddmmBackward>)
  15162. landmarks are: tensor([[[0.6145, 0.4007, 0.8775, 0.4533, 0.4563, 0.5533, 0.6087, 0.5533],
  15163. [0.6275, 0.4024, 0.8500, 0.5383, 0.3913, 0.4883, 0.6288, 0.5100],
  15164. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  15165. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385],
  15166. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  15167. [0.6199, 0.4059, 0.8700, 0.4900, 0.4162, 0.5000, 0.6162, 0.5467],
  15168. [0.6364, 0.4165, 0.9087, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  15169. [0.6198, 0.4101, 0.8838, 0.5283, 0.3762, 0.5267, 0.5913, 0.5567]]],
  15170. device='cuda:0')
  15171. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  15172. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  15173. loss_train: 0.12267718787188642
  15174. step: 82
  15175. running loss: 0.0014960632667303222
  15176. Train Steps: 82/90 Loss: 0.0015 torch.Size([8, 600, 800])
  15177. torch.Size([8, 8])
  15178. tensor([[0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  15179. [0.6043, 0.4022, 0.6887, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136],
  15180. [0.6198, 0.3997, 0.8582, 0.5361, 0.4117, 0.5016, 0.5942, 0.5134],
  15181. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  15182. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  15183. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  15184. [0.6076, 0.3953, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954],
  15185. [0.6267, 0.4065, 0.8313, 0.2467, 0.4788, 0.1733, 0.6312, 0.5133]],
  15186. device='cuda:0', dtype=torch.float64)
  15187. predictions are: tensor([[0.5914, 0.3665, 0.7146, 0.2745, 0.4171, 0.2598, 0.5285, 0.5583],
  15188. [0.6297, 0.3911, 0.7009, 0.2508, 0.3974, 0.2863, 0.5636, 0.5424],
  15189. [0.5483, 0.3389, 0.8805, 0.5764, 0.4454, 0.5368, 0.5776, 0.5156],
  15190. [0.6216, 0.3915, 0.8177, 0.3364, 0.3854, 0.4146, 0.5627, 0.5295],
  15191. [0.6665, 0.4182, 0.8659, 0.3129, 0.5133, 0.2287, 0.6636, 0.5608],
  15192. [0.5599, 0.3312, 0.8511, 0.6297, 0.4080, 0.5062, 0.5952, 0.4846],
  15193. [0.6774, 0.4302, 0.8439, 0.4182, 0.3562, 0.4139, 0.5344, 0.5245],
  15194. [0.6530, 0.4147, 0.8346, 0.2688, 0.4997, 0.1857, 0.6275, 0.5368]],
  15195. device='cuda:0', grad_fn=<AddmmBackward>)
  15196. landmarks are: tensor([[[0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  15197. [0.6043, 0.4022, 0.6888, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136],
  15198. [0.6198, 0.3997, 0.8582, 0.5361, 0.4117, 0.5016, 0.5942, 0.5134],
  15199. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  15200. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  15201. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  15202. [0.6076, 0.3952, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954],
  15203. [0.6266, 0.4065, 0.8313, 0.2467, 0.4787, 0.1733, 0.6313, 0.5133]]],
  15204. device='cuda:0')
  15205. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  15206. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  15207. loss_train: 0.12356227377313189
  15208. step: 83
  15209. running loss: 0.0014887020936521914
  15210. Train Steps: 83/90 Loss: 0.0015 torch.Size([8, 600, 800])
  15211. torch.Size([8, 8])
  15212. tensor([[0.6104, 0.4029, 0.8738, 0.4900, 0.4088, 0.4533, 0.5070, 0.5510],
  15213. [0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166],
  15214. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  15215. [0.6212, 0.4159, 0.8675, 0.5783, 0.4088, 0.4317, 0.5613, 0.5917],
  15216. [0.6185, 0.4079, 0.8838, 0.4617, 0.4838, 0.5650, 0.6175, 0.5850],
  15217. [0.6175, 0.3957, 0.8700, 0.4817, 0.4662, 0.5133, 0.5800, 0.5517],
  15218. [0.6264, 0.4049, 0.8988, 0.4633, 0.3813, 0.4983, 0.6326, 0.4843],
  15219. [0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6088, 0.5583]],
  15220. device='cuda:0', dtype=torch.float64)
  15221. predictions are: tensor([[0.6121, 0.3956, 0.8771, 0.5035, 0.4204, 0.4715, 0.4994, 0.5032],
  15222. [0.6205, 0.3745, 0.9002, 0.5052, 0.3913, 0.5386, 0.6705, 0.4973],
  15223. [0.6060, 0.3819, 0.7070, 0.3297, 0.3670, 0.3024, 0.5258, 0.5516],
  15224. [0.6212, 0.4016, 0.8674, 0.5632, 0.4227, 0.4285, 0.5475, 0.5770],
  15225. [0.6462, 0.4000, 0.9094, 0.4689, 0.4834, 0.5711, 0.6178, 0.5542],
  15226. [0.5842, 0.3633, 0.8862, 0.4659, 0.4760, 0.5078, 0.5525, 0.5156],
  15227. [0.6476, 0.3985, 0.9158, 0.4879, 0.3884, 0.5060, 0.6152, 0.4727],
  15228. [0.6824, 0.4276, 0.7175, 0.2365, 0.3930, 0.2785, 0.5990, 0.5228]],
  15229. device='cuda:0', grad_fn=<AddmmBackward>)
  15230. landmarks are: tensor([[[0.6104, 0.4029, 0.8737, 0.4900, 0.4087, 0.4533, 0.5070, 0.5510],
  15231. [0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166],
  15232. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  15233. [0.6212, 0.4159, 0.8675, 0.5783, 0.4087, 0.4317, 0.5612, 0.5917],
  15234. [0.6184, 0.4079, 0.8838, 0.4617, 0.4837, 0.5650, 0.6175, 0.5850],
  15235. [0.6175, 0.3957, 0.8700, 0.4817, 0.4663, 0.5133, 0.5800, 0.5517],
  15236. [0.6264, 0.4049, 0.8988, 0.4633, 0.3812, 0.4983, 0.6326, 0.4843],
  15237. [0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6087, 0.5583]]],
  15238. device='cuda:0')
  15239. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  15240. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  15241. loss_train: 0.12402151894639246
  15242. step: 84
  15243. running loss: 0.0014764466541237198
  15244.  
  15245. Train Steps: 84/90 Loss: 0.0015 torch.Size([8, 600, 800])
  15246. torch.Size([8, 8])
  15247. tensor([[0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5413, 0.5433],
  15248. [0.6109, 0.4015, 0.7668, 0.3639, 0.3513, 0.3667, 0.5200, 0.5641],
  15249. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  15250. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  15251. [0.6277, 0.4083, 0.8350, 0.2717, 0.4562, 0.1800, 0.5918, 0.4878],
  15252. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  15253. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  15254. [0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742]],
  15255. device='cuda:0', dtype=torch.float64)
  15256. predictions are: tensor([[0.5706, 0.3611, 0.8415, 0.3558, 0.3894, 0.3482, 0.5369, 0.5443],
  15257. [0.6374, 0.4090, 0.7719, 0.3737, 0.3650, 0.3895, 0.5323, 0.5581],
  15258. [0.5539, 0.3452, 0.8682, 0.3352, 0.3994, 0.3267, 0.5938, 0.5154],
  15259. [0.6228, 0.4052, 0.7520, 0.2746, 0.4052, 0.2969, 0.5474, 0.5626],
  15260. [0.5389, 0.3360, 0.8147, 0.2744, 0.4840, 0.2404, 0.5875, 0.4859],
  15261. [0.5996, 0.3857, 0.8043, 0.3106, 0.3818, 0.4226, 0.5740, 0.5148],
  15262. [0.6152, 0.3860, 0.7280, 0.2786, 0.4149, 0.3206, 0.6044, 0.5611],
  15263. [0.6196, 0.3619, 0.8757, 0.3720, 0.4327, 0.2936, 0.6399, 0.4696]],
  15264. device='cuda:0', grad_fn=<AddmmBackward>)
  15265. landmarks are: tensor([[[0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5412, 0.5433],
  15266. [0.6109, 0.4015, 0.7668, 0.3639, 0.3512, 0.3667, 0.5200, 0.5641],
  15267. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  15268. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  15269. [0.6277, 0.4083, 0.8350, 0.2717, 0.4563, 0.1800, 0.5918, 0.4878],
  15270. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  15271. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  15272. [0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742]]],
  15273. device='cuda:0')
  15274. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  15275. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  15276. loss_train: 0.1248362893529702
  15277. step: 85
  15278. running loss: 0.0014686622276820023
  15279. Train Steps: 85/90 Loss: 0.0015 torch.Size([8, 600, 800])
  15280. torch.Size([8, 8])
  15281. tensor([[0.6233, 0.4091, 0.8100, 0.2950, 0.3563, 0.3883, 0.6013, 0.5200],
  15282. [0.6185, 0.4067, 0.8838, 0.4450, 0.4037, 0.4733, 0.5213, 0.5142],
  15283. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  15284. [0.6277, 0.4029, 0.8250, 0.2433, 0.4325, 0.2100, 0.6366, 0.5207],
  15285. [0.6126, 0.4067, 0.8638, 0.5383, 0.4188, 0.4850, 0.5016, 0.5392],
  15286. [0.6200, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  15287. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  15288. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5363, 0.5550]],
  15289. device='cuda:0', dtype=torch.float64)
  15290. predictions are: tensor([[0.6439, 0.4169, 0.7913, 0.2965, 0.3638, 0.4242, 0.6197, 0.5277],
  15291. [0.6116, 0.3866, 0.8863, 0.4314, 0.4029, 0.4827, 0.5048, 0.5232],
  15292. [0.6431, 0.4087, 0.8547, 0.3027, 0.3808, 0.3170, 0.5720, 0.5248],
  15293. [0.6157, 0.3800, 0.8009, 0.2299, 0.4458, 0.2494, 0.6712, 0.5019],
  15294. [0.6382, 0.4040, 0.8611, 0.5433, 0.4226, 0.5305, 0.5289, 0.5456],
  15295. [0.6433, 0.3967, 0.7910, 0.2740, 0.4048, 0.3044, 0.6042, 0.5222],
  15296. [0.6362, 0.4055, 0.8299, 0.3050, 0.3622, 0.4071, 0.6183, 0.5390],
  15297. [0.6083, 0.3849, 0.6878, 0.2513, 0.4030, 0.2424, 0.5473, 0.5537]],
  15298. device='cuda:0', grad_fn=<AddmmBackward>)
  15299. landmarks are: tensor([[[0.6233, 0.4091, 0.8100, 0.2950, 0.3562, 0.3883, 0.6012, 0.5200],
  15300. [0.6185, 0.4067, 0.8838, 0.4450, 0.4038, 0.4733, 0.5213, 0.5142],
  15301. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  15302. [0.6277, 0.4029, 0.8250, 0.2433, 0.4325, 0.2100, 0.6366, 0.5207],
  15303. [0.6126, 0.4067, 0.8637, 0.5383, 0.4187, 0.4850, 0.5016, 0.5392],
  15304. [0.6201, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  15305. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  15306. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5362, 0.5550]]],
  15307. device='cuda:0')
  15308. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  15309. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  15310. loss_train: 0.12514446672867052
  15311. step: 86
  15312. running loss: 0.0014551682177752386
  15313. Train Steps: 86/90 Loss: 0.0015 torch.Size([8, 600, 800])
  15314. torch.Size([8, 8])
  15315. tensor([[0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  15316. [0.6179, 0.3998, 0.8396, 0.3505, 0.3552, 0.3768, 0.6064, 0.5329],
  15317. [0.6108, 0.4011, 0.8037, 0.3400, 0.3700, 0.2933, 0.5658, 0.5617],
  15318. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  15319. [0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  15320. [0.6087, 0.3976, 0.8337, 0.3867, 0.3713, 0.3117, 0.5938, 0.5300],
  15321. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  15322. [ nan, nan, 0.6412, 0.1900, 0.4238, 0.1883, 0.5487, 0.5700]],
  15323. device='cuda:0', dtype=torch.float64)
  15324. predictions are: tensor([[0.5966, 0.3957, 0.8675, 0.5086, 0.3638, 0.5120, 0.6810, 0.5625],
  15325. [0.7951, 0.5214, 0.8327, 0.3075, 0.3375, 0.3722, 0.5612, 0.5145],
  15326. [0.6586, 0.4217, 0.8096, 0.3004, 0.3758, 0.3178, 0.5457, 0.5269],
  15327. [0.6461, 0.4297, 0.8695, 0.5199, 0.3957, 0.4628, 0.5302, 0.5536],
  15328. [0.6294, 0.4333, 0.8799, 0.4842, 0.3766, 0.5092, 0.5607, 0.4978],
  15329. [0.6841, 0.4662, 0.8497, 0.3571, 0.3594, 0.3482, 0.5710, 0.5243],
  15330. [0.6373, 0.4358, 0.8713, 0.2233, 0.4394, 0.2694, 0.6967, 0.5514],
  15331. [0.2011, 0.1628, 0.6991, 0.1745, 0.4166, 0.2143, 0.5079, 0.5705]],
  15332. device='cuda:0', grad_fn=<AddmmBackward>)
  15333. landmarks are: tensor([[[0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  15334. [0.6179, 0.3998, 0.8396, 0.3505, 0.3552, 0.3768, 0.6064, 0.5329],
  15335. [0.6108, 0.4011, 0.8037, 0.3400, 0.3700, 0.2933, 0.5658, 0.5617],
  15336. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  15337. [0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  15338. [0.6087, 0.3976, 0.8338, 0.3867, 0.3713, 0.3117, 0.5938, 0.5300],
  15339. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  15340. [0.0000, 0.0000, 0.6413, 0.1900, 0.4238, 0.1883, 0.5487, 0.5700]]],
  15341. device='cuda:0')
  15342. loss_train_step before backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  15343. loss_train_step after backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  15344. loss_train: 0.12772876137751155
  15345. step: 87
  15346. running loss: 0.0014681466825001326
  15347. Train Steps: 87/90 Loss: 0.0015 torch.Size([8, 600, 800])
  15348. torch.Size([8, 8])
  15349. tensor([[0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  15350. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864],
  15351. [0.6079, 0.3964, 0.7420, 0.2958, 0.3563, 0.2917, 0.5351, 0.4980],
  15352. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  15353. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901],
  15354. [0.6236, 0.3966, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  15355. [0.6357, 0.4118, 0.8400, 0.2500, 0.5413, 0.1633, 0.6725, 0.5586],
  15356. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650]],
  15357. device='cuda:0', dtype=torch.float64)
  15358. predictions are: tensor([[0.5803, 0.3842, 0.8580, 0.5116, 0.3559, 0.4980, 0.6933, 0.5874],
  15359. [0.6300, 0.4038, 0.8794, 0.4717, 0.3554, 0.5003, 0.7029, 0.5902],
  15360. [0.6064, 0.4055, 0.7442, 0.2507, 0.3430, 0.2914, 0.5109, 0.5141],
  15361. [0.5785, 0.3940, 0.8552, 0.3603, 0.3619, 0.5005, 0.5622, 0.5352],
  15362. [0.5799, 0.3884, 0.7843, 0.2502, 0.3571, 0.2743, 0.5150, 0.5242],
  15363. [0.5797, 0.3719, 0.8946, 0.4518, 0.3488, 0.4259, 0.5736, 0.5385],
  15364. [0.5465, 0.3934, 0.8544, 0.2093, 0.5169, 0.1659, 0.6492, 0.5658],
  15365. [0.6566, 0.4615, 0.8477, 0.3786, 0.3487, 0.3242, 0.5568, 0.5618]],
  15366. device='cuda:0', grad_fn=<AddmmBackward>)
  15367. landmarks are: tensor([[[0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  15368. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864],
  15369. [0.6079, 0.3964, 0.7420, 0.2958, 0.3562, 0.2917, 0.5351, 0.4980],
  15370. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  15371. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901],
  15372. [0.6236, 0.3965, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  15373. [0.6357, 0.4118, 0.8400, 0.2500, 0.5412, 0.1633, 0.6725, 0.5586],
  15374. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650]]],
  15375. device='cuda:0')
  15376. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  15377. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  15378. loss_train: 0.12843954053823836
  15379. step: 88
  15380. running loss: 0.0014595402333890722
  15381.  
  15382. Train Steps: 88/90 Loss: 0.0015 torch.Size([8, 600, 800])
  15383. torch.Size([8, 8])
  15384. tensor([[0.6068, 0.3963, 0.8650, 0.4317, 0.4037, 0.5083, 0.5253, 0.4999],
  15385. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5787, 0.5117],
  15386. [0.6170, 0.4102, 0.7468, 0.3695, 0.3463, 0.3767, 0.5238, 0.5823],
  15387. [0.6154, 0.4112, 0.7037, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  15388. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  15389. [0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  15390. [0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767],
  15391. [0.6131, 0.4037, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680]],
  15392. device='cuda:0', dtype=torch.float64)
  15393. predictions are: tensor([[0.6554, 0.4362, 0.8934, 0.4392, 0.3842, 0.4940, 0.5825, 0.5158],
  15394. [0.6204, 0.4117, 0.7705, 0.1987, 0.3993, 0.1871, 0.6053, 0.5255],
  15395. [0.5925, 0.3959, 0.7826, 0.3638, 0.3346, 0.3605, 0.5689, 0.5980],
  15396. [0.5463, 0.3805, 0.7074, 0.2054, 0.4121, 0.1620, 0.5610, 0.5656],
  15397. [0.6076, 0.4064, 0.8212, 0.2460, 0.3942, 0.2600, 0.6509, 0.5222],
  15398. [0.6327, 0.4301, 0.8894, 0.4261, 0.3661, 0.5427, 0.6239, 0.5256],
  15399. [0.5979, 0.3990, 0.9151, 0.5030, 0.3897, 0.4423, 0.5752, 0.5977],
  15400. [0.6053, 0.3951, 0.7183, 0.2489, 0.3607, 0.2643, 0.5485, 0.5817]],
  15401. device='cuda:0', grad_fn=<AddmmBackward>)
  15402. landmarks are: tensor([[[0.6068, 0.3963, 0.8650, 0.4317, 0.4038, 0.5083, 0.5253, 0.4999],
  15403. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5788, 0.5117],
  15404. [0.6170, 0.4102, 0.7468, 0.3695, 0.3462, 0.3767, 0.5238, 0.5823],
  15405. [0.6154, 0.4112, 0.7038, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  15406. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  15407. [0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  15408. [0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767],
  15409. [0.6131, 0.4036, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680]]],
  15410. device='cuda:0')
  15411. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  15412. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  15413. loss_train: 0.12903116384404711
  15414. step: 89
  15415. running loss: 0.0014497883577982821
  15416. Train Steps: 89/90 Loss: 0.0014 torch.Size([8, 600, 800])
  15417. torch.Size([8, 8])
  15418. tensor([[0.6148, 0.4076, 0.8666, 0.4820, 0.4138, 0.5067, 0.5250, 0.5767],
  15419. [0.6204, 0.4110, 0.7913, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367],
  15420. [0.6216, 0.4099, 0.7225, 0.2033, 0.4188, 0.2217, 0.5975, 0.5283],
  15421. [0.6108, 0.4011, 0.8037, 0.3400, 0.3700, 0.2933, 0.5658, 0.5617],
  15422. [0.6196, 0.4088, 0.8888, 0.4583, 0.4500, 0.5683, 0.6138, 0.5883],
  15423. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  15424. [0.6126, 0.3954, 0.8538, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  15425. [0.6198, 0.4115, 0.7762, 0.2717, 0.3713, 0.3200, 0.5837, 0.5683]],
  15426. device='cuda:0', dtype=torch.float64)
  15427. predictions are: tensor([[0.6364, 0.4298, 0.8848, 0.4947, 0.3976, 0.4907, 0.5543, 0.5624],
  15428. [0.5528, 0.3813, 0.7996, 0.2573, 0.3886, 0.2307, 0.6113, 0.5471],
  15429. [0.6042, 0.4352, 0.7250, 0.2135, 0.3953, 0.1917, 0.5677, 0.5342],
  15430. [0.6216, 0.3989, 0.8161, 0.3435, 0.3605, 0.2686, 0.5757, 0.5322],
  15431. [0.5356, 0.3706, 0.9046, 0.4557, 0.4116, 0.5370, 0.6324, 0.5777],
  15432. [0.4950, 0.3409, 0.7071, 0.2959, 0.3453, 0.2621, 0.5593, 0.5702],
  15433. [0.6456, 0.4085, 0.8733, 0.4935, 0.3967, 0.4418, 0.5593, 0.5491],
  15434. [0.6524, 0.4392, 0.7946, 0.2740, 0.3508, 0.2938, 0.5992, 0.5466]],
  15435. device='cuda:0', grad_fn=<AddmmBackward>)
  15436. landmarks are: tensor([[[0.6148, 0.4076, 0.8666, 0.4820, 0.4137, 0.5067, 0.5250, 0.5767],
  15437. [0.6204, 0.4110, 0.7912, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367],
  15438. [0.6216, 0.4099, 0.7225, 0.2033, 0.4187, 0.2217, 0.5975, 0.5283],
  15439. [0.6108, 0.4011, 0.8037, 0.3400, 0.3700, 0.2933, 0.5658, 0.5617],
  15440. [0.6196, 0.4088, 0.8888, 0.4583, 0.4500, 0.5683, 0.6137, 0.5883],
  15441. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  15442. [0.6126, 0.3954, 0.8537, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  15443. [0.6198, 0.4115, 0.7763, 0.2717, 0.3713, 0.3200, 0.5838, 0.5683]]],
  15444. device='cuda:0')
  15445. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  15446. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  15447. loss_train: 0.12994244889705442
  15448. step: 90
  15449. running loss: 0.001443804987745049
  15450. Valid Steps: 10/10 Loss: nan 3.3351
  15451. --------------------------------------------------
  15452. Epoch: 5 Train Loss: 0.0014 Valid Loss: nan
  15453. --------------------------------------------------
  15454. size of train loader is: 90
  15455. torch.Size([8, 600, 800])
  15456. torch.Size([8, 8])
  15457. tensor([[0.6271, 0.4040, 0.9138, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  15458. [ nan, nan, 0.8850, 0.3000, 0.5363, 0.2250, 0.7343, 0.5771],
  15459. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  15460. [0.6117, 0.4019, 0.8538, 0.4067, 0.3513, 0.3583, 0.5663, 0.5133],
  15461. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200],
  15462. [0.6243, 0.4128, 0.7762, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417],
  15463. [0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  15464. [0.6248, 0.4032, 0.7738, 0.1900, 0.4813, 0.1400, 0.5941, 0.4904]],
  15465. device='cuda:0', dtype=torch.float64)
  15466. predictions are: tensor([[ 0.6332, 0.4250, 0.8920, 0.3640, 0.4325, 0.2550, 0.6897, 0.5502],
  15467. [-0.0614, -0.0035, 0.8266, 0.2532, 0.4840, 0.2003, 0.6934, 0.5828],
  15468. [ 0.5656, 0.3930, 0.8296, 0.4740, 0.3736, 0.4827, 0.5273, 0.5308],
  15469. [ 0.6408, 0.4271, 0.8211, 0.3995, 0.3206, 0.3348, 0.5158, 0.5318],
  15470. [ 0.6423, 0.4379, 0.8163, 0.2955, 0.4129, 0.2113, 0.5810, 0.5201],
  15471. [ 0.6365, 0.4531, 0.7563, 0.2654, 0.3638, 0.2782, 0.5941, 0.5576],
  15472. [ 0.5648, 0.4057, 0.7927, 0.5257, 0.3614, 0.3913, 0.6561, 0.5536],
  15473. [ 0.5543, 0.3796, 0.7082, 0.1927, 0.4567, 0.1175, 0.5567, 0.5030]],
  15474. device='cuda:0', grad_fn=<AddmmBackward>)
  15475. landmarks are: tensor([[[0.6271, 0.4040, 0.9137, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  15476. [0.0000, 0.0000, 0.8850, 0.3000, 0.5362, 0.2250, 0.7343, 0.5771],
  15477. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  15478. [0.6116, 0.4019, 0.8537, 0.4067, 0.3512, 0.3583, 0.5663, 0.5133],
  15479. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200],
  15480. [0.6243, 0.4128, 0.7763, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417],
  15481. [0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  15482. [0.6248, 0.4032, 0.7738, 0.1900, 0.4812, 0.1400, 0.5941, 0.4904]]],
  15483. device='cuda:0')
  15484. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  15485. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  15486. loss_train: 0.0010950572323054075
  15487. step: 1
  15488. running loss: 0.0010950572323054075
  15489. Train Steps: 1/90 Loss: 0.0011 torch.Size([8, 600, 800])
  15490. torch.Size([8, 8])
  15491. tensor([[0.6178, 0.4059, 0.7525, 0.2250, 0.4313, 0.1783, 0.5404, 0.5240],
  15492. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  15493. [ nan, nan, 0.7515, 0.2708, 0.3987, 0.2267, 0.5162, 0.5567],
  15494. [0.6364, 0.4165, 0.9088, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  15495. [0.6197, 0.4118, 0.8688, 0.5517, 0.4037, 0.5233, 0.5875, 0.5600],
  15496. [0.6250, 0.4146, 0.8838, 0.3933, 0.3588, 0.4283, 0.6162, 0.5367],
  15497. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5463, 0.5800],
  15498. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123]],
  15499. device='cuda:0', dtype=torch.float64)
  15500. predictions are: tensor([[0.4767, 0.3139, 0.7132, 0.2130, 0.4388, 0.1563, 0.5310, 0.5519],
  15501. [0.5141, 0.3293, 0.8923, 0.3159, 0.5079, 0.2378, 0.7557, 0.5437],
  15502. [0.0024, 0.0341, 0.7160, 0.2549, 0.4101, 0.2055, 0.5066, 0.5869],
  15503. [0.6908, 0.4591, 0.9057, 0.4473, 0.4126, 0.2974, 0.6531, 0.5544],
  15504. [0.7181, 0.4650, 0.8476, 0.5692, 0.3929, 0.5315, 0.6112, 0.5648],
  15505. [0.6992, 0.4510, 0.8456, 0.4030, 0.3658, 0.4038, 0.6252, 0.5446],
  15506. [0.5817, 0.4095, 0.7255, 0.2901, 0.3826, 0.2522, 0.5198, 0.5925],
  15507. [0.5966, 0.4142, 0.7107, 0.2639, 0.3920, 0.2604, 0.5655, 0.5287]],
  15508. device='cuda:0', grad_fn=<AddmmBackward>)
  15509. landmarks are: tensor([[[0.6178, 0.4059, 0.7525, 0.2250, 0.4313, 0.1783, 0.5404, 0.5240],
  15510. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  15511. [0.0000, 0.0000, 0.7515, 0.2708, 0.3988, 0.2267, 0.5163, 0.5567],
  15512. [0.6364, 0.4165, 0.9087, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  15513. [0.6197, 0.4118, 0.8687, 0.5517, 0.4038, 0.5233, 0.5875, 0.5600],
  15514. [0.6250, 0.4146, 0.8838, 0.3933, 0.3587, 0.4283, 0.6162, 0.5367],
  15515. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5462, 0.5800],
  15516. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123]]],
  15517. device='cuda:0')
  15518. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  15519. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  15520. loss_train: 0.002500983187928796
  15521. step: 2
  15522. running loss: 0.001250491593964398
  15523.  
  15524. Train Steps: 2/90 Loss: 0.0013 torch.Size([8, 600, 800])
  15525. torch.Size([8, 8])
  15526. tensor([[0.6260, 0.4214, 0.8538, 0.5500, 0.3663, 0.3767, 0.5587, 0.5983],
  15527. [0.6266, 0.4067, 0.8588, 0.2867, 0.4300, 0.2850, 0.6325, 0.5267],
  15528. [0.6149, 0.4054, 0.6713, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  15529. [0.6307, 0.4029, 0.8650, 0.5200, 0.3763, 0.4017, 0.7311, 0.5366],
  15530. [0.6202, 0.4066, 0.8746, 0.3376, 0.3717, 0.3090, 0.5842, 0.5165],
  15531. [ nan, nan, 0.7515, 0.2708, 0.3987, 0.2267, 0.5162, 0.5567],
  15532. [0.6171, 0.4127, 0.8900, 0.4800, 0.4325, 0.5783, 0.5769, 0.5090],
  15533. [0.6218, 0.4137, 0.7263, 0.2233, 0.4075, 0.2650, 0.6212, 0.5783]],
  15534. device='cuda:0', dtype=torch.float64)
  15535. predictions are: tensor([[0.5942, 0.3765, 0.8441, 0.5555, 0.3895, 0.3756, 0.5572, 0.5978],
  15536. [0.6839, 0.4461, 0.8516, 0.2989, 0.4596, 0.2601, 0.6697, 0.5343],
  15537. [0.5452, 0.3702, 0.6386, 0.2365, 0.4353, 0.1893, 0.5224, 0.5777],
  15538. [0.6697, 0.4203, 0.8474, 0.5115, 0.4019, 0.3696, 0.7155, 0.5159],
  15539. [0.6741, 0.4410, 0.8573, 0.3529, 0.3951, 0.2907, 0.6152, 0.5146],
  15540. [0.0102, 0.0326, 0.7223, 0.2657, 0.4187, 0.2061, 0.5159, 0.5756],
  15541. [0.6716, 0.4447, 0.9013, 0.4962, 0.4554, 0.5630, 0.6017, 0.5144],
  15542. [0.6185, 0.3953, 0.7324, 0.2696, 0.4223, 0.2574, 0.6101, 0.5704]],
  15543. device='cuda:0', grad_fn=<AddmmBackward>)
  15544. landmarks are: tensor([[[0.6260, 0.4214, 0.8537, 0.5500, 0.3663, 0.3767, 0.5587, 0.5983],
  15545. [0.6266, 0.4067, 0.8587, 0.2867, 0.4300, 0.2850, 0.6325, 0.5267],
  15546. [0.6149, 0.4054, 0.6712, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  15547. [0.6307, 0.4029, 0.8650, 0.5200, 0.3762, 0.4017, 0.7311, 0.5366],
  15548. [0.6202, 0.4066, 0.8746, 0.3376, 0.3717, 0.3090, 0.5842, 0.5165],
  15549. [0.0000, 0.0000, 0.7515, 0.2708, 0.3988, 0.2267, 0.5163, 0.5567],
  15550. [0.6171, 0.4127, 0.8900, 0.4800, 0.4325, 0.5783, 0.5769, 0.5090],
  15551. [0.6218, 0.4137, 0.7262, 0.2233, 0.4075, 0.2650, 0.6212, 0.5783]]],
  15552. device='cuda:0')
  15553. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  15554. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  15555. loss_train: 0.0031757151591591537
  15556. step: 3
  15557. running loss: 0.001058571719719718
  15558. Train Steps: 3/90 Loss: 0.0011 torch.Size([8, 600, 800])
  15559. torch.Size([8, 8])
  15560. tensor([[0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  15561. [0.6040, 0.4002, 0.7338, 0.2267, 0.3975, 0.2100, 0.5231, 0.4778],
  15562. [0.6133, 0.4094, 0.8495, 0.4028, 0.3588, 0.3200, 0.5003, 0.5407],
  15563. [0.6144, 0.4032, 0.8563, 0.3283, 0.3525, 0.4200, 0.5775, 0.5583],
  15564. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  15565. [0.6311, 0.3998, 0.7975, 0.5767, 0.3838, 0.4850, 0.7327, 0.5343],
  15566. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  15567. [0.6241, 0.4143, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550]],
  15568. device='cuda:0', dtype=torch.float64)
  15569. predictions are: tensor([[0.5998, 0.3881, 0.8489, 0.5266, 0.4269, 0.5364, 0.5653, 0.5266],
  15570. [0.6322, 0.4066, 0.7096, 0.2361, 0.4164, 0.1925, 0.5609, 0.5039],
  15571. [0.6330, 0.4206, 0.8351, 0.4194, 0.3907, 0.3015, 0.5033, 0.5504],
  15572. [0.6071, 0.3998, 0.8333, 0.3566, 0.3592, 0.4133, 0.5639, 0.5432],
  15573. [0.6248, 0.3960, 0.8594, 0.5358, 0.4098, 0.5365, 0.6945, 0.5580],
  15574. [0.5699, 0.3728, 0.7798, 0.5499, 0.4036, 0.4531, 0.6909, 0.5258],
  15575. [0.5120, 0.3214, 0.8967, 0.3325, 0.5200, 0.2344, 0.7362, 0.5293],
  15576. [0.6220, 0.4074, 0.8722, 0.4873, 0.4371, 0.5165, 0.6189, 0.5546]],
  15577. device='cuda:0', grad_fn=<AddmmBackward>)
  15578. landmarks are: tensor([[[0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  15579. [0.6040, 0.4002, 0.7337, 0.2267, 0.3975, 0.2100, 0.5231, 0.4778],
  15580. [0.6133, 0.4094, 0.8495, 0.4028, 0.3587, 0.3200, 0.5003, 0.5407],
  15581. [0.6144, 0.4032, 0.8562, 0.3283, 0.3525, 0.4200, 0.5775, 0.5583],
  15582. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  15583. [0.6311, 0.3998, 0.7975, 0.5767, 0.3837, 0.4850, 0.7327, 0.5343],
  15584. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  15585. [0.6241, 0.4142, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550]]],
  15586. device='cuda:0')
  15587. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  15588. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  15589. loss_train: 0.003960852511227131
  15590. step: 4
  15591. running loss: 0.0009902131278067827
  15592. Train Steps: 4/90 Loss: 0.0010 torch.Size([8, 600, 800])
  15593. torch.Size([8, 8])
  15594. tensor([[0.6125, 0.4076, 0.8488, 0.3883, 0.3700, 0.3683, 0.5026, 0.5505],
  15595. [0.6141, 0.4038, 0.8650, 0.4833, 0.4839, 0.5176, 0.5787, 0.5600],
  15596. [0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367],
  15597. [0.6265, 0.4088, 0.8025, 0.1850, 0.4163, 0.2500, 0.6290, 0.4947],
  15598. [ nan, nan, 0.6900, 0.1917, 0.3937, 0.2367, 0.5240, 0.5246],
  15599. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  15600. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  15601. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323]],
  15602. device='cuda:0', dtype=torch.float64)
  15603. predictions are: tensor([[0.6783, 0.4287, 0.8345, 0.4148, 0.3889, 0.3670, 0.5106, 0.5487],
  15604. [0.6346, 0.3948, 0.8567, 0.5327, 0.4835, 0.5207, 0.6150, 0.5423],
  15605. [0.6474, 0.3986, 0.8573, 0.3465, 0.4161, 0.2739, 0.6702, 0.5225],
  15606. [0.6479, 0.4147, 0.7972, 0.2488, 0.4486, 0.2405, 0.6617, 0.5058],
  15607. [0.0196, 0.0039, 0.7029, 0.2429, 0.4289, 0.2278, 0.5220, 0.5362],
  15608. [0.5678, 0.3465, 0.8812, 0.5202, 0.3915, 0.5057, 0.6697, 0.4872],
  15609. [0.6323, 0.3935, 0.8804, 0.3828, 0.3995, 0.3069, 0.6250, 0.5382],
  15610. [0.6513, 0.4062, 0.7634, 0.2697, 0.4707, 0.2048, 0.6481, 0.5327]],
  15611. device='cuda:0', grad_fn=<AddmmBackward>)
  15612. landmarks are: tensor([[[0.6125, 0.4076, 0.8487, 0.3883, 0.3700, 0.3683, 0.5026, 0.5505],
  15613. [0.6141, 0.4038, 0.8650, 0.4833, 0.4839, 0.5176, 0.5788, 0.5600],
  15614. [0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367],
  15615. [0.6265, 0.4088, 0.8025, 0.1850, 0.4162, 0.2500, 0.6290, 0.4947],
  15616. [0.0000, 0.0000, 0.6900, 0.1917, 0.3938, 0.2367, 0.5240, 0.5246],
  15617. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  15618. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  15619. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323]]],
  15620. device='cuda:0')
  15621. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  15622. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  15623. loss_train: 0.00465098989661783
  15624. step: 5
  15625. running loss: 0.0009301979793235659
  15626. Train Steps: 5/90 Loss: 0.0009 torch.Size([8, 600, 800])
  15627. torch.Size([8, 8])
  15628. tensor([[0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  15629. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  15630. [0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  15631. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092],
  15632. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  15633. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  15634. [0.6182, 0.4058, 0.8738, 0.4350, 0.3563, 0.3400, 0.5290, 0.5822],
  15635. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100]],
  15636. device='cuda:0', dtype=torch.float64)
  15637. predictions are: tensor([[0.6552, 0.4114, 0.8774, 0.4658, 0.3782, 0.4515, 0.6856, 0.5181],
  15638. [0.6604, 0.4108, 0.9086, 0.4488, 0.4350, 0.3231, 0.7029, 0.5382],
  15639. [0.5866, 0.3753, 0.8584, 0.4249, 0.4075, 0.5679, 0.5804, 0.5033],
  15640. [0.6180, 0.3723, 0.8611, 0.5224, 0.4083, 0.5331, 0.5848, 0.5022],
  15641. [0.6441, 0.4107, 0.8707, 0.4167, 0.3678, 0.4041, 0.6107, 0.5390],
  15642. [0.5974, 0.3602, 0.8996, 0.5247, 0.3950, 0.3934, 0.6271, 0.4822],
  15643. [0.6110, 0.3776, 0.8502, 0.4215, 0.3777, 0.3466, 0.5408, 0.5714],
  15644. [0.5673, 0.3445, 0.8787, 0.4207, 0.3819, 0.3933, 0.6621, 0.5070]],
  15645. device='cuda:0', grad_fn=<AddmmBackward>)
  15646. landmarks are: tensor([[[0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  15647. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  15648. [0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  15649. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092],
  15650. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  15651. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  15652. [0.6182, 0.4058, 0.8737, 0.4350, 0.3562, 0.3400, 0.5290, 0.5822],
  15653. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100]]],
  15654. device='cuda:0')
  15655. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  15656. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  15657. loss_train: 0.00511719353380613
  15658. step: 6
  15659. running loss: 0.0008528655889676884
  15660.  
  15661. Train Steps: 6/90 Loss: 0.0009 torch.Size([8, 600, 800])
  15662. torch.Size([8, 8])
  15663. tensor([[0.6266, 0.4070, 0.8712, 0.5600, 0.3713, 0.4783, 0.5775, 0.6100],
  15664. [ nan, nan, 0.7268, 0.2333, 0.4125, 0.1933, 0.5112, 0.5383],
  15665. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5787, 0.5117],
  15666. [0.6201, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  15667. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  15668. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  15669. [0.6293, 0.3982, 0.8700, 0.5300, 0.3763, 0.4717, 0.7050, 0.5297],
  15670. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461]],
  15671. device='cuda:0', dtype=torch.float64)
  15672. predictions are: tensor([[0.6628, 0.4258, 0.8829, 0.5384, 0.3582, 0.5162, 0.6053, 0.5552],
  15673. [0.1885, 0.1040, 0.7453, 0.2245, 0.3961, 0.2227, 0.5406, 0.5308],
  15674. [0.6685, 0.4067, 0.7905, 0.2101, 0.4101, 0.2576, 0.5969, 0.4991],
  15675. [0.6771, 0.4192, 0.8103, 0.2036, 0.4245, 0.2512, 0.6132, 0.4906],
  15676. [0.6017, 0.3746, 0.8931, 0.4406, 0.4620, 0.3175, 0.5585, 0.5956],
  15677. [0.6587, 0.4070, 0.8789, 0.5625, 0.3910, 0.4548, 0.6079, 0.5264],
  15678. [0.6750, 0.4058, 0.8986, 0.5292, 0.3593, 0.5000, 0.6943, 0.5169],
  15679. [0.6772, 0.4275, 0.8966, 0.5189, 0.4330, 0.5449, 0.5616, 0.5367]],
  15680. device='cuda:0', grad_fn=<AddmmBackward>)
  15681. landmarks are: tensor([[[0.6266, 0.4070, 0.8712, 0.5600, 0.3713, 0.4783, 0.5775, 0.6100],
  15682. [0.0000, 0.0000, 0.7268, 0.2333, 0.4125, 0.1933, 0.5113, 0.5383],
  15683. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5788, 0.5117],
  15684. [0.6202, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  15685. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  15686. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  15687. [0.6293, 0.3982, 0.8700, 0.5300, 0.3762, 0.4717, 0.7050, 0.5297],
  15688. [0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461]]],
  15689. device='cuda:0')
  15690. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  15691. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  15692. loss_train: 0.006649552000453696
  15693. step: 7
  15694. running loss: 0.0009499360000648137
  15695. Train Steps: 7/90 Loss: 0.0009 torch.Size([8, 600, 800])
  15696. torch.Size([8, 8])
  15697. tensor([[0.6261, 0.3987, 0.9045, 0.4208, 0.3600, 0.4633, 0.6570, 0.5162],
  15698. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  15699. [0.6200, 0.4071, 0.7338, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  15700. [0.6126, 0.4039, 0.8237, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138],
  15701. [0.6115, 0.4005, 0.8838, 0.3867, 0.3763, 0.4700, 0.5800, 0.5550],
  15702. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  15703. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791],
  15704. [0.6248, 0.4185, 0.8500, 0.5767, 0.4463, 0.4550, 0.5613, 0.5917]],
  15705. device='cuda:0', dtype=torch.float64)
  15706. predictions are: tensor([[0.5887, 0.3806, 0.9167, 0.4414, 0.3467, 0.4701, 0.6545, 0.4994],
  15707. [0.6237, 0.4254, 0.9156, 0.4351, 0.4059, 0.5172, 0.5768, 0.5177],
  15708. [0.6543, 0.4100, 0.7849, 0.2122, 0.3987, 0.2520, 0.6252, 0.5278],
  15709. [0.6230, 0.3921, 0.8515, 0.4160, 0.3300, 0.3534, 0.6035, 0.5913],
  15710. [0.6065, 0.3897, 0.8989, 0.3883, 0.3638, 0.4858, 0.5655, 0.5183],
  15711. [0.6534, 0.4116, 0.8756, 0.5226, 0.3947, 0.5489, 0.7233, 0.5426],
  15712. [0.6262, 0.3883, 0.9243, 0.4744, 0.3783, 0.5022, 0.6143, 0.4519],
  15713. [0.6220, 0.3931, 0.8964, 0.5730, 0.4454, 0.4293, 0.5510, 0.5697]],
  15714. device='cuda:0', grad_fn=<AddmmBackward>)
  15715. landmarks are: tensor([[[0.6261, 0.3987, 0.9045, 0.4208, 0.3600, 0.4633, 0.6570, 0.5162],
  15716. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  15717. [0.6200, 0.4071, 0.7337, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  15718. [0.6126, 0.4038, 0.8238, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138],
  15719. [0.6115, 0.4005, 0.8838, 0.3867, 0.3762, 0.4700, 0.5800, 0.5550],
  15720. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  15721. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791],
  15722. [0.6248, 0.4185, 0.8500, 0.5767, 0.4462, 0.4550, 0.5612, 0.5917]]],
  15723. device='cuda:0')
  15724. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  15725. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  15726. loss_train: 0.007122047507436946
  15727. step: 8
  15728. running loss: 0.0008902559384296183
  15729. Train Steps: 8/90 Loss: 0.0009 torch.Size([8, 600, 800])
  15730. torch.Size([8, 8])
  15731. tensor([[0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  15732. [0.6273, 0.4110, 0.8900, 0.3817, 0.4188, 0.2167, 0.5858, 0.4835],
  15733. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6362, 0.5200],
  15734. [ nan, nan, 0.6512, 0.1717, 0.4100, 0.1983, 0.5253, 0.5240],
  15735. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6038, 0.4833],
  15736. [0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767],
  15737. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6038, 0.6167],
  15738. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250]],
  15739. device='cuda:0', dtype=torch.float64)
  15740. predictions are: tensor([[0.6781, 0.4366, 0.8292, 0.2767, 0.3947, 0.3024, 0.6299, 0.5228],
  15741. [0.6502, 0.4163, 0.9356, 0.3927, 0.4165, 0.2404, 0.6043, 0.5037],
  15742. [0.6243, 0.4212, 0.8959, 0.3969, 0.3492, 0.5258, 0.6535, 0.5265],
  15743. [0.0754, 0.0272, 0.7135, 0.2025, 0.3964, 0.2239, 0.5353, 0.5389],
  15744. [0.7175, 0.4695, 0.9051, 0.5078, 0.3554, 0.4924, 0.5934, 0.5303],
  15745. [0.6314, 0.4253, 0.9020, 0.5113, 0.4033, 0.4586, 0.5315, 0.6031],
  15746. [0.6845, 0.4592, 0.8323, 0.3439, 0.3670, 0.4061, 0.5821, 0.6231],
  15747. [0.6510, 0.4404, 0.9164, 0.4900, 0.4029, 0.5442, 0.6044, 0.5414]],
  15748. device='cuda:0', grad_fn=<AddmmBackward>)
  15749. landmarks are: tensor([[[0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  15750. [0.6273, 0.4110, 0.8900, 0.3817, 0.4187, 0.2167, 0.5858, 0.4835],
  15751. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6363, 0.5200],
  15752. [0.0000, 0.0000, 0.6513, 0.1717, 0.4100, 0.1983, 0.5253, 0.5240],
  15753. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6037, 0.4833],
  15754. [0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767],
  15755. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6037, 0.6167],
  15756. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250]]],
  15757. device='cuda:0')
  15758. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  15759. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  15760. loss_train: 0.008067115588346496
  15761. step: 9
  15762. running loss: 0.000896346176482944
  15763. Train Steps: 9/90 Loss: 0.0009 torch.Size([8, 600, 800])
  15764. torch.Size([8, 8])
  15765. tensor([[0.6250, 0.4131, 0.8688, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  15766. [0.6353, 0.4128, 0.9138, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  15767. [ nan, nan, 0.7512, 0.2117, 0.4288, 0.2000, 0.5600, 0.5617],
  15768. [0.6275, 0.4157, 0.8337, 0.5800, 0.3763, 0.4200, 0.5547, 0.6125],
  15769. [0.6204, 0.4055, 0.8438, 0.5733, 0.4574, 0.4801, 0.5487, 0.5617],
  15770. [0.6265, 0.4088, 0.8025, 0.1850, 0.4163, 0.2500, 0.6290, 0.4947],
  15771. [0.6339, 0.4102, 0.9088, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  15772. [0.6273, 0.4100, 0.7137, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633]],
  15773. device='cuda:0', dtype=torch.float64)
  15774. predictions are: tensor([[0.6229, 0.4118, 0.8863, 0.2859, 0.4184, 0.2557, 0.5658, 0.5340],
  15775. [0.6148, 0.4197, 0.9315, 0.3354, 0.4334, 0.3215, 0.6852, 0.5903],
  15776. [0.0754, 0.0627, 0.7600, 0.2218, 0.3954, 0.2259, 0.5242, 0.5600],
  15777. [0.5883, 0.4249, 0.8646, 0.5609, 0.3616, 0.4497, 0.5613, 0.6069],
  15778. [0.5997, 0.4042, 0.8556, 0.5515, 0.4356, 0.4546, 0.4996, 0.5693],
  15779. [0.6244, 0.4190, 0.8221, 0.1943, 0.4030, 0.2628, 0.5984, 0.5144],
  15780. [0.6360, 0.4487, 0.9374, 0.4746, 0.3753, 0.5383, 0.7273, 0.5499],
  15781. [0.5903, 0.3867, 0.7336, 0.2222, 0.3727, 0.2676, 0.5683, 0.5639]],
  15782. device='cuda:0', grad_fn=<AddmmBackward>)
  15783. landmarks are: tensor([[[0.6250, 0.4131, 0.8687, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  15784. [0.6353, 0.4128, 0.9137, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  15785. [0.0000, 0.0000, 0.7513, 0.2117, 0.4288, 0.2000, 0.5600, 0.5617],
  15786. [0.6275, 0.4157, 0.8338, 0.5800, 0.3762, 0.4200, 0.5547, 0.6125],
  15787. [0.6204, 0.4055, 0.8438, 0.5733, 0.4574, 0.4801, 0.5487, 0.5617],
  15788. [0.6265, 0.4088, 0.8025, 0.1850, 0.4162, 0.2500, 0.6290, 0.4947],
  15789. [0.6339, 0.4102, 0.9087, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  15790. [0.6273, 0.4099, 0.7138, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633]]],
  15791. device='cuda:0')
  15792. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  15793. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  15794. loss_train: 0.008671639865497127
  15795. step: 10
  15796. running loss: 0.0008671639865497127
  15797.  
  15798. Train Steps: 10/90 Loss: 0.0009 torch.Size([8, 600, 800])
  15799. torch.Size([8, 8])
  15800. tensor([[0.6332, 0.4165, 0.9100, 0.3350, 0.4188, 0.3683, 0.7438, 0.5528],
  15801. [0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767],
  15802. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  15803. [0.6289, 0.4081, 0.8720, 0.3487, 0.3900, 0.3183, 0.6703, 0.5376],
  15804. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  15805. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  15806. [0.6182, 0.4058, 0.8738, 0.4350, 0.3563, 0.3400, 0.5290, 0.5822],
  15807. [0.6175, 0.4093, 0.8800, 0.4433, 0.4075, 0.4367, 0.5128, 0.4901]],
  15808. device='cuda:0', dtype=torch.float64)
  15809. predictions are: tensor([[0.5651, 0.3872, 0.9091, 0.3589, 0.4060, 0.3687, 0.7129, 0.5629],
  15810. [0.5442, 0.3813, 0.8742, 0.4722, 0.4190, 0.4401, 0.5225, 0.5954],
  15811. [0.5596, 0.3881, 0.8712, 0.4179, 0.3868, 0.4584, 0.5164, 0.5327],
  15812. [0.5353, 0.3653, 0.8768, 0.3212, 0.3934, 0.3146, 0.6572, 0.5564],
  15813. [0.5771, 0.3766, 0.8791, 0.4666, 0.4206, 0.5151, 0.6954, 0.5690],
  15814. [0.5786, 0.3886, 0.7983, 0.4074, 0.3795, 0.4221, 0.5060, 0.5502],
  15815. [0.5158, 0.3602, 0.8586, 0.4000, 0.3567, 0.3225, 0.5129, 0.5861],
  15816. [0.6043, 0.4215, 0.8886, 0.4290, 0.4082, 0.4468, 0.5101, 0.5186]],
  15817. device='cuda:0', grad_fn=<AddmmBackward>)
  15818. landmarks are: tensor([[[0.6332, 0.4165, 0.9100, 0.3350, 0.4187, 0.3683, 0.7438, 0.5528],
  15819. [0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767],
  15820. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  15821. [0.6289, 0.4081, 0.8720, 0.3487, 0.3900, 0.3183, 0.6703, 0.5376],
  15822. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  15823. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  15824. [0.6182, 0.4058, 0.8737, 0.4350, 0.3562, 0.3400, 0.5290, 0.5822],
  15825. [0.6175, 0.4093, 0.8800, 0.4433, 0.4075, 0.4367, 0.5128, 0.4901]]],
  15826. device='cuda:0')
  15827. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  15828. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  15829. loss_train: 0.00954322874895297
  15830. step: 11
  15831. running loss: 0.0008675662499048154
  15832. Train Steps: 11/90 Loss: 0.0009 torch.Size([8, 600, 800])
  15833. torch.Size([8, 8])
  15834. tensor([[0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  15835. [0.6048, 0.3928, 0.8538, 0.5433, 0.3875, 0.5117, 0.5266, 0.4719],
  15836. [0.6243, 0.4128, 0.7762, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417],
  15837. [0.6361, 0.4076, 0.8862, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297],
  15838. [0.6203, 0.4021, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  15839. [0.6273, 0.4100, 0.7137, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633],
  15840. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  15841. [0.6175, 0.3957, 0.8700, 0.4817, 0.4662, 0.5133, 0.5800, 0.5517]],
  15842. device='cuda:0', dtype=torch.float64)
  15843. predictions are: tensor([[0.5483, 0.3622, 0.8770, 0.4779, 0.3990, 0.5093, 0.6017, 0.5437],
  15844. [0.5784, 0.3904, 0.8539, 0.5166, 0.3958, 0.4806, 0.5558, 0.5333],
  15845. [0.5637, 0.3742, 0.7941, 0.2559, 0.3872, 0.3076, 0.6280, 0.5515],
  15846. [0.5998, 0.4185, 0.8796, 0.5261, 0.3549, 0.4438, 0.6606, 0.5561],
  15847. [0.6237, 0.4060, 0.8862, 0.4990, 0.3595, 0.3872, 0.5895, 0.5282],
  15848. [0.5888, 0.3756, 0.7120, 0.2081, 0.3834, 0.2500, 0.6050, 0.5684],
  15849. [0.5443, 0.3876, 0.8513, 0.4988, 0.4388, 0.5221, 0.5204, 0.5598],
  15850. [0.5425, 0.3668, 0.8625, 0.4351, 0.4530, 0.4921, 0.5606, 0.5558]],
  15851. device='cuda:0', grad_fn=<AddmmBackward>)
  15852. landmarks are: tensor([[[0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  15853. [0.6048, 0.3928, 0.8537, 0.5433, 0.3875, 0.5117, 0.5266, 0.4719],
  15854. [0.6243, 0.4128, 0.7763, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417],
  15855. [0.6361, 0.4076, 0.8863, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297],
  15856. [0.6203, 0.4020, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  15857. [0.6273, 0.4099, 0.7138, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633],
  15858. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  15859. [0.6175, 0.3957, 0.8700, 0.4817, 0.4663, 0.5133, 0.5800, 0.5517]]],
  15860. device='cuda:0')
  15861. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  15862. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  15863. loss_train: 0.010210297914454713
  15864. step: 12
  15865. running loss: 0.0008508581595378928
  15866. Train Steps: 12/90 Loss: 0.0009 torch.Size([8, 600, 800])
  15867. torch.Size([8, 8])
  15868. tensor([[0.6190, 0.4135, 0.8000, 0.4883, 0.3566, 0.3647, 0.5613, 0.5900],
  15869. [0.6173, 0.4114, 0.7325, 0.2500, 0.4213, 0.1917, 0.5338, 0.5700],
  15870. [ nan, nan, 0.7225, 0.2167, 0.3987, 0.2283, 0.5427, 0.5181],
  15871. [0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6088, 0.5583],
  15872. [0.6262, 0.4163, 0.8850, 0.5183, 0.3763, 0.4150, 0.6025, 0.5500],
  15873. [0.6266, 0.4070, 0.8712, 0.5600, 0.3713, 0.4783, 0.5775, 0.6100],
  15874. [ nan, nan, 0.8525, 0.2217, 0.5413, 0.2367, 0.7367, 0.5482],
  15875. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864]],
  15876. device='cuda:0', dtype=torch.float64)
  15877. predictions are: tensor([[0.6665, 0.4454, 0.8174, 0.4735, 0.3770, 0.3649, 0.5131, 0.5624],
  15878. [0.5387, 0.3739, 0.7161, 0.2188, 0.4464, 0.2063, 0.5391, 0.5532],
  15879. [0.2135, 0.1407, 0.6820, 0.1918, 0.4179, 0.2471, 0.5318, 0.5153],
  15880. [0.6443, 0.4274, 0.6857, 0.2046, 0.4088, 0.2694, 0.6036, 0.5470],
  15881. [0.6675, 0.4486, 0.8739, 0.4873, 0.3920, 0.4055, 0.5779, 0.5486],
  15882. [0.6897, 0.4687, 0.8466, 0.5476, 0.3902, 0.4891, 0.5805, 0.5826],
  15883. [0.2292, 0.1626, 0.8446, 0.2121, 0.5398, 0.2417, 0.7055, 0.5412],
  15884. [0.6525, 0.4344, 0.8758, 0.5061, 0.3929, 0.4984, 0.7030, 0.5711]],
  15885. device='cuda:0', grad_fn=<AddmmBackward>)
  15886. landmarks are: tensor([[[0.6190, 0.4135, 0.8000, 0.4883, 0.3566, 0.3647, 0.5612, 0.5900],
  15887. [0.6173, 0.4114, 0.7325, 0.2500, 0.4212, 0.1917, 0.5337, 0.5700],
  15888. [0.0000, 0.0000, 0.7225, 0.2167, 0.3988, 0.2283, 0.5427, 0.5181],
  15889. [0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6087, 0.5583],
  15890. [0.6262, 0.4163, 0.8850, 0.5183, 0.3762, 0.4150, 0.6025, 0.5500],
  15891. [0.6266, 0.4070, 0.8712, 0.5600, 0.3713, 0.4783, 0.5775, 0.6100],
  15892. [0.0000, 0.0000, 0.8525, 0.2217, 0.5412, 0.2367, 0.7367, 0.5482],
  15893. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864]]],
  15894. device='cuda:0')
  15895. loss_train_step before backward: tensor(0.0029, device='cuda:0', grad_fn=<MseLossBackward>)
  15896. loss_train_step after backward: tensor(0.0029, device='cuda:0', grad_fn=<MseLossBackward>)
  15897. loss_train: 0.013074482005322352
  15898. step: 13
  15899. running loss: 0.0010057293850247962
  15900. Train Steps: 13/90 Loss: 0.0010 torch.Size([8, 600, 800])
  15901. torch.Size([8, 8])
  15902. tensor([[0.6277, 0.4057, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  15903. [0.6087, 0.3951, 0.8387, 0.5833, 0.4188, 0.4933, 0.5146, 0.4830],
  15904. [0.6034, 0.4011, 0.7350, 0.2533, 0.3438, 0.3367, 0.5516, 0.5084],
  15905. [0.6272, 0.4071, 0.8738, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  15906. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  15907. [0.6275, 0.4003, 0.9100, 0.3783, 0.4388, 0.3133, 0.7058, 0.5343],
  15908. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  15909. [0.6260, 0.4120, 0.8013, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895]],
  15910. device='cuda:0', dtype=torch.float64)
  15911. predictions are: tensor([[0.6209, 0.3946, 0.7829, 0.2354, 0.4578, 0.2138, 0.5985, 0.5199],
  15912. [0.5912, 0.3801, 0.8002, 0.5580, 0.4329, 0.5129, 0.5151, 0.5371],
  15913. [0.5538, 0.3627, 0.6871, 0.2188, 0.3659, 0.3580, 0.5647, 0.5449],
  15914. [0.5544, 0.3579, 0.8420, 0.5472, 0.3969, 0.4023, 0.6108, 0.4973],
  15915. [0.5533, 0.3713, 0.8420, 0.4789, 0.3996, 0.3498, 0.6161, 0.5338],
  15916. [0.5912, 0.3869, 0.8465, 0.3534, 0.4348, 0.3184, 0.6907, 0.5468],
  15917. [0.6274, 0.3913, 0.8715, 0.4603, 0.3700, 0.3854, 0.6073, 0.5570],
  15918. [0.5417, 0.3396, 0.7500, 0.1926, 0.4977, 0.1744, 0.6412, 0.5392]],
  15919. device='cuda:0', grad_fn=<AddmmBackward>)
  15920. landmarks are: tensor([[[0.6277, 0.4056, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  15921. [0.6087, 0.3951, 0.8388, 0.5833, 0.4187, 0.4933, 0.5146, 0.4830],
  15922. [0.6033, 0.4011, 0.7350, 0.2533, 0.3438, 0.3367, 0.5516, 0.5084],
  15923. [0.6272, 0.4071, 0.8737, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  15924. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  15925. [0.6275, 0.4003, 0.9100, 0.3783, 0.4387, 0.3133, 0.7058, 0.5343],
  15926. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  15927. [0.6259, 0.4120, 0.8012, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895]]],
  15928. device='cuda:0')
  15929. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  15930. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  15931. loss_train: 0.014277539594331756
  15932. step: 14
  15933. running loss: 0.0010198242567379826
  15934.  
  15935. Train Steps: 14/90 Loss: 0.0010 torch.Size([8, 600, 800])
  15936. torch.Size([8, 8])
  15937. tensor([[0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  15938. [0.6083, 0.3957, 0.8638, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  15939. [0.6200, 0.4071, 0.7338, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  15940. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  15941. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  15942. [0.6201, 0.4017, 0.8871, 0.4621, 0.3517, 0.4675, 0.5999, 0.5106],
  15943. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  15944. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274]],
  15945. device='cuda:0', dtype=torch.float64)
  15946. predictions are: tensor([[0.5333, 0.3506, 0.6552, 0.2076, 0.4242, 0.1751, 0.5727, 0.5451],
  15947. [0.4893, 0.3109, 0.8448, 0.5058, 0.4601, 0.5073, 0.5673, 0.4904],
  15948. [0.5904, 0.3878, 0.7141, 0.1966, 0.4354, 0.2181, 0.6287, 0.5480],
  15949. [0.5772, 0.3840, 0.8524, 0.4117, 0.3834, 0.3702, 0.5531, 0.5023],
  15950. [0.6300, 0.4052, 0.8221, 0.5879, 0.3911, 0.4689, 0.6662, 0.5207],
  15951. [0.6077, 0.3779, 0.8499, 0.4657, 0.3675, 0.4600, 0.6185, 0.5338],
  15952. [0.6065, 0.3866, 0.8750, 0.4366, 0.3635, 0.3120, 0.6016, 0.5102],
  15953. [0.5697, 0.3748, 0.8739, 0.4813, 0.3705, 0.4117, 0.6781, 0.5255]],
  15954. device='cuda:0', grad_fn=<AddmmBackward>)
  15955. landmarks are: tensor([[[0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  15956. [0.6083, 0.3957, 0.8637, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  15957. [0.6200, 0.4071, 0.7337, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  15958. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  15959. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  15960. [0.6201, 0.4017, 0.8871, 0.4621, 0.3517, 0.4675, 0.5999, 0.5106],
  15961. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  15962. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274]]],
  15963. device='cuda:0')
  15964. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  15965. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  15966. loss_train: 0.015163687261519954
  15967. step: 15
  15968. running loss: 0.0010109124841013303
  15969. Train Steps: 15/90 Loss: 0.0010 torch.Size([8, 600, 800])
  15970. torch.Size([8, 8])
  15971. tensor([[0.6157, 0.4102, 0.8513, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  15972. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  15973. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  15974. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  15975. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  15976. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  15977. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  15978. [0.6085, 0.4005, 0.8400, 0.4317, 0.3763, 0.4750, 0.5476, 0.5058]],
  15979. device='cuda:0', dtype=torch.float64)
  15980. predictions are: tensor([[0.5410, 0.3491, 0.8061, 0.3897, 0.3737, 0.3296, 0.5710, 0.5361],
  15981. [0.5908, 0.3836, 0.8486, 0.5070, 0.3837, 0.4084, 0.5487, 0.5290],
  15982. [0.5794, 0.3698, 0.8578, 0.4783, 0.4464, 0.5413, 0.6610, 0.5036],
  15983. [0.5673, 0.3616, 0.8306, 0.5433, 0.3512, 0.3234, 0.6101, 0.5062],
  15984. [0.5835, 0.3757, 0.8322, 0.3436, 0.3784, 0.2795, 0.6196, 0.4811],
  15985. [0.5601, 0.3804, 0.8507, 0.4410, 0.4294, 0.4775, 0.6118, 0.5142],
  15986. [0.6267, 0.4019, 0.8304, 0.4744, 0.3647, 0.2856, 0.5849, 0.5461],
  15987. [0.6285, 0.4035, 0.8221, 0.4240, 0.3711, 0.4418, 0.5872, 0.4826]],
  15988. device='cuda:0', grad_fn=<AddmmBackward>)
  15989. landmarks are: tensor([[[0.6157, 0.4102, 0.8512, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  15990. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  15991. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  15992. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  15993. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  15994. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  15995. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  15996. [0.6084, 0.4005, 0.8400, 0.4317, 0.3762, 0.4750, 0.5476, 0.5058]]],
  15997. device='cuda:0')
  15998. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  15999. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  16000. loss_train: 0.016262349992757663
  16001. step: 16
  16002. running loss: 0.001016396874547354
  16003. Train Steps: 16/90 Loss: 0.0010 torch.Size([8, 600, 800])
  16004. torch.Size([8, 8])
  16005. tensor([[0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  16006. [0.6109, 0.4003, 0.8650, 0.4883, 0.4775, 0.4867, 0.5175, 0.5683],
  16007. [0.6200, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  16008. [0.6273, 0.4105, 0.8988, 0.4517, 0.3912, 0.2550, 0.5894, 0.4811],
  16009. [ nan, nan, 0.6412, 0.1900, 0.4238, 0.1883, 0.5487, 0.5700],
  16010. [0.6336, 0.4191, 0.8938, 0.5167, 0.3937, 0.3517, 0.7343, 0.5748],
  16011. [0.6138, 0.4101, 0.8800, 0.5083, 0.4637, 0.5950, 0.5587, 0.5077],
  16012. [0.6109, 0.4041, 0.6975, 0.3167, 0.3513, 0.3383, 0.5153, 0.5319]],
  16013. device='cuda:0', dtype=torch.float64)
  16014. predictions are: tensor([[0.6855, 0.4411, 0.8811, 0.4829, 0.3468, 0.4056, 0.5976, 0.4854],
  16015. [0.6003, 0.3924, 0.8620, 0.5050, 0.4516, 0.4777, 0.5454, 0.5403],
  16016. [0.7025, 0.4502, 0.8847, 0.4959, 0.3405, 0.4132, 0.5959, 0.4986],
  16017. [0.6217, 0.3892, 0.8945, 0.4343, 0.3822, 0.2371, 0.5838, 0.4497],
  16018. [0.0274, 0.0246, 0.6846, 0.2371, 0.3901, 0.1621, 0.5692, 0.5287],
  16019. [0.6936, 0.4408, 0.8705, 0.5191, 0.3849, 0.3495, 0.7009, 0.5208],
  16020. [0.6504, 0.4256, 0.8840, 0.5347, 0.4355, 0.5537, 0.5773, 0.5028],
  16021. [0.6316, 0.4177, 0.7242, 0.3255, 0.3338, 0.3280, 0.5260, 0.4997]],
  16022. device='cuda:0', grad_fn=<AddmmBackward>)
  16023. landmarks are: tensor([[[0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  16024. [0.6109, 0.4003, 0.8650, 0.4883, 0.4775, 0.4867, 0.5175, 0.5683],
  16025. [0.6201, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  16026. [0.6273, 0.4105, 0.8988, 0.4517, 0.3913, 0.2550, 0.5894, 0.4811],
  16027. [0.0000, 0.0000, 0.6413, 0.1900, 0.4238, 0.1883, 0.5487, 0.5700],
  16028. [0.6336, 0.4191, 0.8938, 0.5167, 0.3938, 0.3517, 0.7343, 0.5748],
  16029. [0.6138, 0.4101, 0.8800, 0.5083, 0.4638, 0.5950, 0.5587, 0.5077],
  16030. [0.6109, 0.4041, 0.6975, 0.3167, 0.3512, 0.3383, 0.5153, 0.5319]]],
  16031. device='cuda:0')
  16032. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  16033. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  16034. loss_train: 0.0170363754441496
  16035. step: 17
  16036. running loss: 0.0010021397320088
  16037. Train Steps: 17/90 Loss: 0.0010 torch.Size([8, 600, 800])
  16038. torch.Size([8, 8])
  16039. tensor([[0.6120, 0.4014, 0.6863, 0.2817, 0.3700, 0.2783, 0.5513, 0.5667],
  16040. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  16041. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  16042. [0.6276, 0.4095, 0.8237, 0.2250, 0.4662, 0.1783, 0.6171, 0.4869],
  16043. [ nan, nan, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  16044. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  16045. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  16046. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833]],
  16047. device='cuda:0', dtype=torch.float64)
  16048. predictions are: tensor([[0.6260, 0.4011, 0.7258, 0.3199, 0.3612, 0.2915, 0.5442, 0.5361],
  16049. [0.7346, 0.4767, 0.9288, 0.5064, 0.4090, 0.3807, 0.7193, 0.5532],
  16050. [0.6979, 0.4502, 0.8781, 0.4688, 0.3707, 0.4679, 0.5312, 0.5525],
  16051. [0.6071, 0.3951, 0.8454, 0.2538, 0.4587, 0.1876, 0.6201, 0.4572],
  16052. [0.0940, 0.0501, 0.8083, 0.3363, 0.3641, 0.2925, 0.5379, 0.5242],
  16053. [0.6665, 0.4243, 0.8967, 0.5811, 0.4485, 0.5496, 0.5898, 0.5064],
  16054. [0.6506, 0.4104, 0.7844, 0.2195, 0.4589, 0.1643, 0.5733, 0.4783],
  16055. [0.6472, 0.4191, 0.7120, 0.3332, 0.3684, 0.2936, 0.5573, 0.5656]],
  16056. device='cuda:0', grad_fn=<AddmmBackward>)
  16057. landmarks are: tensor([[[0.6120, 0.4013, 0.6862, 0.2817, 0.3700, 0.2783, 0.5512, 0.5667],
  16058. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  16059. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  16060. [0.6276, 0.4095, 0.8238, 0.2250, 0.4663, 0.1783, 0.6171, 0.4869],
  16061. [0.0000, 0.0000, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  16062. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  16063. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  16064. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833]]],
  16065. device='cuda:0')
  16066. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  16067. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  16068. loss_train: 0.018217981763882563
  16069. step: 18
  16070. running loss: 0.0010121100979934756
  16071.  
  16072. Train Steps: 18/90 Loss: 0.0010 torch.Size([8, 600, 800])
  16073. torch.Size([8, 8])
  16074. tensor([[0.6173, 0.4114, 0.7325, 0.2500, 0.4213, 0.1917, 0.5338, 0.5700],
  16075. [0.6141, 0.4038, 0.8650, 0.4833, 0.4839, 0.5176, 0.5787, 0.5600],
  16076. [0.6170, 0.4102, 0.7468, 0.3695, 0.3463, 0.3767, 0.5238, 0.5823],
  16077. [0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726],
  16078. [0.6154, 0.4112, 0.7037, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  16079. [0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038],
  16080. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  16081. [0.6201, 0.4064, 0.8688, 0.5050, 0.4225, 0.5100, 0.6138, 0.5500]],
  16082. device='cuda:0', dtype=torch.float64)
  16083. predictions are: tensor([[0.5512, 0.3725, 0.7604, 0.2443, 0.4207, 0.1884, 0.5394, 0.5562],
  16084. [0.6294, 0.3869, 0.9030, 0.5222, 0.4701, 0.5182, 0.5802, 0.5584],
  16085. [0.5952, 0.3779, 0.7930, 0.3845, 0.3451, 0.3834, 0.5343, 0.5833],
  16086. [0.6804, 0.4306, 0.8026, 0.3019, 0.3634, 0.2921, 0.5605, 0.4898],
  16087. [0.6347, 0.4243, 0.7262, 0.2597, 0.4223, 0.1819, 0.5501, 0.5555],
  16088. [0.5753, 0.3615, 0.9085, 0.5114, 0.4387, 0.5237, 0.5442, 0.5066],
  16089. [0.5027, 0.3365, 0.8310, 0.3479, 0.3453, 0.2930, 0.5062, 0.5514],
  16090. [0.6423, 0.4073, 0.9061, 0.5245, 0.4004, 0.5121, 0.5782, 0.5466]],
  16091. device='cuda:0', grad_fn=<AddmmBackward>)
  16092. landmarks are: tensor([[[0.6173, 0.4114, 0.7325, 0.2500, 0.4212, 0.1917, 0.5337, 0.5700],
  16093. [0.6141, 0.4038, 0.8650, 0.4833, 0.4839, 0.5176, 0.5788, 0.5600],
  16094. [0.6170, 0.4102, 0.7468, 0.3695, 0.3462, 0.3767, 0.5238, 0.5823],
  16095. [0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726],
  16096. [0.6154, 0.4112, 0.7038, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  16097. [0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038],
  16098. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  16099. [0.6201, 0.4064, 0.8687, 0.5050, 0.4225, 0.5100, 0.6137, 0.5500]]],
  16100. device='cuda:0')
  16101. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  16102. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  16103. loss_train: 0.019125032442389056
  16104. step: 19
  16105. running loss: 0.0010065806548625819
  16106. Train Steps: 19/90 Loss: 0.0010 torch.Size([8, 600, 800])
  16107. torch.Size([8, 8])
  16108. tensor([[0.6201, 0.4082, 0.8827, 0.3715, 0.3825, 0.2712, 0.5845, 0.5412],
  16109. [0.6182, 0.4058, 0.8738, 0.4350, 0.3563, 0.3400, 0.5290, 0.5822],
  16110. [0.6200, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913],
  16111. [0.6179, 0.4040, 0.7412, 0.1850, 0.3825, 0.2783, 0.5837, 0.5600],
  16112. [0.6208, 0.4082, 0.8538, 0.3067, 0.3588, 0.3717, 0.6112, 0.5517],
  16113. [0.6179, 0.3998, 0.8396, 0.3505, 0.3552, 0.3768, 0.6064, 0.5329],
  16114. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  16115. [0.6068, 0.3963, 0.8650, 0.4317, 0.4037, 0.5083, 0.5253, 0.4999]],
  16116. device='cuda:0', dtype=torch.float64)
  16117. predictions are: tensor([[0.6359, 0.4048, 0.9052, 0.3762, 0.3995, 0.2797, 0.5899, 0.5466],
  16118. [0.6064, 0.3951, 0.8949, 0.4519, 0.3761, 0.3334, 0.5104, 0.6051],
  16119. [0.6164, 0.3700, 0.8848, 0.5386, 0.4169, 0.5256, 0.5715, 0.5245],
  16120. [0.6481, 0.4276, 0.7565, 0.2404, 0.3964, 0.2855, 0.5759, 0.5843],
  16121. [0.6212, 0.4118, 0.8703, 0.3438, 0.3663, 0.3934, 0.5951, 0.5677],
  16122. [0.6131, 0.3926, 0.8612, 0.3559, 0.3660, 0.3619, 0.5658, 0.5549],
  16123. [0.5858, 0.3855, 0.7570, 0.2178, 0.4388, 0.2146, 0.5812, 0.5519],
  16124. [0.5676, 0.3535, 0.8760, 0.4405, 0.4186, 0.5101, 0.5055, 0.5193]],
  16125. device='cuda:0', grad_fn=<AddmmBackward>)
  16126. landmarks are: tensor([[[0.6201, 0.4082, 0.8827, 0.3715, 0.3825, 0.2712, 0.5845, 0.5412],
  16127. [0.6182, 0.4058, 0.8737, 0.4350, 0.3562, 0.3400, 0.5290, 0.5822],
  16128. [0.6199, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913],
  16129. [0.6179, 0.4040, 0.7412, 0.1850, 0.3825, 0.2783, 0.5838, 0.5600],
  16130. [0.6208, 0.4082, 0.8537, 0.3067, 0.3587, 0.3717, 0.6112, 0.5517],
  16131. [0.6179, 0.3998, 0.8396, 0.3505, 0.3552, 0.3768, 0.6064, 0.5329],
  16132. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  16133. [0.6068, 0.3963, 0.8650, 0.4317, 0.4038, 0.5083, 0.5253, 0.4999]]],
  16134. device='cuda:0')
  16135. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  16136. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  16137. loss_train: 0.01953231156221591
  16138. step: 20
  16139. running loss: 0.0009766155781107955
  16140. Train Steps: 20/90 Loss: 0.0010 torch.Size([8, 600, 800])
  16141. torch.Size([8, 8])
  16142. tensor([[0.6206, 0.4001, 0.8900, 0.3933, 0.3588, 0.3567, 0.5837, 0.5083],
  16143. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317],
  16144. [ nan, nan, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600],
  16145. [0.6128, 0.4022, 0.8738, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  16146. [0.6108, 0.4011, 0.8037, 0.3400, 0.3700, 0.2933, 0.5658, 0.5617],
  16147. [ nan, nan, 0.7850, 0.2700, 0.4288, 0.1717, 0.5199, 0.4999],
  16148. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  16149. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466]],
  16150. device='cuda:0', dtype=torch.float64)
  16151. predictions are: tensor([[0.7210, 0.4565, 0.9084, 0.3731, 0.3526, 0.3772, 0.5655, 0.5376],
  16152. [0.8087, 0.5395, 0.8758, 0.5171, 0.3853, 0.3925, 0.5778, 0.5738],
  16153. [0.1119, 0.0785, 0.7751, 0.2568, 0.4144, 0.2698, 0.5499, 0.5953],
  16154. [0.6822, 0.4636, 0.8693, 0.4949, 0.4949, 0.5419, 0.5108, 0.5413],
  16155. [0.7736, 0.5026, 0.8111, 0.3225, 0.3772, 0.3114, 0.5589, 0.5785],
  16156. [0.0275, 0.0309, 0.7832, 0.2323, 0.4275, 0.1838, 0.5191, 0.5356],
  16157. [0.7302, 0.4657, 0.9031, 0.4539, 0.3850, 0.5179, 0.6317, 0.5228],
  16158. [0.6900, 0.4722, 0.8493, 0.4946, 0.4511, 0.5650, 0.5092, 0.5610]],
  16159. device='cuda:0', grad_fn=<AddmmBackward>)
  16160. landmarks are: tensor([[[0.6206, 0.4001, 0.8900, 0.3933, 0.3587, 0.3567, 0.5838, 0.5083],
  16161. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317],
  16162. [0.0000, 0.0000, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600],
  16163. [0.6128, 0.4022, 0.8737, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  16164. [0.6108, 0.4011, 0.8037, 0.3400, 0.3700, 0.2933, 0.5658, 0.5617],
  16165. [0.0000, 0.0000, 0.7850, 0.2700, 0.4288, 0.1717, 0.5199, 0.4999],
  16166. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  16167. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466]]],
  16168. device='cuda:0')
  16169. loss_train_step before backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  16170. loss_train_step after backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  16171. loss_train: 0.022330504696583375
  16172. step: 21
  16173. running loss: 0.0010633573665039702
  16174. Train Steps: 21/90 Loss: 0.0011 torch.Size([8, 600, 800])
  16175. torch.Size([8, 8])
  16176. tensor([[0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  16177. [0.6111, 0.4019, 0.8350, 0.3867, 0.3500, 0.4283, 0.5480, 0.5719],
  16178. [0.6127, 0.4084, 0.8700, 0.4467, 0.3987, 0.4317, 0.5013, 0.5471],
  16179. [0.6108, 0.4011, 0.8037, 0.3400, 0.3700, 0.2933, 0.5658, 0.5617],
  16180. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  16181. [ nan, nan, 0.8888, 0.3100, 0.5262, 0.2817, 0.7145, 0.6003],
  16182. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5463, 0.5800],
  16183. [0.6126, 0.4039, 0.8237, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138]],
  16184. device='cuda:0', dtype=torch.float64)
  16185. predictions are: tensor([[0.6185, 0.3869, 0.8579, 0.5263, 0.4154, 0.5089, 0.5843, 0.5349],
  16186. [0.6103, 0.4033, 0.8438, 0.3815, 0.3781, 0.4424, 0.5075, 0.5624],
  16187. [0.6637, 0.4565, 0.8744, 0.4178, 0.4033, 0.4657, 0.4512, 0.5411],
  16188. [0.6865, 0.4526, 0.8055, 0.3225, 0.3858, 0.3246, 0.5445, 0.5547],
  16189. [0.6587, 0.4332, 0.8624, 0.2587, 0.5050, 0.2517, 0.6967, 0.5345],
  16190. [0.1376, 0.0907, 0.8921, 0.2710, 0.5097, 0.2876, 0.7036, 0.5853],
  16191. [0.6178, 0.4209, 0.7452, 0.2694, 0.3747, 0.3018, 0.5023, 0.5806],
  16192. [0.6423, 0.4236, 0.8180, 0.3788, 0.3601, 0.3860, 0.5589, 0.6233]],
  16193. device='cuda:0', grad_fn=<AddmmBackward>)
  16194. landmarks are: tensor([[[0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  16195. [0.6111, 0.4019, 0.8350, 0.3867, 0.3500, 0.4283, 0.5480, 0.5719],
  16196. [0.6127, 0.4084, 0.8700, 0.4467, 0.3988, 0.4317, 0.5013, 0.5471],
  16197. [0.6108, 0.4011, 0.8037, 0.3400, 0.3700, 0.2933, 0.5658, 0.5617],
  16198. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  16199. [0.0000, 0.0000, 0.8888, 0.3100, 0.5263, 0.2817, 0.7145, 0.6003],
  16200. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5462, 0.5800],
  16201. [0.6126, 0.4038, 0.8238, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138]]],
  16202. device='cuda:0')
  16203. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  16204. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  16205. loss_train: 0.023386229382595047
  16206. step: 22
  16207. running loss: 0.001063010426481593
  16208.  
  16209. Train Steps: 22/90 Loss: 0.0011 torch.Size([8, 600, 800])
  16210. torch.Size([8, 8])
  16211. tensor([[0.6048, 0.3928, 0.8538, 0.5433, 0.3875, 0.5117, 0.5266, 0.4719],
  16212. [0.6275, 0.4024, 0.8600, 0.2283, 0.5350, 0.1800, 0.7074, 0.5413],
  16213. [0.6149, 0.4054, 0.6713, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  16214. [0.6259, 0.4156, 0.8812, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960],
  16215. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  16216. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  16217. [0.6196, 0.4094, 0.7562, 0.2817, 0.3937, 0.3183, 0.6013, 0.6183],
  16218. [0.6234, 0.4179, 0.7825, 0.3450, 0.3813, 0.2867, 0.5675, 0.5617]],
  16219. device='cuda:0', dtype=torch.float64)
  16220. predictions are: tensor([[0.5246, 0.3330, 0.8417, 0.5379, 0.3996, 0.5314, 0.5079, 0.5074],
  16221. [0.5499, 0.3492, 0.8426, 0.2148, 0.5311, 0.2304, 0.6864, 0.5399],
  16222. [0.5512, 0.3678, 0.6624, 0.2023, 0.4101, 0.2173, 0.4776, 0.5696],
  16223. [0.5862, 0.3871, 0.8917, 0.2849, 0.4597, 0.2297, 0.5993, 0.5250],
  16224. [0.5173, 0.3275, 0.7987, 0.2612, 0.3914, 0.2580, 0.5748, 0.5194],
  16225. [0.5717, 0.3551, 0.8947, 0.4054, 0.3440, 0.4823, 0.6041, 0.5235],
  16226. [0.5840, 0.3897, 0.7675, 0.2568, 0.3996, 0.3258, 0.5925, 0.6365],
  16227. [0.5287, 0.3531, 0.7851, 0.3243, 0.3914, 0.2944, 0.5629, 0.5665]],
  16228. device='cuda:0', grad_fn=<AddmmBackward>)
  16229. landmarks are: tensor([[[0.6048, 0.3928, 0.8537, 0.5433, 0.3875, 0.5117, 0.5266, 0.4719],
  16230. [0.6275, 0.4024, 0.8600, 0.2283, 0.5350, 0.1800, 0.7074, 0.5413],
  16231. [0.6149, 0.4054, 0.6712, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  16232. [0.6259, 0.4156, 0.8813, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960],
  16233. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  16234. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  16235. [0.6196, 0.4094, 0.7563, 0.2817, 0.3938, 0.3183, 0.6012, 0.6183],
  16236. [0.6234, 0.4179, 0.7825, 0.3450, 0.3812, 0.2867, 0.5675, 0.5617]]],
  16237. device='cuda:0')
  16238. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  16239. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  16240. loss_train: 0.024698010383872315
  16241. step: 23
  16242. running loss: 0.001073826538429231
  16243. Train Steps: 23/90 Loss: 0.0011 torch.Size([8, 600, 800])
  16244. torch.Size([8, 8])
  16245. tensor([[0.6248, 0.4185, 0.8500, 0.5767, 0.4463, 0.4550, 0.5613, 0.5917],
  16246. [0.6132, 0.4066, 0.7259, 0.2402, 0.3588, 0.3300, 0.6000, 0.5600],
  16247. [0.6203, 0.4078, 0.8800, 0.5083, 0.3900, 0.5000, 0.6100, 0.5583],
  16248. [0.6254, 0.4076, 0.8700, 0.3267, 0.4150, 0.3083, 0.7050, 0.5609],
  16249. [0.6357, 0.4118, 0.8400, 0.2500, 0.5413, 0.1633, 0.6725, 0.5586],
  16250. [0.6157, 0.4102, 0.8513, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  16251. [0.6075, 0.4000, 0.8513, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  16252. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351]],
  16253. device='cuda:0', dtype=torch.float64)
  16254. predictions are: tensor([[0.5903, 0.3847, 0.8431, 0.5428, 0.4399, 0.4542, 0.5502, 0.5792],
  16255. [0.5266, 0.3435, 0.7182, 0.2179, 0.3495, 0.3179, 0.5764, 0.5612],
  16256. [0.5280, 0.3317, 0.8819, 0.4689, 0.3673, 0.4914, 0.5831, 0.5414],
  16257. [0.6012, 0.3880, 0.8877, 0.2876, 0.4039, 0.2774, 0.6887, 0.5500],
  16258. [0.5292, 0.3549, 0.8352, 0.2285, 0.5218, 0.1683, 0.6503, 0.5430],
  16259. [0.5647, 0.3668, 0.8404, 0.3564, 0.3523, 0.3591, 0.5268, 0.5653],
  16260. [0.5416, 0.3683, 0.8392, 0.5033, 0.4493, 0.5229, 0.5332, 0.5184],
  16261. [0.4986, 0.3387, 0.7712, 0.2292, 0.4214, 0.1707, 0.5749, 0.5097]],
  16262. device='cuda:0', grad_fn=<AddmmBackward>)
  16263. landmarks are: tensor([[[0.6248, 0.4185, 0.8500, 0.5767, 0.4462, 0.4550, 0.5612, 0.5917],
  16264. [0.6132, 0.4066, 0.7259, 0.2402, 0.3587, 0.3300, 0.6000, 0.5600],
  16265. [0.6203, 0.4078, 0.8800, 0.5083, 0.3900, 0.5000, 0.6100, 0.5583],
  16266. [0.6254, 0.4076, 0.8700, 0.3267, 0.4150, 0.3083, 0.7050, 0.5609],
  16267. [0.6357, 0.4118, 0.8400, 0.2500, 0.5412, 0.1633, 0.6725, 0.5586],
  16268. [0.6157, 0.4102, 0.8512, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  16269. [0.6075, 0.4000, 0.8512, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  16270. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351]]],
  16271. device='cuda:0')
  16272. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  16273. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  16274. loss_train: 0.02608701281133108
  16275. step: 24
  16276. running loss: 0.0010869588671387949
  16277. Train Steps: 24/90 Loss: 0.0011 torch.Size([8, 600, 800])
  16278. torch.Size([8, 8])
  16279. tensor([[0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  16280. [ nan, nan, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  16281. [0.6092, 0.4001, 0.8638, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  16282. [0.6185, 0.4067, 0.8838, 0.4450, 0.4037, 0.4733, 0.5213, 0.5142],
  16283. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  16284. [ nan, nan, 0.7268, 0.2333, 0.4125, 0.1933, 0.5112, 0.5383],
  16285. [0.6332, 0.4165, 0.9100, 0.3350, 0.4188, 0.3683, 0.7438, 0.5528],
  16286. [ nan, nan, 0.8363, 0.3317, 0.3563, 0.3367, 0.5329, 0.5142]],
  16287. device='cuda:0', dtype=torch.float64)
  16288. predictions are: tensor([[0.7749, 0.5050, 0.8667, 0.5227, 0.4023, 0.5295, 0.7262, 0.5520],
  16289. [0.1078, 0.0715, 0.8621, 0.3088, 0.5078, 0.1750, 0.6640, 0.5756],
  16290. [0.7126, 0.4748, 0.8329, 0.4779, 0.4340, 0.5367, 0.5249, 0.5027],
  16291. [0.7413, 0.5045, 0.8681, 0.4425, 0.4100, 0.4426, 0.5186, 0.5116],
  16292. [0.7770, 0.5188, 0.8618, 0.3211, 0.4173, 0.4002, 0.6991, 0.5632],
  16293. [0.0340, 0.0423, 0.6935, 0.2104, 0.4144, 0.1369, 0.5162, 0.5456],
  16294. [0.7058, 0.4849, 0.8911, 0.3628, 0.4073, 0.3391, 0.7341, 0.5740],
  16295. [0.1683, 0.1309, 0.7931, 0.3116, 0.3325, 0.3041, 0.5446, 0.5366]],
  16296. device='cuda:0', grad_fn=<AddmmBackward>)
  16297. landmarks are: tensor([[[0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  16298. [0.0000, 0.0000, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  16299. [0.6092, 0.4001, 0.8637, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  16300. [0.6185, 0.4067, 0.8838, 0.4450, 0.4038, 0.4733, 0.5213, 0.5142],
  16301. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  16302. [0.0000, 0.0000, 0.7268, 0.2333, 0.4125, 0.1933, 0.5113, 0.5383],
  16303. [0.6332, 0.4165, 0.9100, 0.3350, 0.4187, 0.3683, 0.7438, 0.5528],
  16304. [0.0000, 0.0000, 0.8363, 0.3317, 0.3562, 0.3367, 0.5329, 0.5142]]],
  16305. device='cuda:0')
  16306. loss_train_step before backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  16307. loss_train_step after backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  16308. loss_train: 0.02925560690346174
  16309. step: 25
  16310. running loss: 0.0011702242761384696
  16311. Train Steps: 25/90 Loss: 0.0012 torch.Size([8, 600, 800])
  16312. torch.Size([8, 8])
  16313. tensor([[0.6185, 0.4080, 0.8625, 0.3483, 0.3788, 0.2650, 0.5320, 0.5272],
  16314. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  16315. [0.6192, 0.4128, 0.8513, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  16316. [0.6229, 0.4066, 0.7612, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  16317. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  16318. [0.6201, 0.4064, 0.8688, 0.5050, 0.4225, 0.5100, 0.6138, 0.5500],
  16319. [0.6109, 0.4041, 0.6975, 0.3167, 0.3513, 0.3383, 0.5153, 0.5319],
  16320. [0.6182, 0.3972, 0.8552, 0.5914, 0.3683, 0.4181, 0.5688, 0.5378]],
  16321. device='cuda:0', dtype=torch.float64)
  16322. predictions are: tensor([[0.5933, 0.3969, 0.8577, 0.3590, 0.4118, 0.2590, 0.5607, 0.5364],
  16323. [0.5607, 0.3531, 0.8947, 0.4054, 0.3543, 0.4176, 0.6658, 0.5124],
  16324. [0.5784, 0.3753, 0.8585, 0.5589, 0.4376, 0.4992, 0.6307, 0.5515],
  16325. [0.5546, 0.3648, 0.7714, 0.2658, 0.4262, 0.2079, 0.6115, 0.5436],
  16326. [0.4776, 0.3190, 0.8711, 0.3610, 0.3545, 0.3455, 0.6230, 0.5217],
  16327. [0.5311, 0.3501, 0.8663, 0.4774, 0.4244, 0.4933, 0.6378, 0.5392],
  16328. [0.5744, 0.3876, 0.7364, 0.2986, 0.3520, 0.3252, 0.5476, 0.5329],
  16329. [0.6285, 0.4031, 0.8640, 0.5768, 0.3735, 0.3970, 0.6315, 0.5407]],
  16330. device='cuda:0', grad_fn=<AddmmBackward>)
  16331. landmarks are: tensor([[[0.6186, 0.4080, 0.8625, 0.3483, 0.3787, 0.2650, 0.5320, 0.5272],
  16332. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  16333. [0.6192, 0.4128, 0.8512, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  16334. [0.6229, 0.4066, 0.7613, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  16335. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  16336. [0.6201, 0.4064, 0.8687, 0.5050, 0.4225, 0.5100, 0.6137, 0.5500],
  16337. [0.6109, 0.4041, 0.6975, 0.3167, 0.3512, 0.3383, 0.5153, 0.5319],
  16338. [0.6182, 0.3972, 0.8552, 0.5914, 0.3683, 0.4181, 0.5688, 0.5378]]],
  16339. device='cuda:0')
  16340. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  16341. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  16342. loss_train: 0.03047404598328285
  16343. step: 26
  16344. running loss: 0.001172078691664725
  16345.  
  16346. Train Steps: 26/90 Loss: 0.0012 torch.Size([8, 600, 800])
  16347. torch.Size([8, 8])
  16348. tensor([[0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  16349. [ nan, nan, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  16350. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  16351. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  16352. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  16353. [0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  16354. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  16355. [0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915]],
  16356. device='cuda:0', dtype=torch.float64)
  16357. predictions are: tensor([[ 0.5571, 0.3781, 0.8592, 0.4977, 0.3940, 0.5043, 0.6304, 0.5167],
  16358. [-0.0641, -0.0320, 0.8451, 0.2732, 0.5266, 0.1686, 0.6831, 0.5694],
  16359. [ 0.5501, 0.3686, 0.8577, 0.4885, 0.4755, 0.5070, 0.5333, 0.5305],
  16360. [ 0.6006, 0.4094, 0.8789, 0.5209, 0.3491, 0.3783, 0.6467, 0.4951],
  16361. [ 0.5343, 0.3618, 0.8626, 0.4401, 0.4468, 0.5363, 0.6245, 0.5325],
  16362. [ 0.5859, 0.4038, 0.8746, 0.5054, 0.3550, 0.4501, 0.6062, 0.5472],
  16363. [ 0.5593, 0.3798, 0.8754, 0.3642, 0.4343, 0.2383, 0.6505, 0.5330],
  16364. [ 0.6434, 0.4352, 0.8833, 0.4716, 0.3494, 0.3238, 0.6562, 0.5142]],
  16365. device='cuda:0', grad_fn=<AddmmBackward>)
  16366. landmarks are: tensor([[[0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  16367. [0.0000, 0.0000, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  16368. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  16369. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  16370. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  16371. [0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  16372. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  16373. [0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915]]],
  16374. device='cuda:0')
  16375. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  16376. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  16377. loss_train: 0.03130909425090067
  16378. step: 27
  16379. running loss: 0.0011595960833666915
  16380. Train Steps: 27/90 Loss: 0.0012 torch.Size([8, 600, 800])
  16381. torch.Size([8, 8])
  16382. tensor([[0.6203, 0.4078, 0.8800, 0.5083, 0.3900, 0.5000, 0.6100, 0.5583],
  16383. [0.6153, 0.4117, 0.8688, 0.5167, 0.4895, 0.5647, 0.5524, 0.5136],
  16384. [0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5837, 0.5317],
  16385. [0.6179, 0.3998, 0.8396, 0.3505, 0.3552, 0.3768, 0.6064, 0.5329],
  16386. [0.6270, 0.4267, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  16387. [ nan, nan, 0.8525, 0.2217, 0.5413, 0.2367, 0.7367, 0.5482],
  16388. [0.6151, 0.4125, 0.8738, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  16389. [ nan, nan, 0.6469, 0.1943, 0.4025, 0.2000, 0.5125, 0.5533]],
  16390. device='cuda:0', dtype=torch.float64)
  16391. predictions are: tensor([[ 0.6480, 0.4196, 0.8940, 0.5324, 0.3837, 0.5083, 0.6373, 0.5335],
  16392. [ 0.6795, 0.4661, 0.8864, 0.5449, 0.4735, 0.5319, 0.6194, 0.5147],
  16393. [ 0.5716, 0.3851, 0.8499, 0.3645, 0.3668, 0.3217, 0.6308, 0.5389],
  16394. [ 0.6398, 0.4191, 0.8459, 0.3649, 0.3495, 0.3682, 0.6308, 0.5194],
  16395. [ 0.6781, 0.4620, 0.7139, 0.3219, 0.4760, 0.1850, 0.5917, 0.5935],
  16396. [-0.0955, -0.0567, 0.8607, 0.2603, 0.5467, 0.2307, 0.7484, 0.5299],
  16397. [ 0.6616, 0.4385, 0.8714, 0.4713, 0.3620, 0.3759, 0.5641, 0.5374],
  16398. [ 0.1023, 0.0534, 0.6864, 0.2369, 0.4190, 0.1628, 0.5558, 0.5566]],
  16399. device='cuda:0', grad_fn=<AddmmBackward>)
  16400. landmarks are: tensor([[[0.6203, 0.4078, 0.8800, 0.5083, 0.3900, 0.5000, 0.6100, 0.5583],
  16401. [0.6154, 0.4117, 0.8687, 0.5167, 0.4895, 0.5647, 0.5524, 0.5136],
  16402. [0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5838, 0.5317],
  16403. [0.6179, 0.3998, 0.8396, 0.3505, 0.3552, 0.3768, 0.6064, 0.5329],
  16404. [0.6270, 0.4266, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  16405. [0.0000, 0.0000, 0.8525, 0.2217, 0.5412, 0.2367, 0.7367, 0.5482],
  16406. [0.6151, 0.4125, 0.8737, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  16407. [0.0000, 0.0000, 0.6469, 0.1943, 0.4025, 0.2000, 0.5125, 0.5533]]],
  16408. device='cuda:0')
  16409. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  16410. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  16411. loss_train: 0.03243761681369506
  16412. step: 28
  16413. running loss: 0.0011584863147748234
  16414. Train Steps: 28/90 Loss: 0.0012 torch.Size([8, 600, 800])
  16415. torch.Size([8, 8])
  16416. tensor([[0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  16417. [0.6208, 0.4082, 0.8538, 0.3067, 0.3588, 0.3717, 0.6112, 0.5517],
  16418. [0.6076, 0.3953, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954],
  16419. [0.6332, 0.4118, 0.9238, 0.4267, 0.4012, 0.4733, 0.7525, 0.5436],
  16420. [0.6272, 0.4071, 0.8738, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  16421. [0.6296, 0.4076, 0.8400, 0.5583, 0.3700, 0.4367, 0.6876, 0.5494],
  16422. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  16423. [0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767]],
  16424. device='cuda:0', dtype=torch.float64)
  16425. predictions are: tensor([[0.5333, 0.3519, 0.8224, 0.5819, 0.4291, 0.4459, 0.5833, 0.5994],
  16426. [0.5432, 0.3570, 0.8515, 0.3269, 0.3724, 0.3711, 0.6106, 0.5296],
  16427. [0.5232, 0.3436, 0.8307, 0.3767, 0.3578, 0.3805, 0.5443, 0.4934],
  16428. [0.6259, 0.4124, 0.9127, 0.4304, 0.4193, 0.4777, 0.7340, 0.5407],
  16429. [0.5638, 0.3596, 0.8785, 0.5704, 0.3951, 0.3730, 0.6073, 0.4628],
  16430. [0.5675, 0.3640, 0.8701, 0.5401, 0.4035, 0.4203, 0.6903, 0.5244],
  16431. [0.5548, 0.3612, 0.8543, 0.4221, 0.3882, 0.4384, 0.5354, 0.5408],
  16432. [0.5765, 0.3649, 0.8960, 0.4664, 0.4072, 0.3573, 0.5822, 0.5698]],
  16433. device='cuda:0', grad_fn=<AddmmBackward>)
  16434. landmarks are: tensor([[[0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  16435. [0.6208, 0.4082, 0.8537, 0.3067, 0.3587, 0.3717, 0.6112, 0.5517],
  16436. [0.6076, 0.3952, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954],
  16437. [0.6332, 0.4118, 0.9237, 0.4267, 0.4013, 0.4733, 0.7525, 0.5436],
  16438. [0.6272, 0.4071, 0.8737, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  16439. [0.6296, 0.4076, 0.8400, 0.5583, 0.3700, 0.4367, 0.6876, 0.5494],
  16440. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  16441. [0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767]]],
  16442. device='cuda:0')
  16443. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  16444. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  16445. loss_train: 0.03350451713777147
  16446. step: 29
  16447. running loss: 0.0011553281771645334
  16448. Train Steps: 29/90 Loss: 0.0012 torch.Size([8, 600, 800])
  16449. torch.Size([8, 8])
  16450. tensor([[0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6038, 0.6167],
  16451. [0.6317, 0.4038, 0.8287, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  16452. [0.6200, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  16453. [0.6261, 0.4029, 0.8720, 0.3364, 0.3665, 0.3753, 0.6531, 0.5183],
  16454. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  16455. [0.6097, 0.4000, 0.7325, 0.2667, 0.3450, 0.3517, 0.5284, 0.5045],
  16456. [0.6219, 0.4114, 0.8175, 0.2817, 0.3925, 0.2783, 0.5900, 0.5350],
  16457. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123]],
  16458. device='cuda:0', dtype=torch.float64)
  16459. predictions are: tensor([[0.5300, 0.3320, 0.8162, 0.3578, 0.3960, 0.3841, 0.5884, 0.6032],
  16460. [0.5531, 0.3594, 0.8705, 0.6327, 0.3978, 0.4552, 0.6268, 0.5013],
  16461. [0.6140, 0.3880, 0.8141, 0.3311, 0.4078, 0.2734, 0.5962, 0.5032],
  16462. [0.5975, 0.3754, 0.8768, 0.3859, 0.3943, 0.3580, 0.6395, 0.5108],
  16463. [0.5600, 0.3510, 0.8141, 0.3194, 0.3890, 0.3210, 0.6084, 0.5404],
  16464. [0.5502, 0.3410, 0.7539, 0.3059, 0.3619, 0.3273, 0.5426, 0.5130],
  16465. [0.5694, 0.3608, 0.8149, 0.3133, 0.4233, 0.2520, 0.5978, 0.5442],
  16466. [0.5955, 0.3840, 0.7417, 0.2904, 0.3905, 0.2555, 0.5303, 0.5149]],
  16467. device='cuda:0', grad_fn=<AddmmBackward>)
  16468. landmarks are: tensor([[[0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6037, 0.6167],
  16469. [0.6317, 0.4038, 0.8288, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  16470. [0.6201, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  16471. [0.6261, 0.4029, 0.8720, 0.3364, 0.3665, 0.3753, 0.6531, 0.5183],
  16472. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  16473. [0.6097, 0.4000, 0.7325, 0.2667, 0.3450, 0.3517, 0.5284, 0.5045],
  16474. [0.6219, 0.4114, 0.8175, 0.2817, 0.3925, 0.2783, 0.5900, 0.5350],
  16475. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123]]],
  16476. device='cuda:0')
  16477. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  16478. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  16479. loss_train: 0.03453243928379379
  16480. step: 30
  16481. running loss: 0.001151081309459793
  16482.  
  16483. Train Steps: 30/90 Loss: 0.0012 torch.Size([8, 600, 800])
  16484. torch.Size([8, 8])
  16485. tensor([[0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  16486. [0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  16487. [0.6064, 0.3953, 0.8738, 0.4417, 0.3663, 0.4683, 0.5511, 0.5416],
  16488. [0.6272, 0.4045, 0.8538, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  16489. [ nan, nan, 0.8850, 0.2817, 0.5112, 0.2183, 0.7184, 0.5436],
  16490. [0.6357, 0.4097, 0.9038, 0.3883, 0.4213, 0.2950, 0.6686, 0.5390],
  16491. [0.6236, 0.3966, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  16492. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5837, 0.5583]],
  16493. device='cuda:0', dtype=torch.float64)
  16494. predictions are: tensor([[0.6427, 0.4202, 0.8314, 0.6064, 0.3901, 0.5030, 0.5913, 0.4913],
  16495. [0.6279, 0.4157, 0.8776, 0.5284, 0.3637, 0.4915, 0.5434, 0.5713],
  16496. [0.6431, 0.4021, 0.8803, 0.4585, 0.3645, 0.4955, 0.5290, 0.5244],
  16497. [0.6660, 0.4271, 0.8366, 0.5968, 0.3659, 0.4531, 0.5882, 0.4971],
  16498. [0.2643, 0.1526, 0.8646, 0.2992, 0.5143, 0.2550, 0.6820, 0.5459],
  16499. [0.6371, 0.4096, 0.8860, 0.3984, 0.3991, 0.3012, 0.6437, 0.5549],
  16500. [0.6351, 0.4026, 0.8868, 0.4923, 0.3529, 0.4365, 0.5662, 0.5320],
  16501. [0.6200, 0.4089, 0.8712, 0.4522, 0.4741, 0.5675, 0.5527, 0.5659]],
  16502. device='cuda:0', grad_fn=<AddmmBackward>)
  16503. landmarks are: tensor([[[0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  16504. [0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  16505. [0.6064, 0.3952, 0.8737, 0.4417, 0.3663, 0.4683, 0.5511, 0.5416],
  16506. [0.6271, 0.4045, 0.8537, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  16507. [0.0000, 0.0000, 0.8850, 0.2817, 0.5113, 0.2183, 0.7184, 0.5436],
  16508. [0.6357, 0.4097, 0.9038, 0.3883, 0.4212, 0.2950, 0.6686, 0.5390],
  16509. [0.6236, 0.3965, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  16510. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5838, 0.5583]]],
  16511. device='cuda:0')
  16512. loss_train_step before backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  16513. loss_train_step after backward: tensor(0.0018, device='cuda:0', grad_fn=<MseLossBackward>)
  16514. loss_train: 0.03629230541992001
  16515. step: 31
  16516. running loss: 0.0011707195296748391
  16517. Train Steps: 31/90 Loss: 0.0012 torch.Size([8, 600, 800])
  16518. torch.Size([8, 8])
  16519. tensor([[0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  16520. [0.6127, 0.4066, 0.8550, 0.5567, 0.4662, 0.5141, 0.5070, 0.5412],
  16521. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  16522. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  16523. [0.6159, 0.4085, 0.6900, 0.2283, 0.4088, 0.1950, 0.5123, 0.5397],
  16524. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  16525. [0.6055, 0.4015, 0.7425, 0.2033, 0.4113, 0.1883, 0.5217, 0.4823],
  16526. [0.6110, 0.4047, 0.8700, 0.4483, 0.3713, 0.3967, 0.5088, 0.5517]],
  16527. device='cuda:0', dtype=torch.float64)
  16528. predictions are: tensor([[0.6312, 0.4087, 0.8552, 0.2854, 0.4418, 0.2874, 0.6899, 0.5743],
  16529. [0.6343, 0.4257, 0.8424, 0.5669, 0.4541, 0.5228, 0.5158, 0.5439],
  16530. [0.5952, 0.3849, 0.8683, 0.5273, 0.4134, 0.5349, 0.5800, 0.5326],
  16531. [0.6040, 0.3875, 0.8703, 0.3866, 0.3464, 0.3959, 0.5834, 0.5296],
  16532. [0.5615, 0.3480, 0.6886, 0.2595, 0.4079, 0.2172, 0.4974, 0.5599],
  16533. [0.6166, 0.3884, 0.8692, 0.4099, 0.3551, 0.4637, 0.5817, 0.5406],
  16534. [0.5760, 0.3547, 0.7143, 0.2348, 0.3887, 0.2061, 0.5262, 0.5102],
  16535. [0.5566, 0.3679, 0.8671, 0.4734, 0.3623, 0.4301, 0.5038, 0.5463]],
  16536. device='cuda:0', grad_fn=<AddmmBackward>)
  16537. landmarks are: tensor([[[0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  16538. [0.6127, 0.4066, 0.8550, 0.5567, 0.4662, 0.5141, 0.5070, 0.5412],
  16539. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  16540. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  16541. [0.6159, 0.4085, 0.6900, 0.2283, 0.4087, 0.1950, 0.5123, 0.5397],
  16542. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  16543. [0.6055, 0.4015, 0.7425, 0.2033, 0.4112, 0.1883, 0.5217, 0.4823],
  16544. [0.6110, 0.4047, 0.8700, 0.4483, 0.3713, 0.3967, 0.5088, 0.5517]]],
  16545. device='cuda:0')
  16546. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  16547. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  16548. loss_train: 0.03677627723664045
  16549. step: 32
  16550. running loss: 0.0011492586636450142
  16551. Train Steps: 32/90 Loss: 0.0011 torch.Size([8, 600, 800])
  16552. torch.Size([8, 8])
  16553. tensor([[0.6125, 0.3999, 0.8750, 0.4883, 0.4750, 0.4700, 0.5533, 0.5617],
  16554. [0.6196, 0.4094, 0.7562, 0.2817, 0.3937, 0.3183, 0.6013, 0.6183],
  16555. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  16556. [0.6273, 0.4110, 0.8900, 0.3817, 0.4188, 0.2167, 0.5858, 0.4835],
  16557. [0.6068, 0.3963, 0.8650, 0.4317, 0.4037, 0.5083, 0.5253, 0.4999],
  16558. [0.6122, 0.4006, 0.8850, 0.4217, 0.4088, 0.5517, 0.6063, 0.5517],
  16559. [ nan, nan, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  16560. [0.6043, 0.4022, 0.6887, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136]],
  16561. device='cuda:0', dtype=torch.float64)
  16562. predictions are: tensor([[0.6957, 0.4481, 0.8929, 0.5009, 0.4728, 0.5124, 0.5399, 0.5542],
  16563. [0.6731, 0.4449, 0.7781, 0.3020, 0.3991, 0.3333, 0.6170, 0.6263],
  16564. [0.6312, 0.4144, 0.6878, 0.2701, 0.3600, 0.2677, 0.5420, 0.5699],
  16565. [0.6798, 0.4307, 0.9188, 0.4046, 0.4165, 0.2358, 0.5908, 0.4885],
  16566. [0.6334, 0.4017, 0.8684, 0.4303, 0.3911, 0.5226, 0.5477, 0.4909],
  16567. [0.6938, 0.4446, 0.8939, 0.4294, 0.4075, 0.5651, 0.5999, 0.5187],
  16568. [0.1302, 0.0652, 0.6944, 0.2737, 0.3742, 0.2653, 0.5407, 0.5516],
  16569. [0.6296, 0.4025, 0.6964, 0.2256, 0.3679, 0.2693, 0.5542, 0.5215]],
  16570. device='cuda:0', grad_fn=<AddmmBackward>)
  16571. landmarks are: tensor([[[0.6125, 0.3999, 0.8750, 0.4883, 0.4750, 0.4700, 0.5533, 0.5617],
  16572. [0.6196, 0.4094, 0.7563, 0.2817, 0.3938, 0.3183, 0.6012, 0.6183],
  16573. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  16574. [0.6273, 0.4110, 0.8900, 0.3817, 0.4187, 0.2167, 0.5858, 0.4835],
  16575. [0.6068, 0.3963, 0.8650, 0.4317, 0.4038, 0.5083, 0.5253, 0.4999],
  16576. [0.6122, 0.4006, 0.8850, 0.4217, 0.4087, 0.5517, 0.6062, 0.5517],
  16577. [0.0000, 0.0000, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  16578. [0.6043, 0.4022, 0.6888, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136]]],
  16579. device='cuda:0')
  16580. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  16581. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  16582. loss_train: 0.037722738285083324
  16583. step: 33
  16584. running loss: 0.0011431132813661613
  16585. Train Steps: 33/90 Loss: 0.0011 torch.Size([8, 600, 800])
  16586. torch.Size([8, 8])
  16587. tensor([[0.6273, 0.4100, 0.7137, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633],
  16588. [0.6212, 0.4033, 0.8938, 0.4167, 0.3813, 0.4267, 0.5613, 0.5583],
  16589. [0.6085, 0.4008, 0.8588, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  16590. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  16591. [0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533],
  16592. [0.6053, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  16593. [0.6200, 0.4071, 0.7338, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  16594. [0.6346, 0.4086, 0.7938, 0.5500, 0.3962, 0.4867, 0.7343, 0.5702]],
  16595. device='cuda:0', dtype=torch.float64)
  16596. predictions are: tensor([[0.6207, 0.4087, 0.7299, 0.2346, 0.3832, 0.2794, 0.5773, 0.5598],
  16597. [0.6340, 0.4120, 0.9302, 0.4444, 0.3620, 0.4391, 0.5457, 0.5405],
  16598. [0.6096, 0.4085, 0.8734, 0.5473, 0.5027, 0.5276, 0.4847, 0.5311],
  16599. [0.6392, 0.4208, 0.9139, 0.4806, 0.3919, 0.4799, 0.5381, 0.5347],
  16600. [0.6476, 0.4382, 0.7389, 0.2453, 0.3673, 0.3284, 0.5825, 0.5613],
  16601. [0.5024, 0.3274, 0.6944, 0.2168, 0.3967, 0.2365, 0.5230, 0.5248],
  16602. [0.5912, 0.4161, 0.7449, 0.2064, 0.4182, 0.2596, 0.5937, 0.5524],
  16603. [0.6630, 0.4463, 0.8269, 0.5501, 0.3866, 0.5200, 0.6644, 0.5626]],
  16604. device='cuda:0', grad_fn=<AddmmBackward>)
  16605. landmarks are: tensor([[[0.6273, 0.4099, 0.7138, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633],
  16606. [0.6212, 0.4033, 0.8938, 0.4167, 0.3812, 0.4267, 0.5612, 0.5583],
  16607. [0.6084, 0.4008, 0.8587, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  16608. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  16609. [0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533],
  16610. [0.6054, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  16611. [0.6200, 0.4071, 0.7337, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  16612. [0.6346, 0.4086, 0.7937, 0.5500, 0.3963, 0.4867, 0.7343, 0.5702]]],
  16613. device='cuda:0')
  16614. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  16615.  
  16616. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  16617. loss_train: 0.03848643408855423
  16618. step: 34
  16619. running loss: 0.001131953943781007
  16620. Train Steps: 34/90 Loss: 0.0011 torch.Size([8, 600, 800])
  16621. torch.Size([8, 8])
  16622. tensor([[0.6128, 0.4084, 0.8738, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  16623. [0.6251, 0.4163, 0.8662, 0.4467, 0.3625, 0.3567, 0.6038, 0.5533],
  16624. [0.6222, 0.3957, 0.8838, 0.5017, 0.3937, 0.4600, 0.5900, 0.5017],
  16625. [0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500],
  16626. [0.6224, 0.4097, 0.7438, 0.2267, 0.3850, 0.2850, 0.5988, 0.5250],
  16627. [0.6200, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  16628. [0.6030, 0.3969, 0.7988, 0.3917, 0.3450, 0.3667, 0.5266, 0.4700],
  16629. [0.6275, 0.4048, 0.8488, 0.2883, 0.4463, 0.2033, 0.6321, 0.5155]],
  16630. device='cuda:0', dtype=torch.float64)
  16631. predictions are: tensor([[0.6183, 0.4021, 0.8597, 0.4573, 0.3796, 0.3890, 0.5009, 0.5528],
  16632. [0.6546, 0.4281, 0.8510, 0.4200, 0.3696, 0.3607, 0.5795, 0.5611],
  16633. [0.6465, 0.4086, 0.8639, 0.4932, 0.3910, 0.4767, 0.5734, 0.5107],
  16634. [0.6153, 0.4060, 0.7664, 0.2453, 0.3581, 0.4111, 0.6184, 0.5636],
  16635. [0.6052, 0.4063, 0.7235, 0.2073, 0.3895, 0.3052, 0.5874, 0.5488],
  16636. [0.6432, 0.4107, 0.8695, 0.4533, 0.3793, 0.4444, 0.5860, 0.5381],
  16637. [0.6142, 0.4123, 0.8144, 0.3713, 0.3490, 0.3675, 0.5293, 0.5152],
  16638. [0.7097, 0.4605, 0.8296, 0.2603, 0.4559, 0.2119, 0.6177, 0.5250]],
  16639. device='cuda:0', grad_fn=<AddmmBackward>)
  16640. landmarks are: tensor([[[0.6127, 0.4084, 0.8737, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  16641. [0.6252, 0.4162, 0.8662, 0.4467, 0.3625, 0.3567, 0.6037, 0.5533],
  16642. [0.6222, 0.3957, 0.8838, 0.5017, 0.3938, 0.4600, 0.5900, 0.5017],
  16643. [0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500],
  16644. [0.6224, 0.4097, 0.7437, 0.2267, 0.3850, 0.2850, 0.5987, 0.5250],
  16645. [0.6201, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  16646. [0.6030, 0.3969, 0.7987, 0.3917, 0.3450, 0.3667, 0.5266, 0.4700],
  16647. [0.6275, 0.4048, 0.8487, 0.2883, 0.4462, 0.2033, 0.6321, 0.5155]]],
  16648. device='cuda:0')
  16649. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  16650. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  16651. loss_train: 0.03889687912305817
  16652. step: 35
  16653. running loss: 0.0011113394035159477
  16654. Train Steps: 35/90 Loss: 0.0011 torch.Size([8, 600, 800])
  16655. torch.Size([8, 8])
  16656. tensor([[0.6268, 0.4102, 0.8938, 0.3667, 0.4025, 0.2833, 0.6275, 0.5183],
  16657. [0.6186, 0.4060, 0.8750, 0.5050, 0.3538, 0.4367, 0.5813, 0.6083],
  16658. [0.6138, 0.4054, 0.8750, 0.4750, 0.4363, 0.5017, 0.5086, 0.5822],
  16659. [0.6147, 0.4081, 0.8538, 0.3400, 0.3663, 0.3150, 0.5142, 0.4875],
  16660. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  16661. [0.6053, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  16662. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  16663. [0.6339, 0.4102, 0.8588, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089]],
  16664. device='cuda:0', dtype=torch.float64)
  16665. predictions are: tensor([[0.6272, 0.4205, 0.8835, 0.3393, 0.3981, 0.2917, 0.6411, 0.5323],
  16666. [0.5420, 0.3742, 0.8473, 0.4490, 0.3551, 0.4554, 0.6004, 0.5777],
  16667. [0.6386, 0.4225, 0.8350, 0.4646, 0.4252, 0.4898, 0.5376, 0.5558],
  16668. [0.6313, 0.4140, 0.8614, 0.3493, 0.3677, 0.3204, 0.5472, 0.5091],
  16669. [0.5479, 0.3567, 0.8535, 0.3825, 0.3773, 0.4866, 0.6150, 0.5179],
  16670. [0.5576, 0.3567, 0.6646, 0.1769, 0.3972, 0.2270, 0.5550, 0.5160],
  16671. [0.6751, 0.4624, 0.7140, 0.2613, 0.4381, 0.2245, 0.5698, 0.5634],
  16672. [0.6748, 0.4515, 0.8537, 0.2882, 0.4649, 0.2031, 0.6719, 0.5120]],
  16673. device='cuda:0', grad_fn=<AddmmBackward>)
  16674. landmarks are: tensor([[[0.6268, 0.4102, 0.8938, 0.3667, 0.4025, 0.2833, 0.6275, 0.5183],
  16675. [0.6186, 0.4060, 0.8750, 0.5050, 0.3537, 0.4367, 0.5813, 0.6083],
  16676. [0.6138, 0.4054, 0.8750, 0.4750, 0.4363, 0.5017, 0.5086, 0.5822],
  16677. [0.6147, 0.4081, 0.8537, 0.3400, 0.3663, 0.3150, 0.5142, 0.4875],
  16678. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  16679. [0.6054, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  16680. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  16681. [0.6339, 0.4102, 0.8587, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089]]],
  16682. device='cuda:0')
  16683. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  16684. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  16685. loss_train: 0.03962359408615157
  16686. step: 36
  16687. running loss: 0.001100655391281988
  16688. Train Steps: 36/90 Loss: 0.0011 torch.Size([8, 600, 800])
  16689. torch.Size([8, 8])
  16690. tensor([[0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5463, 0.5800],
  16691. [0.6128, 0.4084, 0.8738, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  16692. [0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6262, 0.5167],
  16693. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  16694. [0.6200, 0.4118, 0.8287, 0.4017, 0.3775, 0.2833, 0.5391, 0.5799],
  16695. [0.6202, 0.4079, 0.8025, 0.2500, 0.3763, 0.3217, 0.6125, 0.5533],
  16696. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  16697. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783]],
  16698. device='cuda:0', dtype=torch.float64)
  16699. predictions are: tensor([[0.6731, 0.4660, 0.7442, 0.2541, 0.3794, 0.2471, 0.5337, 0.5815],
  16700. [0.5771, 0.3816, 0.8796, 0.4264, 0.3821, 0.3529, 0.5279, 0.5435],
  16701. [0.5705, 0.3869, 0.9043, 0.3565, 0.3812, 0.3285, 0.6388, 0.5319],
  16702. [0.5005, 0.3500, 0.8975, 0.4178, 0.3760, 0.4214, 0.6897, 0.5422],
  16703. [0.5129, 0.3521, 0.8153, 0.3398, 0.3702, 0.2708, 0.5315, 0.5734],
  16704. [0.5919, 0.4103, 0.7996, 0.1988, 0.3947, 0.3031, 0.6229, 0.5450],
  16705. [0.6546, 0.4317, 0.6917, 0.1848, 0.4577, 0.1434, 0.5603, 0.5350],
  16706. [0.6887, 0.4524, 0.8398, 0.3022, 0.3768, 0.2874, 0.5113, 0.5499]],
  16707. device='cuda:0', grad_fn=<AddmmBackward>)
  16708. landmarks are: tensor([[[0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5462, 0.5800],
  16709. [0.6127, 0.4084, 0.8737, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  16710. [0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6263, 0.5167],
  16711. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  16712. [0.6200, 0.4118, 0.8288, 0.4017, 0.3775, 0.2833, 0.5391, 0.5799],
  16713. [0.6202, 0.4079, 0.8025, 0.2500, 0.3762, 0.3217, 0.6125, 0.5533],
  16714. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  16715. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783]]],
  16716. device='cuda:0')
  16717. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  16718. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  16719. loss_train: 0.04108484456082806
  16720. step: 37
  16721. running loss: 0.0011104012043467043
  16722. Train Steps: 37/90 Loss: 0.0011 torch.Size([8, 600, 800])
  16723. torch.Size([8, 8])
  16724. tensor([[0.6132, 0.4037, 0.6963, 0.2217, 0.4100, 0.1950, 0.5395, 0.5175],
  16725. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  16726. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  16727. [0.6243, 0.4128, 0.7762, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417],
  16728. [0.6293, 0.4024, 0.8750, 0.5000, 0.4012, 0.5733, 0.7121, 0.5633],
  16729. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  16730. [0.6109, 0.4041, 0.6975, 0.3167, 0.3513, 0.3383, 0.5153, 0.5319],
  16731. [0.6251, 0.4163, 0.8662, 0.4467, 0.3625, 0.3567, 0.6038, 0.5533]],
  16732. device='cuda:0', dtype=torch.float64)
  16733. predictions are: tensor([[0.7004, 0.4781, 0.6918, 0.2217, 0.4056, 0.1450, 0.5147, 0.5004],
  16734. [0.6227, 0.4156, 0.8690, 0.3062, 0.3745, 0.2790, 0.5618, 0.5130],
  16735. [0.5683, 0.3657, 0.9082, 0.3851, 0.4262, 0.3278, 0.6973, 0.5923],
  16736. [0.6048, 0.4049, 0.7707, 0.2541, 0.3962, 0.2739, 0.6194, 0.5442],
  16737. [0.5766, 0.3658, 0.8741, 0.4772, 0.4067, 0.5390, 0.6700, 0.5635],
  16738. [0.6280, 0.4125, 0.8869, 0.4231, 0.3597, 0.2912, 0.6195, 0.4975],
  16739. [0.5745, 0.3832, 0.7137, 0.2926, 0.3510, 0.3062, 0.5140, 0.5181],
  16740. [0.6141, 0.4022, 0.8545, 0.4257, 0.3587, 0.3034, 0.5714, 0.5359]],
  16741. device='cuda:0', grad_fn=<AddmmBackward>)
  16742. landmarks are: tensor([[[0.6132, 0.4037, 0.6963, 0.2217, 0.4100, 0.1950, 0.5395, 0.5175],
  16743. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  16744. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  16745. [0.6243, 0.4128, 0.7763, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417],
  16746. [0.6293, 0.4024, 0.8750, 0.5000, 0.4013, 0.5733, 0.7121, 0.5633],
  16747. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  16748. [0.6109, 0.4041, 0.6975, 0.3167, 0.3512, 0.3383, 0.5153, 0.5319],
  16749. [0.6252, 0.4162, 0.8662, 0.4467, 0.3625, 0.3567, 0.6037, 0.5533]]],
  16750. device='cuda:0')
  16751. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  16752. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  16753. loss_train: 0.04196064005373046
  16754. step: 38
  16755. running loss: 0.001104227369835012
  16756.  
  16757. Train Steps: 38/90 Loss: 0.0011 torch.Size([8, 600, 800])
  16758. torch.Size([8, 8])
  16759. tensor([[0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274],
  16760. [0.6059, 0.4002, 0.7562, 0.2767, 0.3538, 0.3033, 0.5529, 0.5455],
  16761. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  16762. [0.6196, 0.4088, 0.8888, 0.4583, 0.4500, 0.5683, 0.6138, 0.5883],
  16763. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  16764. [0.6138, 0.4054, 0.8750, 0.4750, 0.4363, 0.5017, 0.5086, 0.5822],
  16765. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6063, 0.6083],
  16766. [0.6189, 0.4029, 0.8375, 0.5767, 0.4745, 0.4829, 0.5551, 0.5598]],
  16767. device='cuda:0', dtype=torch.float64)
  16768. predictions are: tensor([[0.6322, 0.4343, 0.8624, 0.3162, 0.4473, 0.1855, 0.6437, 0.4917],
  16769. [0.6319, 0.4302, 0.7483, 0.2713, 0.3327, 0.2648, 0.5815, 0.5005],
  16770. [0.6248, 0.3952, 0.8612, 0.4800, 0.4064, 0.4749, 0.7014, 0.5224],
  16771. [0.6160, 0.4161, 0.8835, 0.4645, 0.4237, 0.5160, 0.5874, 0.5665],
  16772. [0.6446, 0.4308, 0.6868, 0.2093, 0.3712, 0.1665, 0.5641, 0.5247],
  16773. [0.6253, 0.4151, 0.8410, 0.4777, 0.4036, 0.4330, 0.5291, 0.5450],
  16774. [0.5632, 0.3792, 0.8631, 0.3644, 0.3287, 0.3986, 0.6136, 0.5751],
  16775. [0.6078, 0.4056, 0.8220, 0.5465, 0.4432, 0.4151, 0.5733, 0.5377]],
  16776. device='cuda:0', grad_fn=<AddmmBackward>)
  16777. landmarks are: tensor([[[0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274],
  16778. [0.6059, 0.4002, 0.7563, 0.2767, 0.3537, 0.3033, 0.5529, 0.5455],
  16779. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  16780. [0.6196, 0.4088, 0.8888, 0.4583, 0.4500, 0.5683, 0.6137, 0.5883],
  16781. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  16782. [0.6138, 0.4054, 0.8750, 0.4750, 0.4363, 0.5017, 0.5086, 0.5822],
  16783. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6062, 0.6083],
  16784. [0.6189, 0.4029, 0.8375, 0.5767, 0.4745, 0.4829, 0.5551, 0.5598]]],
  16785. device='cuda:0')
  16786. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  16787. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  16788. loss_train: 0.042734951653983444
  16789. step: 39
  16790. running loss: 0.0010957679911277806
  16791. Train Steps: 39/90 Loss: 0.0011 torch.Size([8, 600, 800])
  16792. torch.Size([8, 8])
  16793. tensor([[0.6289, 0.4019, 0.8113, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332],
  16794. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  16795. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5463, 0.5800],
  16796. [0.6161, 0.4024, 0.8838, 0.4583, 0.3688, 0.3733, 0.5311, 0.5344],
  16797. [0.6305, 0.3983, 0.8950, 0.4833, 0.3688, 0.4683, 0.6375, 0.5117],
  16798. [0.6085, 0.4005, 0.8400, 0.4317, 0.3763, 0.4750, 0.5476, 0.5058],
  16799. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6188, 0.5283],
  16800. [0.6307, 0.4029, 0.8988, 0.4817, 0.3937, 0.3500, 0.7311, 0.5378]],
  16801. device='cuda:0', dtype=torch.float64)
  16802. predictions are: tensor([[0.6414, 0.4290, 0.8107, 0.5311, 0.3755, 0.4731, 0.6912, 0.5280],
  16803. [0.6003, 0.3886, 0.7825, 0.4177, 0.3560, 0.4139, 0.5226, 0.5284],
  16804. [0.6446, 0.4353, 0.7344, 0.3098, 0.3624, 0.2710, 0.5068, 0.5727],
  16805. [0.5883, 0.3823, 0.8884, 0.4842, 0.3630, 0.3670, 0.5167, 0.5230],
  16806. [0.6610, 0.4082, 0.8864, 0.4826, 0.3667, 0.4435, 0.6328, 0.5101],
  16807. [0.6120, 0.3974, 0.8367, 0.4157, 0.3598, 0.4524, 0.5596, 0.4991],
  16808. [0.7246, 0.4780, 0.9052, 0.3613, 0.3900, 0.2460, 0.6239, 0.5306],
  16809. [0.6220, 0.4103, 0.8976, 0.4861, 0.3990, 0.3335, 0.7187, 0.5409]],
  16810. device='cuda:0', grad_fn=<AddmmBackward>)
  16811. landmarks are: tensor([[[0.6289, 0.4019, 0.8112, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332],
  16812. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  16813. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5462, 0.5800],
  16814. [0.6161, 0.4024, 0.8838, 0.4583, 0.3688, 0.3733, 0.5311, 0.5344],
  16815. [0.6305, 0.3983, 0.8950, 0.4833, 0.3688, 0.4683, 0.6375, 0.5117],
  16816. [0.6084, 0.4005, 0.8400, 0.4317, 0.3762, 0.4750, 0.5476, 0.5058],
  16817. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6187, 0.5283],
  16818. [0.6307, 0.4029, 0.8988, 0.4817, 0.3938, 0.3500, 0.7311, 0.5378]]],
  16819. device='cuda:0')
  16820. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  16821. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  16822. loss_train: 0.0431966645992361
  16823. step: 40
  16824. running loss: 0.0010799166149809026
  16825. Train Steps: 40/90 Loss: 0.0011 torch.Size([8, 600, 800])
  16826. torch.Size([8, 8])
  16827. tensor([[0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  16828. [0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5837, 0.5317],
  16829. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  16830. [0.6229, 0.4066, 0.8513, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  16831. [0.6321, 0.4048, 0.8738, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  16832. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  16833. [0.6255, 0.4017, 0.8688, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  16834. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6063, 0.5617]],
  16835. device='cuda:0', dtype=torch.float64)
  16836. predictions are: tensor([[0.6134, 0.3949, 0.8595, 0.3987, 0.3494, 0.3724, 0.5910, 0.5731],
  16837. [0.6684, 0.4354, 0.8558, 0.3522, 0.3534, 0.3221, 0.6058, 0.5432],
  16838. [0.6412, 0.4086, 0.8021, 0.3002, 0.3626, 0.3407, 0.6261, 0.5365],
  16839. [0.6907, 0.4500, 0.8277, 0.5763, 0.4382, 0.4893, 0.5934, 0.5552],
  16840. [0.6440, 0.4196, 0.8511, 0.5767, 0.3726, 0.4335, 0.6587, 0.4990],
  16841. [0.6427, 0.4119, 0.8444, 0.4761, 0.3769, 0.4700, 0.5662, 0.5726],
  16842. [0.6348, 0.3972, 0.8611, 0.3470, 0.3555, 0.3311, 0.6533, 0.5098],
  16843. [0.6770, 0.4490, 0.8951, 0.5009, 0.3560, 0.4613, 0.6049, 0.5530]],
  16844. device='cuda:0', grad_fn=<AddmmBackward>)
  16845. landmarks are: tensor([[[0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  16846. [0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5838, 0.5317],
  16847. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  16848. [0.6229, 0.4066, 0.8512, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  16849. [0.6321, 0.4048, 0.8737, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  16850. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  16851. [0.6255, 0.4017, 0.8687, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  16852. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6062, 0.5617]]],
  16853. device='cuda:0')
  16854. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  16855. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  16856. loss_train: 0.04362316819606349
  16857. step: 41
  16858. running loss: 0.0010639797120991095
  16859. Train Steps: 41/90 Loss: 0.0011 torch.Size([8, 600, 800])
  16860. torch.Size([8, 8])
  16861. tensor([[0.6164, 0.4119, 0.7913, 0.2650, 0.3538, 0.3500, 0.5614, 0.5038],
  16862. [0.6192, 0.4128, 0.8513, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  16863. [0.6274, 0.4099, 0.8625, 0.3233, 0.4400, 0.1983, 0.5876, 0.4869],
  16864. [0.6198, 0.4115, 0.7762, 0.2717, 0.3713, 0.3200, 0.5837, 0.5683],
  16865. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290],
  16866. [0.6128, 0.4084, 0.8738, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  16867. [0.6203, 0.4021, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  16868. [0.6222, 0.4169, 0.8638, 0.5650, 0.4313, 0.4783, 0.5637, 0.5633]],
  16869. device='cuda:0', dtype=torch.float64)
  16870. predictions are: tensor([[0.5952, 0.3894, 0.7800, 0.2822, 0.3461, 0.3827, 0.6112, 0.5220],
  16871. [0.6692, 0.4349, 0.8421, 0.5854, 0.4277, 0.5497, 0.5884, 0.5794],
  16872. [0.7066, 0.4452, 0.8616, 0.3625, 0.4512, 0.2253, 0.6105, 0.5051],
  16873. [0.6223, 0.3899, 0.7771, 0.2893, 0.3718, 0.3561, 0.5877, 0.5753],
  16874. [0.5444, 0.3587, 0.7548, 0.2279, 0.4270, 0.2348, 0.6474, 0.5424],
  16875. [0.6348, 0.4049, 0.8640, 0.4997, 0.3731, 0.3949, 0.5174, 0.5608],
  16876. [0.6215, 0.3988, 0.8642, 0.5334, 0.3727, 0.4115, 0.6142, 0.5432],
  16877. [0.6647, 0.4366, 0.8561, 0.5870, 0.4228, 0.4839, 0.5994, 0.5959]],
  16878. device='cuda:0', grad_fn=<AddmmBackward>)
  16879. landmarks are: tensor([[[0.6164, 0.4119, 0.7912, 0.2650, 0.3537, 0.3500, 0.5614, 0.5038],
  16880. [0.6192, 0.4128, 0.8512, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  16881. [0.6274, 0.4099, 0.8625, 0.3233, 0.4400, 0.1983, 0.5876, 0.4869],
  16882. [0.6198, 0.4115, 0.7763, 0.2717, 0.3713, 0.3200, 0.5838, 0.5683],
  16883. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290],
  16884. [0.6127, 0.4084, 0.8737, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  16885. [0.6203, 0.4020, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  16886. [0.6222, 0.4169, 0.8637, 0.5650, 0.4313, 0.4783, 0.5638, 0.5633]]],
  16887. device='cuda:0')
  16888. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  16889. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  16890. loss_train: 0.04430640529608354
  16891. step: 42
  16892. running loss: 0.001054914411811513
  16893.  
  16894. Train Steps: 42/90 Loss: 0.0011 torch.Size([8, 600, 800])
  16895. torch.Size([8, 8])
  16896. tensor([[0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  16897. [0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5413, 0.5433],
  16898. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6063, 0.5617],
  16899. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  16900. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5637, 0.5633],
  16901. [0.6087, 0.3976, 0.8337, 0.3867, 0.3713, 0.3117, 0.5938, 0.5300],
  16902. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  16903. [0.6271, 0.4040, 0.9138, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413]],
  16904. device='cuda:0', dtype=torch.float64)
  16905. predictions are: tensor([[0.6467, 0.4283, 0.8193, 0.5591, 0.4115, 0.5895, 0.7145, 0.5837],
  16906. [0.6037, 0.3921, 0.8437, 0.3812, 0.3790, 0.3736, 0.5460, 0.5361],
  16907. [0.6682, 0.4433, 0.8867, 0.5180, 0.3720, 0.4951, 0.5859, 0.5464],
  16908. [0.6078, 0.4041, 0.8402, 0.3575, 0.3668, 0.4851, 0.6394, 0.5646],
  16909. [0.6845, 0.4671, 0.8428, 0.5293, 0.3853, 0.4209, 0.5545, 0.5740],
  16910. [0.6057, 0.3982, 0.8355, 0.4236, 0.3830, 0.3535, 0.5858, 0.5416],
  16911. [0.5911, 0.3794, 0.8517, 0.4044, 0.3831, 0.3585, 0.5884, 0.5192],
  16912. [0.6004, 0.3727, 0.9194, 0.3995, 0.4778, 0.3034, 0.7175, 0.5427]],
  16913. device='cuda:0', grad_fn=<AddmmBackward>)
  16914. landmarks are: tensor([[[0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  16915. [0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5412, 0.5433],
  16916. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6062, 0.5617],
  16917. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  16918. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5638, 0.5633],
  16919. [0.6087, 0.3976, 0.8338, 0.3867, 0.3713, 0.3117, 0.5938, 0.5300],
  16920. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  16921. [0.6271, 0.4040, 0.9137, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413]]],
  16922. device='cuda:0')
  16923. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  16924. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  16925. loss_train: 0.044957086036447436
  16926. step: 43
  16927. running loss: 0.0010455136287545914
  16928. Train Steps: 43/90 Loss: 0.0010 torch.Size([8, 600, 800])
  16929. torch.Size([8, 8])
  16930. tensor([[0.6201, 0.4151, 0.8588, 0.5467, 0.3700, 0.3950, 0.5637, 0.5933],
  16931. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  16932. [0.6250, 0.4054, 0.8770, 0.4723, 0.4662, 0.5367, 0.6162, 0.5433],
  16933. [0.6124, 0.4075, 0.7696, 0.4153, 0.3475, 0.3767, 0.5157, 0.5427],
  16934. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  16935. [0.6263, 0.4233, 0.7924, 0.4626, 0.3788, 0.2883, 0.5573, 0.6047],
  16936. [0.6204, 0.4110, 0.7913, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367],
  16937. [0.6102, 0.4020, 0.8638, 0.3717, 0.3625, 0.5017, 0.6038, 0.5500]],
  16938. device='cuda:0', dtype=torch.float64)
  16939. predictions are: tensor([[0.6396, 0.4144, 0.8675, 0.5608, 0.3827, 0.4323, 0.5664, 0.5795],
  16940. [0.5833, 0.3797, 0.8016, 0.2433, 0.3960, 0.2990, 0.6173, 0.5316],
  16941. [0.6888, 0.4475, 0.8880, 0.4995, 0.4760, 0.5530, 0.6200, 0.5388],
  16942. [0.6149, 0.3904, 0.8058, 0.4255, 0.3589, 0.4215, 0.5320, 0.5407],
  16943. [0.6182, 0.4064, 0.8864, 0.5289, 0.4396, 0.5638, 0.6034, 0.5243],
  16944. [0.6576, 0.4455, 0.8199, 0.5035, 0.4000, 0.3319, 0.5725, 0.6120],
  16945. [0.5834, 0.3808, 0.8157, 0.2679, 0.4248, 0.2818, 0.6214, 0.5289],
  16946. [0.6289, 0.4154, 0.8847, 0.3929, 0.3923, 0.5196, 0.6435, 0.5359]],
  16947. device='cuda:0', grad_fn=<AddmmBackward>)
  16948. landmarks are: tensor([[[0.6202, 0.4151, 0.8587, 0.5467, 0.3700, 0.3950, 0.5638, 0.5933],
  16949. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  16950. [0.6250, 0.4054, 0.8770, 0.4723, 0.4663, 0.5367, 0.6162, 0.5433],
  16951. [0.6124, 0.4075, 0.7696, 0.4153, 0.3475, 0.3767, 0.5157, 0.5427],
  16952. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  16953. [0.6263, 0.4232, 0.7924, 0.4626, 0.3787, 0.2883, 0.5573, 0.6047],
  16954. [0.6204, 0.4110, 0.7912, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367],
  16955. [0.6102, 0.4020, 0.8637, 0.3717, 0.3625, 0.5017, 0.6037, 0.5500]]],
  16956. device='cuda:0')
  16957. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  16958. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  16959. loss_train: 0.04550424014450982
  16960. step: 44
  16961. running loss: 0.0010341872760115868
  16962. Train Steps: 44/90 Loss: 0.0010 torch.Size([8, 600, 800])
  16963. torch.Size([8, 8])
  16964. tensor([[0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6262, 0.5167],
  16965. [ nan, nan, 0.7725, 0.2611, 0.3675, 0.2733, 0.5413, 0.5167],
  16966. [0.6127, 0.4115, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  16967. [0.6048, 0.3928, 0.8538, 0.5433, 0.3875, 0.5117, 0.5266, 0.4719],
  16968. [0.6275, 0.4157, 0.8337, 0.5800, 0.3763, 0.4200, 0.5547, 0.6125],
  16969. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  16970. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  16971. [0.6087, 0.3976, 0.8337, 0.3867, 0.3713, 0.3117, 0.5938, 0.5300]],
  16972. device='cuda:0', dtype=torch.float64)
  16973. predictions are: tensor([[0.6898, 0.4383, 0.9413, 0.4102, 0.4053, 0.3490, 0.6271, 0.5214],
  16974. [0.0654, 0.0371, 0.8057, 0.2584, 0.3954, 0.3384, 0.5145, 0.5398],
  16975. [0.5854, 0.3809, 0.7735, 0.2698, 0.3950, 0.3140, 0.5210, 0.5573],
  16976. [0.7269, 0.4606, 0.8708, 0.5288, 0.4172, 0.5184, 0.5386, 0.5147],
  16977. [0.7218, 0.4794, 0.8484, 0.5600, 0.4143, 0.4491, 0.5731, 0.6127],
  16978. [0.6632, 0.4350, 0.8738, 0.5665, 0.3998, 0.5053, 0.6562, 0.5334],
  16979. [0.6927, 0.4557, 0.8649, 0.5187, 0.4277, 0.5654, 0.7123, 0.5794],
  16980. [0.6533, 0.4278, 0.8803, 0.3830, 0.3984, 0.3294, 0.5873, 0.5378]],
  16981. device='cuda:0', grad_fn=<AddmmBackward>)
  16982. landmarks are: tensor([[[0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6263, 0.5167],
  16983. [0.0000, 0.0000, 0.7725, 0.2611, 0.3675, 0.2733, 0.5412, 0.5167],
  16984. [0.6127, 0.4114, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  16985. [0.6048, 0.3928, 0.8537, 0.5433, 0.3875, 0.5117, 0.5266, 0.4719],
  16986. [0.6275, 0.4157, 0.8338, 0.5800, 0.3762, 0.4200, 0.5547, 0.6125],
  16987. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  16988. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  16989. [0.6087, 0.3976, 0.8338, 0.3867, 0.3713, 0.3117, 0.5938, 0.5300]]],
  16990. device='cuda:0')
  16991. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  16992. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  16993. loss_train: 0.04684005380840972
  16994. step: 45
  16995. running loss: 0.0010408900846313272
  16996. Train Steps: 45/90 Loss: 0.0010 torch.Size([8, 600, 800])
  16997. torch.Size([8, 8])
  16998. tensor([[ nan, nan, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  16999. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  17000. [0.6162, 0.4134, 0.6700, 0.2467, 0.3962, 0.2533, 0.5737, 0.5467],
  17001. [0.6222, 0.3957, 0.8838, 0.5017, 0.3937, 0.4600, 0.5900, 0.5017],
  17002. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  17003. [0.6199, 0.4071, 0.7600, 0.2117, 0.4037, 0.2767, 0.6138, 0.5550],
  17004. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5363, 0.5550],
  17005. [0.6165, 0.4106, 0.7575, 0.1733, 0.3838, 0.2650, 0.5680, 0.5116]],
  17006. device='cuda:0', dtype=torch.float64)
  17007. predictions are: tensor([[0.0948, 0.0649, 0.7369, 0.2828, 0.4063, 0.2761, 0.5456, 0.5900],
  17008. [0.6272, 0.3999, 0.7922, 0.2179, 0.4729, 0.1858, 0.5854, 0.5116],
  17009. [0.6376, 0.4229, 0.7174, 0.2835, 0.3986, 0.2754, 0.5660, 0.5660],
  17010. [0.6890, 0.4232, 0.9184, 0.5697, 0.4056, 0.4773, 0.5971, 0.5330],
  17011. [0.6022, 0.3880, 0.8296, 0.3167, 0.3761, 0.3408, 0.6021, 0.5541],
  17012. [0.6351, 0.4145, 0.8118, 0.2516, 0.4369, 0.2929, 0.6219, 0.5716],
  17013. [0.5690, 0.3789, 0.7400, 0.2806, 0.4204, 0.2384, 0.5463, 0.5773],
  17014. [0.6463, 0.4115, 0.7827, 0.2099, 0.3979, 0.2565, 0.5745, 0.5275]],
  17015. device='cuda:0', grad_fn=<AddmmBackward>)
  17016. landmarks are: tensor([[[0.0000, 0.0000, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  17017. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  17018. [0.6162, 0.4134, 0.6700, 0.2467, 0.3963, 0.2533, 0.5738, 0.5467],
  17019. [0.6222, 0.3957, 0.8838, 0.5017, 0.3938, 0.4600, 0.5900, 0.5017],
  17020. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  17021. [0.6199, 0.4071, 0.7600, 0.2117, 0.4038, 0.2767, 0.6137, 0.5550],
  17022. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5362, 0.5550],
  17023. [0.6165, 0.4106, 0.7575, 0.1733, 0.3837, 0.2650, 0.5680, 0.5116]]],
  17024. device='cuda:0')
  17025. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  17026. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  17027. loss_train: 0.04774374206317589
  17028. step: 46
  17029. running loss: 0.0010379074361559976
  17030.  
  17031. Train Steps: 46/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17032. torch.Size([8, 8])
  17033. tensor([[0.6200, 0.3961, 0.8461, 0.5497, 0.4142, 0.4577, 0.5892, 0.5402],
  17034. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092],
  17035. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  17036. [0.6263, 0.4039, 0.9000, 0.4400, 0.3625, 0.4667, 0.6424, 0.4804],
  17037. [0.6157, 0.4102, 0.8513, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  17038. [0.6193, 0.4034, 0.7757, 0.2347, 0.3733, 0.2919, 0.5930, 0.4926],
  17039. [0.6325, 0.4165, 0.9000, 0.4617, 0.3813, 0.4900, 0.7485, 0.5447],
  17040. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383]],
  17041. device='cuda:0', dtype=torch.float64)
  17042. predictions are: tensor([[0.6132, 0.3930, 0.8448, 0.5366, 0.3979, 0.4417, 0.5856, 0.5404],
  17043. [0.6065, 0.3869, 0.8564, 0.5038, 0.3883, 0.5033, 0.5811, 0.4995],
  17044. [0.6180, 0.4011, 0.8389, 0.5640, 0.3996, 0.4710, 0.5505, 0.6195],
  17045. [0.5635, 0.3521, 0.8894, 0.4405, 0.3446, 0.4432, 0.5997, 0.4692],
  17046. [0.5423, 0.3544, 0.8496, 0.3760, 0.3585, 0.3441, 0.4819, 0.5728],
  17047. [0.5672, 0.3738, 0.7683, 0.2288, 0.3706, 0.2683, 0.5741, 0.4807],
  17048. [0.5929, 0.3905, 0.9018, 0.4519, 0.3811, 0.4729, 0.7004, 0.5482],
  17049. [0.5863, 0.3843, 0.8827, 0.4588, 0.4425, 0.5560, 0.6066, 0.5455]],
  17050. device='cuda:0', grad_fn=<AddmmBackward>)
  17051. landmarks are: tensor([[[0.6200, 0.3961, 0.8461, 0.5497, 0.4142, 0.4577, 0.5892, 0.5402],
  17052. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092],
  17053. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  17054. [0.6263, 0.4038, 0.9000, 0.4400, 0.3625, 0.4667, 0.6424, 0.4804],
  17055. [0.6157, 0.4102, 0.8512, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  17056. [0.6193, 0.4034, 0.7757, 0.2347, 0.3733, 0.2919, 0.5930, 0.4926],
  17057. [0.6325, 0.4165, 0.9000, 0.4617, 0.3812, 0.4900, 0.7485, 0.5447],
  17058. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383]]],
  17059. device='cuda:0')
  17060. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  17061. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  17062. loss_train: 0.04828619211912155
  17063. step: 47
  17064. running loss: 0.0010273657897685437
  17065. Train Steps: 47/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17066. torch.Size([8, 8])
  17067. tensor([[0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  17068. [0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  17069. [0.6161, 0.4024, 0.8662, 0.4683, 0.4935, 0.5364, 0.6063, 0.5567],
  17070. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  17071. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  17072. [0.6204, 0.4049, 0.7975, 0.2700, 0.3937, 0.2567, 0.5700, 0.5183],
  17073. [0.6272, 0.4071, 0.8738, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  17074. [0.6222, 0.3937, 0.8350, 0.5617, 0.4138, 0.4600, 0.5800, 0.5233]],
  17075. device='cuda:0', dtype=torch.float64)
  17076. predictions are: tensor([[0.5316, 0.3312, 0.7556, 0.1395, 0.4007, 0.2116, 0.6300, 0.4945],
  17077. [0.5704, 0.3699, 0.8897, 0.3588, 0.4087, 0.3668, 0.6990, 0.5320],
  17078. [0.5642, 0.3640, 0.8545, 0.4413, 0.4406, 0.5011, 0.5789, 0.5432],
  17079. [0.5409, 0.3582, 0.8844, 0.4137, 0.3971, 0.3509, 0.6847, 0.5424],
  17080. [0.5854, 0.3841, 0.8754, 0.4690, 0.3467, 0.4304, 0.5603, 0.5154],
  17081. [0.5621, 0.3721, 0.7616, 0.2443, 0.3572, 0.2442, 0.5230, 0.5137],
  17082. [0.5795, 0.3681, 0.8496, 0.5353, 0.3407, 0.3531, 0.5610, 0.4714],
  17083. [0.5794, 0.3526, 0.7994, 0.5492, 0.3817, 0.4437, 0.5184, 0.5368]],
  17084. device='cuda:0', grad_fn=<AddmmBackward>)
  17085. landmarks are: tensor([[[0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  17086. [0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  17087. [0.6161, 0.4024, 0.8662, 0.4683, 0.4935, 0.5364, 0.6062, 0.5567],
  17088. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  17089. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  17090. [0.6204, 0.4049, 0.7975, 0.2700, 0.3938, 0.2567, 0.5700, 0.5183],
  17091. [0.6272, 0.4071, 0.8737, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  17092. [0.6222, 0.3937, 0.8350, 0.5617, 0.4137, 0.4600, 0.5800, 0.5233]]],
  17093. device='cuda:0')
  17094. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  17095. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  17096. loss_train: 0.04967287927865982
  17097. step: 48
  17098. running loss: 0.0010348516516387463
  17099. Train Steps: 48/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17100. torch.Size([8, 8])
  17101. tensor([[0.6163, 0.4114, 0.7650, 0.2017, 0.3763, 0.2867, 0.5631, 0.5071],
  17102. [0.6135, 0.4115, 0.8838, 0.4667, 0.4288, 0.6050, 0.5778, 0.5097],
  17103. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  17104. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5787, 0.5117],
  17105. [0.6132, 0.4118, 0.8200, 0.3633, 0.3563, 0.5400, 0.5787, 0.5136],
  17106. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  17107. [0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6262, 0.5367],
  17108. [0.6212, 0.4159, 0.8675, 0.5783, 0.4088, 0.4317, 0.5613, 0.5917]],
  17109. device='cuda:0', dtype=torch.float64)
  17110. predictions are: tensor([[0.5932, 0.3752, 0.7299, 0.1961, 0.3496, 0.2364, 0.5796, 0.4742],
  17111. [0.5746, 0.3886, 0.8618, 0.4477, 0.4209, 0.5416, 0.5816, 0.5256],
  17112. [0.5612, 0.3511, 0.7895, 0.2552, 0.3983, 0.1881, 0.5990, 0.5168],
  17113. [0.5476, 0.3570, 0.7310, 0.1807, 0.3870, 0.1791, 0.5720, 0.4853],
  17114. [0.5533, 0.3635, 0.7871, 0.3705, 0.3497, 0.4683, 0.5949, 0.5131],
  17115. [0.5736, 0.3728, 0.8460, 0.5015, 0.4372, 0.4897, 0.6130, 0.5065],
  17116. [0.5700, 0.3722, 0.8619, 0.4676, 0.4099, 0.5433, 0.6251, 0.5171],
  17117. [0.5712, 0.3966, 0.8282, 0.5571, 0.3881, 0.3772, 0.5808, 0.5863]],
  17118. device='cuda:0', grad_fn=<AddmmBackward>)
  17119. landmarks are: tensor([[[0.6163, 0.4114, 0.7650, 0.2017, 0.3762, 0.2867, 0.5631, 0.5071],
  17120. [0.6135, 0.4115, 0.8838, 0.4667, 0.4288, 0.6050, 0.5778, 0.5097],
  17121. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  17122. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5788, 0.5117],
  17123. [0.6132, 0.4118, 0.8200, 0.3633, 0.3562, 0.5400, 0.5787, 0.5136],
  17124. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  17125. [0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6263, 0.5367],
  17126. [0.6212, 0.4159, 0.8675, 0.5783, 0.4087, 0.4317, 0.5612, 0.5917]]],
  17127. device='cuda:0')
  17128. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  17129. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  17130. loss_train: 0.05076544382609427
  17131. step: 49
  17132. running loss: 0.0010360294658386586
  17133. Train Steps: 49/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17134. torch.Size([8, 8])
  17135. tensor([[0.6058, 0.3986, 0.8324, 0.4626, 0.3838, 0.4983, 0.5147, 0.5466],
  17136. [0.6270, 0.4267, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  17137. [0.6040, 0.4002, 0.7338, 0.2267, 0.3975, 0.2100, 0.5231, 0.4778],
  17138. [0.6182, 0.4099, 0.7812, 0.3000, 0.3937, 0.2367, 0.5325, 0.5750],
  17139. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  17140. [0.6353, 0.4128, 0.8488, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  17141. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  17142. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483]],
  17143. device='cuda:0', dtype=torch.float64)
  17144. predictions are: tensor([[0.5817, 0.3841, 0.8039, 0.4701, 0.3774, 0.5051, 0.5660, 0.5194],
  17145. [0.4629, 0.3292, 0.7094, 0.2834, 0.4573, 0.1952, 0.5821, 0.5922],
  17146. [0.5646, 0.3773, 0.7173, 0.1904, 0.3819, 0.2052, 0.5569, 0.4717],
  17147. [0.5157, 0.3567, 0.7806, 0.2952, 0.3890, 0.2561, 0.5382, 0.5372],
  17148. [0.6007, 0.4074, 0.8742, 0.4553, 0.4202, 0.5225, 0.6016, 0.5305],
  17149. [0.4842, 0.3445, 0.8362, 0.2285, 0.5071, 0.1936, 0.6952, 0.5239],
  17150. [0.5824, 0.3880, 0.8562, 0.4701, 0.3626, 0.5318, 0.6304, 0.4917],
  17151. [0.5742, 0.3909, 0.7183, 0.1900, 0.3832, 0.2580, 0.6323, 0.5240]],
  17152. device='cuda:0', grad_fn=<AddmmBackward>)
  17153. landmarks are: tensor([[[0.6058, 0.3986, 0.8324, 0.4626, 0.3837, 0.4983, 0.5147, 0.5466],
  17154. [0.6270, 0.4266, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  17155. [0.6040, 0.4002, 0.7337, 0.2267, 0.3975, 0.2100, 0.5231, 0.4778],
  17156. [0.6182, 0.4099, 0.7812, 0.3000, 0.3938, 0.2367, 0.5325, 0.5750],
  17157. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  17158. [0.6353, 0.4128, 0.8487, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  17159. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  17160. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483]]],
  17161. device='cuda:0')
  17162. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  17163. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  17164. loss_train: 0.05246755573898554
  17165. step: 50
  17166. running loss: 0.0010493511147797107
  17167.  
  17168. Train Steps: 50/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17169. torch.Size([8, 8])
  17170. tensor([[0.6107, 0.4013, 0.8700, 0.4650, 0.5049, 0.5176, 0.5850, 0.5550],
  17171. [0.6245, 0.4100, 0.7762, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  17172. [0.6055, 0.4015, 0.7425, 0.2033, 0.4113, 0.1883, 0.5217, 0.4823],
  17173. [0.6145, 0.4008, 0.8750, 0.5383, 0.3975, 0.4650, 0.5563, 0.5533],
  17174. [ nan, nan, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  17175. [0.6224, 0.4061, 0.8988, 0.4300, 0.3838, 0.4750, 0.6112, 0.5483],
  17176. [ nan, nan, 0.7525, 0.2291, 0.3838, 0.3017, 0.6050, 0.5667],
  17177. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517]],
  17178. device='cuda:0', dtype=torch.float64)
  17179. predictions are: tensor([[0.6535, 0.4426, 0.8615, 0.4605, 0.4753, 0.5020, 0.6464, 0.5205],
  17180. [0.6618, 0.4507, 0.7415, 0.2485, 0.4715, 0.1504, 0.6295, 0.5336],
  17181. [0.6118, 0.4104, 0.7157, 0.1890, 0.3934, 0.1675, 0.5420, 0.4826],
  17182. [0.6650, 0.4469, 0.8542, 0.5318, 0.3817, 0.4565, 0.5792, 0.5371],
  17183. [0.0827, 0.0845, 0.7178, 0.2095, 0.4269, 0.2542, 0.5514, 0.5404],
  17184. [0.7079, 0.4768, 0.8986, 0.4412, 0.3717, 0.4759, 0.6466, 0.5368],
  17185. [0.1929, 0.1621, 0.7625, 0.2185, 0.3761, 0.2935, 0.6065, 0.5570],
  17186. [0.7233, 0.4964, 0.8791, 0.4836, 0.4390, 0.5540, 0.6191, 0.5469]],
  17187. device='cuda:0', grad_fn=<AddmmBackward>)
  17188. landmarks are: tensor([[[0.6107, 0.4013, 0.8700, 0.4650, 0.5049, 0.5176, 0.5850, 0.5550],
  17189. [0.6245, 0.4100, 0.7763, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  17190. [0.6055, 0.4015, 0.7425, 0.2033, 0.4112, 0.1883, 0.5217, 0.4823],
  17191. [0.6145, 0.4008, 0.8750, 0.5383, 0.3975, 0.4650, 0.5562, 0.5533],
  17192. [0.0000, 0.0000, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  17193. [0.6224, 0.4061, 0.8988, 0.4300, 0.3837, 0.4750, 0.6112, 0.5483],
  17194. [0.0000, 0.0000, 0.7525, 0.2291, 0.3837, 0.3017, 0.6050, 0.5667],
  17195. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517]]],
  17196. device='cuda:0')
  17197. loss_train_step before backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  17198. loss_train_step after backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  17199. loss_train: 0.05460604024119675
  17200. step: 51
  17201. running loss: 0.0010707066713960148
  17202. Train Steps: 51/90 Loss: 0.0011 torch.Size([8, 600, 800])
  17203. torch.Size([8, 8])
  17204. tensor([[0.6087, 0.3951, 0.8387, 0.5833, 0.4188, 0.4933, 0.5146, 0.4830],
  17205. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683],
  17206. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  17207. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  17208. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  17209. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  17210. [0.6251, 0.4163, 0.8662, 0.4467, 0.3625, 0.3567, 0.6038, 0.5533],
  17211. [0.6239, 0.4206, 0.8750, 0.5400, 0.3688, 0.4850, 0.5737, 0.5700]],
  17212. device='cuda:0', dtype=torch.float64)
  17213. predictions are: tensor([[0.5814, 0.3786, 0.8285, 0.5403, 0.4182, 0.5051, 0.5368, 0.5026],
  17214. [0.6062, 0.3988, 0.8697, 0.4950, 0.3917, 0.3794, 0.5435, 0.5671],
  17215. [0.6641, 0.4272, 0.8921, 0.3767, 0.3677, 0.3942, 0.6647, 0.4933],
  17216. [0.6181, 0.4069, 0.8375, 0.2639, 0.4191, 0.2653, 0.6480, 0.4718],
  17217. [0.5773, 0.3799, 0.8746, 0.4581, 0.5086, 0.5067, 0.5284, 0.5730],
  17218. [0.5787, 0.4009, 0.7336, 0.2481, 0.4412, 0.2274, 0.5866, 0.5746],
  17219. [0.5987, 0.4005, 0.8599, 0.4030, 0.3679, 0.3575, 0.6200, 0.5459],
  17220. [0.6027, 0.3978, 0.8663, 0.5080, 0.4044, 0.5205, 0.6093, 0.5446]],
  17221. device='cuda:0', grad_fn=<AddmmBackward>)
  17222. landmarks are: tensor([[[0.6087, 0.3951, 0.8388, 0.5833, 0.4187, 0.4933, 0.5146, 0.4830],
  17223. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683],
  17224. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  17225. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  17226. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  17227. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  17228. [0.6252, 0.4162, 0.8662, 0.4467, 0.3625, 0.3567, 0.6037, 0.5533],
  17229. [0.6239, 0.4206, 0.8750, 0.5400, 0.3688, 0.4850, 0.5738, 0.5700]]],
  17230. device='cuda:0')
  17231. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  17232. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  17233. loss_train: 0.0550856547197327
  17234. step: 52
  17235. running loss: 0.0010593395138410134
  17236. Train Steps: 52/90 Loss: 0.0011 torch.Size([8, 600, 800])
  17237. torch.Size([8, 8])
  17238. tensor([[0.6128, 0.4022, 0.8738, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  17239. [ nan, nan, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  17240. [0.6115, 0.4005, 0.8838, 0.3867, 0.3763, 0.4700, 0.5800, 0.5550],
  17241. [0.6219, 0.4097, 0.8738, 0.3400, 0.3563, 0.4117, 0.5975, 0.5683],
  17242. [0.6260, 0.4153, 0.9000, 0.4533, 0.4025, 0.2633, 0.6223, 0.4967],
  17243. [0.6227, 0.4083, 0.8938, 0.4800, 0.3800, 0.2950, 0.5737, 0.5350],
  17244. [0.6199, 0.4093, 0.7913, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700],
  17245. [0.6085, 0.4005, 0.8400, 0.4317, 0.3763, 0.4750, 0.5476, 0.5058]],
  17246. device='cuda:0', dtype=torch.float64)
  17247. predictions are: tensor([[0.6195, 0.4219, 0.8560, 0.4903, 0.5090, 0.5246, 0.5346, 0.5340],
  17248. [0.0739, 0.0713, 0.8875, 0.3262, 0.5059, 0.2328, 0.6973, 0.5713],
  17249. [0.6520, 0.4337, 0.8578, 0.3723, 0.4002, 0.4738, 0.5595, 0.5606],
  17250. [0.6410, 0.4396, 0.8600, 0.3413, 0.3624, 0.3962, 0.5996, 0.5702],
  17251. [0.6282, 0.4127, 0.8733, 0.4211, 0.4223, 0.2760, 0.6183, 0.5145],
  17252. [0.6489, 0.4297, 0.8637, 0.4640, 0.4111, 0.3077, 0.5486, 0.5505],
  17253. [0.6476, 0.4493, 0.7594, 0.2379, 0.4411, 0.2562, 0.5998, 0.5810],
  17254. [0.6860, 0.4447, 0.8223, 0.4055, 0.3957, 0.4801, 0.5548, 0.5080]],
  17255. device='cuda:0', grad_fn=<AddmmBackward>)
  17256. landmarks are: tensor([[[0.6128, 0.4022, 0.8737, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  17257. [0.0000, 0.0000, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  17258. [0.6115, 0.4005, 0.8838, 0.3867, 0.3762, 0.4700, 0.5800, 0.5550],
  17259. [0.6219, 0.4097, 0.8737, 0.3400, 0.3562, 0.4117, 0.5975, 0.5683],
  17260. [0.6260, 0.4153, 0.9000, 0.4533, 0.4025, 0.2633, 0.6223, 0.4967],
  17261. [0.6227, 0.4083, 0.8938, 0.4800, 0.3800, 0.2950, 0.5738, 0.5350],
  17262. [0.6198, 0.4093, 0.7912, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700],
  17263. [0.6084, 0.4005, 0.8400, 0.4317, 0.3762, 0.4750, 0.5476, 0.5058]]],
  17264. device='cuda:0')
  17265. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  17266. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  17267. loss_train: 0.05573676276253536
  17268. step: 53
  17269. running loss: 0.0010516370332553841
  17270. Train Steps: 53/90 Loss: 0.0011 torch.Size([8, 600, 800])
  17271. torch.Size([8, 8])
  17272. tensor([[0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  17273. [0.6263, 0.4030, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  17274. [0.6187, 0.4104, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  17275. [0.6236, 0.3966, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  17276. [0.6075, 0.4007, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280],
  17277. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  17278. [0.6068, 0.3963, 0.8650, 0.4317, 0.4037, 0.5083, 0.5253, 0.4999],
  17279. [0.6203, 0.4078, 0.8800, 0.5083, 0.3900, 0.5000, 0.6100, 0.5583]],
  17280. device='cuda:0', dtype=torch.float64)
  17281. predictions are: tensor([[0.6280, 0.4024, 0.9012, 0.5148, 0.4555, 0.4977, 0.6189, 0.5503],
  17282. [0.5665, 0.3724, 0.9334, 0.4956, 0.4231, 0.5474, 0.6169, 0.5207],
  17283. [0.6607, 0.4436, 0.7319, 0.2175, 0.4085, 0.2458, 0.5786, 0.5611],
  17284. [0.6557, 0.4170, 0.9130, 0.4984, 0.4010, 0.4071, 0.5816, 0.5355],
  17285. [0.6025, 0.3909, 0.8496, 0.5025, 0.4549, 0.5250, 0.5244, 0.5597],
  17286. [0.6019, 0.3968, 0.8792, 0.5568, 0.4240, 0.4326, 0.5877, 0.5582],
  17287. [0.6882, 0.4499, 0.8793, 0.4420, 0.4445, 0.5120, 0.5330, 0.5221],
  17288. [0.6171, 0.4100, 0.9059, 0.5125, 0.4226, 0.5082, 0.5894, 0.5972]],
  17289. device='cuda:0', grad_fn=<AddmmBackward>)
  17290. landmarks are: tensor([[[0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  17291. [0.6263, 0.4029, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  17292. [0.6187, 0.4103, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  17293. [0.6236, 0.3965, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  17294. [0.6075, 0.4006, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280],
  17295. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  17296. [0.6068, 0.3963, 0.8650, 0.4317, 0.4038, 0.5083, 0.5253, 0.4999],
  17297. [0.6203, 0.4078, 0.8800, 0.5083, 0.3900, 0.5000, 0.6100, 0.5583]]],
  17298. device='cuda:0')
  17299. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  17300. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  17301. loss_train: 0.056453864090144634
  17302. step: 54
  17303. running loss: 0.001045441927595271
  17304.  
  17305. Train Steps: 54/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17306. torch.Size([8, 8])
  17307. tensor([[0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  17308. [0.6277, 0.4029, 0.8250, 0.2433, 0.4325, 0.2100, 0.6366, 0.5207],
  17309. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250],
  17310. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  17311. [0.6129, 0.4063, 0.8738, 0.5250, 0.4313, 0.4733, 0.5230, 0.5874],
  17312. [0.6152, 0.4131, 0.6863, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  17313. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  17314. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123]],
  17315. device='cuda:0', dtype=torch.float64)
  17316. predictions are: tensor([[0.5476, 0.3369, 0.8878, 0.5874, 0.4094, 0.4444, 0.5650, 0.5610],
  17317. [0.6466, 0.4018, 0.8539, 0.2829, 0.4648, 0.2358, 0.6455, 0.5369],
  17318. [0.5906, 0.3769, 0.9650, 0.4317, 0.3911, 0.3305, 0.6028, 0.5531],
  17319. [0.5458, 0.3493, 0.9107, 0.5839, 0.4247, 0.4551, 0.5193, 0.5251],
  17320. [0.6134, 0.3893, 0.9136, 0.5638, 0.4533, 0.5064, 0.5083, 0.6175],
  17321. [0.6576, 0.4241, 0.7325, 0.2693, 0.3812, 0.3485, 0.5727, 0.5774],
  17322. [0.6544, 0.4179, 0.7269, 0.2816, 0.4004, 0.2486, 0.5374, 0.5922],
  17323. [0.6109, 0.3914, 0.7628, 0.2722, 0.4018, 0.2721, 0.5324, 0.5373]],
  17324. device='cuda:0', grad_fn=<AddmmBackward>)
  17325. landmarks are: tensor([[[0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  17326. [0.6277, 0.4029, 0.8250, 0.2433, 0.4325, 0.2100, 0.6366, 0.5207],
  17327. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250],
  17328. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  17329. [0.6130, 0.4063, 0.8737, 0.5250, 0.4313, 0.4733, 0.5230, 0.5874],
  17330. [0.6152, 0.4131, 0.6862, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  17331. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  17332. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123]]],
  17333. device='cuda:0')
  17334. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  17335. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  17336. loss_train: 0.057424455240834504
  17337. step: 55
  17338. running loss: 0.0010440810043788093
  17339. Train Steps: 55/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17340. torch.Size([8, 8])
  17341. tensor([[0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  17342. [ nan, nan, 0.7412, 0.2200, 0.4450, 0.1517, 0.5312, 0.4983],
  17343. [0.6193, 0.4034, 0.7757, 0.2347, 0.3733, 0.2919, 0.5930, 0.4926],
  17344. [0.6205, 0.4016, 0.8350, 0.2717, 0.3987, 0.2550, 0.5787, 0.5133],
  17345. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783],
  17346. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  17347. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  17348. [0.6076, 0.3953, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954]],
  17349. device='cuda:0', dtype=torch.float64)
  17350. predictions are: tensor([[0.6488, 0.3997, 0.9324, 0.4416, 0.4320, 0.3891, 0.6941, 0.6133],
  17351. [0.0758, 0.0263, 0.7338, 0.2581, 0.4481, 0.1622, 0.5430, 0.5465],
  17352. [0.7226, 0.4630, 0.7617, 0.2752, 0.3950, 0.3043, 0.5740, 0.5084],
  17353. [0.6153, 0.3819, 0.8446, 0.3057, 0.4062, 0.2827, 0.5835, 0.5420],
  17354. [0.5881, 0.3800, 0.8620, 0.4081, 0.3653, 0.3207, 0.4853, 0.5845],
  17355. [0.6604, 0.4283, 0.7799, 0.3036, 0.4455, 0.1807, 0.5666, 0.5509],
  17356. [0.5751, 0.3707, 0.7140, 0.2635, 0.4521, 0.1633, 0.5350, 0.5606],
  17357. [0.6710, 0.4345, 0.8234, 0.4160, 0.3584, 0.4337, 0.5220, 0.5283]],
  17358. device='cuda:0', grad_fn=<AddmmBackward>)
  17359. landmarks are: tensor([[[0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  17360. [0.0000, 0.0000, 0.7412, 0.2200, 0.4450, 0.1517, 0.5312, 0.4983],
  17361. [0.6193, 0.4034, 0.7757, 0.2347, 0.3733, 0.2919, 0.5930, 0.4926],
  17362. [0.6205, 0.4015, 0.8350, 0.2717, 0.3988, 0.2550, 0.5788, 0.5133],
  17363. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783],
  17364. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  17365. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  17366. [0.6076, 0.3952, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954]]],
  17367. device='cuda:0')
  17368. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  17369. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  17370. loss_train: 0.05837335210526362
  17371. step: 56
  17372. running loss: 0.0010423812875939933
  17373. Train Steps: 56/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17374. torch.Size([8, 8])
  17375. tensor([[0.6185, 0.4079, 0.8838, 0.4617, 0.4838, 0.5650, 0.6175, 0.5850],
  17376. [0.6336, 0.4191, 0.8938, 0.5167, 0.3937, 0.3517, 0.7343, 0.5748],
  17377. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  17378. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  17379. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  17380. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  17381. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  17382. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834]],
  17383. device='cuda:0', dtype=torch.float64)
  17384. predictions are: tensor([[0.5535, 0.3603, 0.9047, 0.4861, 0.4687, 0.5778, 0.5736, 0.5600],
  17385. [0.5951, 0.3766, 0.8869, 0.5239, 0.3845, 0.3401, 0.6707, 0.5305],
  17386. [0.6612, 0.4110, 0.8894, 0.3805, 0.3492, 0.4530, 0.6655, 0.4991],
  17387. [0.6235, 0.4039, 0.6751, 0.3450, 0.3461, 0.2827, 0.4931, 0.5618],
  17388. [0.5802, 0.3689, 0.7180, 0.2602, 0.3880, 0.2208, 0.5133, 0.5405],
  17389. [0.6408, 0.4002, 0.8874, 0.3405, 0.3949, 0.4310, 0.6676, 0.5227],
  17390. [0.5665, 0.3689, 0.8803, 0.5668, 0.3373, 0.3560, 0.5179, 0.5180],
  17391. [0.5994, 0.3896, 0.8057, 0.2313, 0.4689, 0.1243, 0.6043, 0.4866]],
  17392. device='cuda:0', grad_fn=<AddmmBackward>)
  17393. landmarks are: tensor([[[0.6184, 0.4079, 0.8838, 0.4617, 0.4837, 0.5650, 0.6175, 0.5850],
  17394. [0.6336, 0.4191, 0.8938, 0.5167, 0.3938, 0.3517, 0.7343, 0.5748],
  17395. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  17396. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  17397. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  17398. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  17399. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  17400. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834]]],
  17401. device='cuda:0')
  17402. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  17403. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  17404. loss_train: 0.059197709371801466
  17405. step: 57
  17406. running loss: 0.0010385563047684467
  17407. Train Steps: 57/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17408. torch.Size([8, 8])
  17409. tensor([[0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  17410. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  17411. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  17412. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  17413. [0.6311, 0.4008, 0.7935, 0.5746, 0.3900, 0.5033, 0.6955, 0.5366],
  17414. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  17415. [0.6193, 0.4050, 0.7313, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  17416. [0.6289, 0.4019, 0.8113, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332]],
  17417. device='cuda:0', dtype=torch.float64)
  17418. predictions are: tensor([[0.6581, 0.4206, 0.8478, 0.3237, 0.3369, 0.3658, 0.5894, 0.4957],
  17419. [0.6343, 0.4063, 0.8662, 0.5718, 0.3784, 0.4163, 0.5256, 0.5283],
  17420. [0.6484, 0.4191, 0.8517, 0.5362, 0.3857, 0.5657, 0.6739, 0.5407],
  17421. [0.6023, 0.3653, 0.8979, 0.5239, 0.3713, 0.5414, 0.6793, 0.5288],
  17422. [0.6106, 0.3916, 0.8149, 0.5506, 0.3792, 0.4707, 0.6510, 0.5134],
  17423. [0.6310, 0.4057, 0.8812, 0.4616, 0.4065, 0.4958, 0.5455, 0.5182],
  17424. [0.6720, 0.4215, 0.7415, 0.2647, 0.3884, 0.1843, 0.5366, 0.5280],
  17425. [0.6216, 0.3953, 0.8320, 0.5320, 0.3525, 0.4859, 0.6587, 0.5009]],
  17426. device='cuda:0', grad_fn=<AddmmBackward>)
  17427. landmarks are: tensor([[[0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  17428. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  17429. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  17430. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  17431. [0.6311, 0.4008, 0.7935, 0.5746, 0.3900, 0.5033, 0.6955, 0.5366],
  17432. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  17433. [0.6193, 0.4050, 0.7312, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  17434. [0.6289, 0.4019, 0.8112, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332]]],
  17435. device='cuda:0')
  17436. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  17437. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  17438. loss_train: 0.05994822469074279
  17439. step: 58
  17440. running loss: 0.0010335900808748756
  17441.  
  17442. Train Steps: 58/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17443. torch.Size([8, 8])
  17444. tensor([[0.6261, 0.4029, 0.8720, 0.3364, 0.3665, 0.3753, 0.6531, 0.5183],
  17445. [0.6186, 0.4060, 0.8750, 0.5050, 0.3538, 0.4367, 0.5813, 0.6083],
  17446. [0.6199, 0.4093, 0.7913, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700],
  17447. [0.6250, 0.4236, 0.8638, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  17448. [0.6222, 0.3937, 0.8350, 0.5617, 0.4138, 0.4600, 0.5800, 0.5233],
  17449. [0.6259, 0.4133, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  17450. [0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  17451. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783]],
  17452. device='cuda:0', dtype=torch.float64)
  17453. predictions are: tensor([[0.6293, 0.4008, 0.8419, 0.3521, 0.3527, 0.3801, 0.6642, 0.4922],
  17454. [0.6315, 0.4160, 0.8540, 0.4946, 0.3421, 0.4459, 0.6039, 0.5588],
  17455. [0.5937, 0.4044, 0.7634, 0.2575, 0.3904, 0.2473, 0.6448, 0.5504],
  17456. [0.6096, 0.3997, 0.8519, 0.3804, 0.3719, 0.3151, 0.6080, 0.5556],
  17457. [0.5817, 0.3542, 0.8235, 0.5666, 0.3866, 0.4522, 0.6036, 0.5165],
  17458. [0.6386, 0.4336, 0.8146, 0.2436, 0.4702, 0.1479, 0.6958, 0.4712],
  17459. [0.5744, 0.3767, 0.8577, 0.4240, 0.3493, 0.5482, 0.5868, 0.4898],
  17460. [0.6189, 0.3986, 0.8186, 0.3587, 0.3306, 0.2866, 0.5438, 0.5318]],
  17461. device='cuda:0', grad_fn=<AddmmBackward>)
  17462. landmarks are: tensor([[[0.6261, 0.4029, 0.8720, 0.3364, 0.3665, 0.3753, 0.6531, 0.5183],
  17463. [0.6186, 0.4060, 0.8750, 0.5050, 0.3537, 0.4367, 0.5813, 0.6083],
  17464. [0.6198, 0.4093, 0.7912, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700],
  17465. [0.6250, 0.4236, 0.8637, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  17466. [0.6222, 0.3937, 0.8350, 0.5617, 0.4137, 0.4600, 0.5800, 0.5233],
  17467. [0.6259, 0.4132, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  17468. [0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  17469. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783]]],
  17470. device='cuda:0')
  17471. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  17472. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  17473. loss_train: 0.060535903030540794
  17474. step: 59
  17475. running loss: 0.0010260322547549288
  17476. Train Steps: 59/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17477. torch.Size([8, 8])
  17478. tensor([[0.6201, 0.4102, 0.7288, 0.2417, 0.4150, 0.2383, 0.6100, 0.5500],
  17479. [0.6115, 0.4081, 0.6725, 0.2433, 0.4088, 0.1933, 0.5167, 0.5544],
  17480. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  17481. [0.6170, 0.4102, 0.7468, 0.3695, 0.3463, 0.3767, 0.5238, 0.5823],
  17482. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  17483. [0.6212, 0.4033, 0.8938, 0.4167, 0.3813, 0.4267, 0.5613, 0.5583],
  17484. [0.6204, 0.4110, 0.7913, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367],
  17485. [0.6076, 0.3953, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954]],
  17486. device='cuda:0', dtype=torch.float64)
  17487. predictions are: tensor([[0.6403, 0.4280, 0.7446, 0.2610, 0.3933, 0.2408, 0.6553, 0.5600],
  17488. [0.6543, 0.4360, 0.6836, 0.2661, 0.4094, 0.1976, 0.5499, 0.5387],
  17489. [0.6499, 0.4227, 0.7998, 0.2759, 0.4333, 0.2275, 0.6982, 0.5251],
  17490. [0.5878, 0.3954, 0.7473, 0.3859, 0.3282, 0.3870, 0.5789, 0.5672],
  17491. [0.5764, 0.3836, 0.8764, 0.5084, 0.3964, 0.4931, 0.5974, 0.5081],
  17492. [0.5695, 0.3758, 0.8933, 0.4360, 0.3550, 0.4061, 0.6115, 0.5394],
  17493. [0.6436, 0.4339, 0.7857, 0.2777, 0.3960, 0.2413, 0.6778, 0.5352],
  17494. [0.5729, 0.3854, 0.8123, 0.3846, 0.3252, 0.4032, 0.5814, 0.4950]],
  17495. device='cuda:0', grad_fn=<AddmmBackward>)
  17496. landmarks are: tensor([[[0.6201, 0.4102, 0.7287, 0.2417, 0.4150, 0.2383, 0.6100, 0.5500],
  17497. [0.6115, 0.4081, 0.6725, 0.2433, 0.4087, 0.1933, 0.5167, 0.5544],
  17498. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  17499. [0.6170, 0.4102, 0.7468, 0.3695, 0.3462, 0.3767, 0.5238, 0.5823],
  17500. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  17501. [0.6212, 0.4033, 0.8938, 0.4167, 0.3812, 0.4267, 0.5612, 0.5583],
  17502. [0.6204, 0.4110, 0.7912, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367],
  17503. [0.6076, 0.3952, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954]]],
  17504. device='cuda:0')
  17505. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  17506. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  17507. loss_train: 0.061192915483843535
  17508. step: 60
  17509. running loss: 0.0010198819247307256
  17510. Train Steps: 60/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17511. torch.Size([8, 8])
  17512. tensor([[0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  17513. [0.6250, 0.4146, 0.8838, 0.3933, 0.3588, 0.4283, 0.6162, 0.5367],
  17514. [0.6087, 0.3976, 0.8337, 0.3867, 0.3713, 0.3117, 0.5938, 0.5300],
  17515. [0.6165, 0.4106, 0.7575, 0.1733, 0.3838, 0.2650, 0.5680, 0.5116],
  17516. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  17517. [0.6229, 0.4107, 0.8137, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  17518. [0.6133, 0.4094, 0.8495, 0.4028, 0.3588, 0.3200, 0.5003, 0.5407],
  17519. [0.6289, 0.4081, 0.8720, 0.3487, 0.3900, 0.3183, 0.6703, 0.5376]],
  17520. device='cuda:0', dtype=torch.float64)
  17521. predictions are: tensor([[0.6098, 0.4095, 0.8948, 0.3430, 0.4369, 0.3946, 0.7409, 0.5581],
  17522. [0.5593, 0.3761, 0.8257, 0.3682, 0.3557, 0.4102, 0.6401, 0.5402],
  17523. [0.5868, 0.4028, 0.8061, 0.3714, 0.3707, 0.3365, 0.6015, 0.5382],
  17524. [0.6645, 0.4425, 0.7266, 0.1562, 0.3846, 0.2444, 0.5894, 0.5023],
  17525. [0.5956, 0.4105, 0.8264, 0.5500, 0.3945, 0.4470, 0.5775, 0.5710],
  17526. [0.6154, 0.4200, 0.7679, 0.2885, 0.4765, 0.1840, 0.6045, 0.5576],
  17527. [0.5763, 0.3963, 0.8310, 0.3714, 0.3726, 0.3122, 0.5121, 0.5635],
  17528. [0.5984, 0.3993, 0.8475, 0.3177, 0.3888, 0.2935, 0.6982, 0.5648]],
  17529. device='cuda:0', grad_fn=<AddmmBackward>)
  17530. landmarks are: tensor([[[0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  17531. [0.6250, 0.4146, 0.8838, 0.3933, 0.3587, 0.4283, 0.6162, 0.5367],
  17532. [0.6087, 0.3976, 0.8338, 0.3867, 0.3713, 0.3117, 0.5938, 0.5300],
  17533. [0.6165, 0.4106, 0.7575, 0.1733, 0.3837, 0.2650, 0.5680, 0.5116],
  17534. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  17535. [0.6229, 0.4107, 0.8138, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  17536. [0.6133, 0.4094, 0.8495, 0.4028, 0.3587, 0.3200, 0.5003, 0.5407],
  17537. [0.6289, 0.4081, 0.8720, 0.3487, 0.3900, 0.3183, 0.6703, 0.5376]]],
  17538. device='cuda:0')
  17539. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  17540. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  17541. loss_train: 0.06174909329274669
  17542. step: 61
  17543. running loss: 0.0010122802179138803
  17544. Train Steps: 61/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17545. torch.Size([8, 8])
  17546. tensor([[0.6186, 0.4060, 0.8750, 0.5050, 0.3538, 0.4367, 0.5813, 0.6083],
  17547. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  17548. [0.6229, 0.4066, 0.8513, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  17549. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6038, 0.4833],
  17550. [0.6200, 0.4070, 0.8938, 0.4183, 0.3538, 0.4567, 0.6175, 0.5400],
  17551. [ nan, nan, 0.6688, 0.2513, 0.4113, 0.2117, 0.5193, 0.5933],
  17552. [0.6132, 0.4066, 0.7259, 0.2402, 0.3588, 0.3300, 0.6000, 0.5600],
  17553. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482]],
  17554. device='cuda:0', dtype=torch.float64)
  17555. predictions are: tensor([[0.6732, 0.4678, 0.8815, 0.4820, 0.3907, 0.4533, 0.6017, 0.6037],
  17556. [0.6412, 0.4282, 0.7319, 0.2063, 0.4344, 0.2330, 0.5760, 0.5694],
  17557. [0.6649, 0.4414, 0.8627, 0.5376, 0.4606, 0.4971, 0.6258, 0.5521],
  17558. [0.7022, 0.4611, 0.8948, 0.4707, 0.3917, 0.5046, 0.6166, 0.4972],
  17559. [0.7338, 0.4807, 0.8952, 0.3996, 0.3814, 0.4844, 0.6488, 0.5259],
  17560. [0.0807, 0.0648, 0.6899, 0.2234, 0.4330, 0.2029, 0.5665, 0.5873],
  17561. [0.6507, 0.4482, 0.7274, 0.2220, 0.3836, 0.3474, 0.6153, 0.5847],
  17562. [0.7269, 0.4700, 0.9090, 0.4028, 0.3918, 0.3560, 0.6513, 0.5766]],
  17563. device='cuda:0', grad_fn=<AddmmBackward>)
  17564. landmarks are: tensor([[[0.6186, 0.4060, 0.8750, 0.5050, 0.3537, 0.4367, 0.5813, 0.6083],
  17565. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  17566. [0.6229, 0.4066, 0.8512, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  17567. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6037, 0.4833],
  17568. [0.6200, 0.4070, 0.8938, 0.4183, 0.3537, 0.4567, 0.6175, 0.5400],
  17569. [0.0000, 0.0000, 0.6688, 0.2513, 0.4112, 0.2117, 0.5193, 0.5933],
  17570. [0.6132, 0.4066, 0.7259, 0.2402, 0.3587, 0.3300, 0.6000, 0.5600],
  17571. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482]]],
  17572. device='cuda:0')
  17573. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  17574. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  17575. loss_train: 0.06322128738975152
  17576. step: 62
  17577. running loss: 0.0010196981837056697
  17578.  
  17579. Train Steps: 62/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17580. torch.Size([8, 8])
  17581. tensor([[0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  17582. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  17583. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  17584. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  17585. [0.6332, 0.4165, 0.9100, 0.3350, 0.4188, 0.3683, 0.7438, 0.5528],
  17586. [0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133],
  17587. [0.6257, 0.4034, 0.8287, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  17588. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540]],
  17589. device='cuda:0', dtype=torch.float64)
  17590. predictions are: tensor([[0.6324, 0.4039, 0.8685, 0.4872, 0.4011, 0.4443, 0.4893, 0.5842],
  17591. [0.6219, 0.4002, 0.7847, 0.1992, 0.4461, 0.2610, 0.6476, 0.5553],
  17592. [0.6258, 0.4004, 0.8205, 0.4090, 0.3847, 0.4614, 0.5261, 0.5988],
  17593. [0.6113, 0.3853, 0.8718, 0.4939, 0.4130, 0.5536, 0.7076, 0.5905],
  17594. [0.6323, 0.4041, 0.8950, 0.3352, 0.4317, 0.3551, 0.7193, 0.5770],
  17595. [0.6194, 0.4062, 0.8558, 0.5051, 0.3982, 0.4593, 0.6008, 0.5352],
  17596. [0.6342, 0.4045, 0.8060, 0.2035, 0.4183, 0.2695, 0.5965, 0.5183],
  17597. [0.6079, 0.3940, 0.8847, 0.4094, 0.4375, 0.3078, 0.6545, 0.5843]],
  17598. device='cuda:0', grad_fn=<AddmmBackward>)
  17599. landmarks are: tensor([[[0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  17600. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  17601. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  17602. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  17603. [0.6332, 0.4165, 0.9100, 0.3350, 0.4187, 0.3683, 0.7438, 0.5528],
  17604. [0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133],
  17605. [0.6257, 0.4034, 0.8288, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  17606. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540]]],
  17607. device='cuda:0')
  17608. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  17609. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  17610. loss_train: 0.06367821450112388
  17611. step: 63
  17612. running loss: 0.0010107653095416489
  17613. Train Steps: 63/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17614. torch.Size([8, 8])
  17615. tensor([[0.6126, 0.4039, 0.8237, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138],
  17616. [0.6218, 0.4185, 0.7338, 0.2650, 0.4625, 0.1950, 0.5687, 0.5800],
  17617. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  17618. [0.6257, 0.4034, 0.8287, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  17619. [0.6179, 0.4008, 0.8600, 0.4015, 0.3932, 0.2515, 0.5711, 0.5438],
  17620. [0.6197, 0.4091, 0.8800, 0.4783, 0.3538, 0.4767, 0.5950, 0.5550],
  17621. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  17622. [0.6200, 0.4071, 0.7338, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517]],
  17623. device='cuda:0', dtype=torch.float64)
  17624. predictions are: tensor([[0.6023, 0.3895, 0.8220, 0.3872, 0.3741, 0.3751, 0.5697, 0.6120],
  17625. [0.5667, 0.3706, 0.7275, 0.2095, 0.4766, 0.2035, 0.5421, 0.6021],
  17626. [0.5960, 0.3922, 0.9036, 0.3543, 0.4618, 0.2125, 0.6223, 0.5303],
  17627. [0.6050, 0.3791, 0.8262, 0.2050, 0.4125, 0.2759, 0.6123, 0.5004],
  17628. [0.5788, 0.3780, 0.8298, 0.3318, 0.4147, 0.2553, 0.5634, 0.5648],
  17629. [0.5880, 0.3790, 0.8971, 0.4535, 0.3892, 0.4829, 0.5863, 0.5748],
  17630. [0.5920, 0.3777, 0.8626, 0.4950, 0.4768, 0.5564, 0.5932, 0.5270],
  17631. [0.5899, 0.3935, 0.7271, 0.1832, 0.4383, 0.2355, 0.6159, 0.5777]],
  17632. device='cuda:0', grad_fn=<AddmmBackward>)
  17633. landmarks are: tensor([[[0.6126, 0.4038, 0.8238, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138],
  17634. [0.6218, 0.4185, 0.7337, 0.2650, 0.4625, 0.1950, 0.5688, 0.5800],
  17635. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  17636. [0.6257, 0.4034, 0.8288, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  17637. [0.6179, 0.4008, 0.8600, 0.4015, 0.3932, 0.2515, 0.5711, 0.5438],
  17638. [0.6197, 0.4091, 0.8800, 0.4783, 0.3537, 0.4767, 0.5950, 0.5550],
  17639. [0.6129, 0.3925, 0.8720, 0.5246, 0.4534, 0.5515, 0.6026, 0.5114],
  17640. [0.6200, 0.4071, 0.7337, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517]]],
  17641. device='cuda:0')
  17642. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  17643. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  17644. loss_train: 0.06423556891968474
  17645. step: 64
  17646. running loss: 0.001003680764370074
  17647. Train Steps: 64/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17648. torch.Size([8, 8])
  17649. tensor([[ nan, nan, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  17650. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  17651. [0.6212, 0.4171, 0.7875, 0.3633, 0.3813, 0.2933, 0.5675, 0.5700],
  17652. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  17653. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  17654. [0.6265, 0.4251, 0.7113, 0.3550, 0.4375, 0.2117, 0.5587, 0.6118],
  17655. [0.6204, 0.4055, 0.8438, 0.5733, 0.4574, 0.4801, 0.5487, 0.5617],
  17656. [0.6043, 0.4022, 0.6887, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136]],
  17657. device='cuda:0', dtype=torch.float64)
  17658. predictions are: tensor([[0.0176, 0.0051, 0.7073, 0.2451, 0.3940, 0.2041, 0.5673, 0.5674],
  17659. [0.7240, 0.4591, 0.8006, 0.2585, 0.3777, 0.3049, 0.6131, 0.5358],
  17660. [0.6416, 0.4175, 0.8055, 0.3550, 0.3874, 0.3014, 0.5880, 0.5704],
  17661. [0.6756, 0.4380, 0.8862, 0.3885, 0.3845, 0.4889, 0.5990, 0.5338],
  17662. [0.6557, 0.4121, 0.8220, 0.2495, 0.4030, 0.2284, 0.5946, 0.5053],
  17663. [0.6400, 0.4365, 0.7531, 0.3356, 0.4336, 0.2066, 0.5468, 0.6130],
  17664. [0.6865, 0.4423, 0.8561, 0.5533, 0.4550, 0.4492, 0.5814, 0.5617],
  17665. [0.6215, 0.4045, 0.7008, 0.1796, 0.4043, 0.2497, 0.5549, 0.5109]],
  17666. device='cuda:0', grad_fn=<AddmmBackward>)
  17667. landmarks are: tensor([[[0.0000, 0.0000, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  17668. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  17669. [0.6212, 0.4171, 0.7875, 0.3633, 0.3812, 0.2933, 0.5675, 0.5700],
  17670. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  17671. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  17672. [0.6265, 0.4251, 0.7113, 0.3550, 0.4375, 0.2117, 0.5587, 0.6118],
  17673. [0.6204, 0.4055, 0.8438, 0.5733, 0.4574, 0.4801, 0.5487, 0.5617],
  17674. [0.6043, 0.4022, 0.6888, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136]]],
  17675. device='cuda:0')
  17676. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  17677. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  17678. loss_train: 0.06490114069310948
  17679. step: 65
  17680. running loss: 0.0009984790875862998
  17681. Train Steps: 65/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17682. torch.Size([8, 8])
  17683. tensor([[0.6115, 0.4081, 0.6725, 0.2433, 0.4088, 0.1933, 0.5167, 0.5544],
  17684. [ nan, nan, 0.8888, 0.3100, 0.5262, 0.2817, 0.7145, 0.6003],
  17685. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  17686. [0.6208, 0.4082, 0.8538, 0.3067, 0.3588, 0.3717, 0.6112, 0.5517],
  17687. [0.6339, 0.4123, 0.8638, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  17688. [0.6261, 0.3987, 0.9045, 0.4208, 0.3600, 0.4633, 0.6570, 0.5162],
  17689. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  17690. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578]],
  17691. device='cuda:0', dtype=torch.float64)
  17692. predictions are: tensor([[0.6413, 0.4297, 0.6962, 0.2717, 0.4131, 0.2019, 0.4723, 0.5475],
  17693. [0.0419, 0.0319, 0.9071, 0.3396, 0.4849, 0.2528, 0.7129, 0.5819],
  17694. [0.6728, 0.4302, 0.8161, 0.2820, 0.4344, 0.2315, 0.6006, 0.5320],
  17695. [0.6691, 0.4475, 0.8545, 0.3254, 0.3570, 0.3591, 0.5835, 0.5511],
  17696. [0.6772, 0.4555, 0.8836, 0.5338, 0.3839, 0.5459, 0.7248, 0.5521],
  17697. [0.6453, 0.4206, 0.9209, 0.4473, 0.3538, 0.4860, 0.6285, 0.5191],
  17698. [0.6039, 0.3915, 0.7655, 0.3256, 0.3388, 0.3197, 0.5228, 0.5558],
  17699. [0.5965, 0.4058, 0.7044, 0.2233, 0.4040, 0.1886, 0.5195, 0.5606]],
  17700. device='cuda:0', grad_fn=<AddmmBackward>)
  17701. landmarks are: tensor([[[0.6115, 0.4081, 0.6725, 0.2433, 0.4087, 0.1933, 0.5167, 0.5544],
  17702. [0.0000, 0.0000, 0.8888, 0.3100, 0.5263, 0.2817, 0.7145, 0.6003],
  17703. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  17704. [0.6208, 0.4082, 0.8537, 0.3067, 0.3587, 0.3717, 0.6112, 0.5517],
  17705. [0.6339, 0.4123, 0.8637, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  17706. [0.6261, 0.3987, 0.9045, 0.4208, 0.3600, 0.4633, 0.6570, 0.5162],
  17707. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  17708. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578]]],
  17709. device='cuda:0')
  17710. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  17711. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  17712. loss_train: 0.06546252308180556
  17713. step: 66
  17714. running loss: 0.0009918564103303872
  17715.  
  17716. Train Steps: 66/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17717. torch.Size([8, 8])
  17718. tensor([[0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  17719. [0.6100, 0.4016, 0.8600, 0.5067, 0.4612, 0.5233, 0.5086, 0.5519],
  17720. [0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5988, 0.5700],
  17721. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373],
  17722. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  17723. [0.6143, 0.4034, 0.8800, 0.4833, 0.4512, 0.5367, 0.5289, 0.5097],
  17724. [0.6275, 0.4111, 0.8463, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  17725. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575]],
  17726. device='cuda:0', dtype=torch.float64)
  17727. predictions are: tensor([[0.6000, 0.3890, 0.8598, 0.5231, 0.3860, 0.4836, 0.6604, 0.5219],
  17728. [0.5418, 0.3529, 0.8517, 0.5108, 0.4181, 0.4930, 0.5155, 0.5245],
  17729. [0.5073, 0.3391, 0.7839, 0.2833, 0.3466, 0.3270, 0.5728, 0.5695],
  17730. [0.6115, 0.4022, 0.8242, 0.5736, 0.3559, 0.4262, 0.5729, 0.5374],
  17731. [0.6262, 0.4260, 0.8762, 0.5047, 0.3387, 0.2864, 0.6100, 0.5022],
  17732. [0.5922, 0.3897, 0.8660, 0.4663, 0.4117, 0.4919, 0.5157, 0.4979],
  17733. [0.5938, 0.3825, 0.8493, 0.2560, 0.4352, 0.2088, 0.6199, 0.4986],
  17734. [0.4856, 0.3118, 0.8906, 0.3083, 0.3727, 0.3904, 0.6850, 0.5458]],
  17735. device='cuda:0', grad_fn=<AddmmBackward>)
  17736. landmarks are: tensor([[[0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  17737. [0.6100, 0.4016, 0.8600, 0.5067, 0.4613, 0.5233, 0.5086, 0.5519],
  17738. [0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5987, 0.5700],
  17739. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373],
  17740. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  17741. [0.6143, 0.4034, 0.8800, 0.4833, 0.4512, 0.5367, 0.5289, 0.5097],
  17742. [0.6275, 0.4111, 0.8462, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  17743. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575]]],
  17744. device='cuda:0')
  17745. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  17746. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  17747. loss_train: 0.06678534619277343
  17748. step: 67
  17749. running loss: 0.0009967962118324393
  17750. Train Steps: 67/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17751. torch.Size([8, 8])
  17752. tensor([[0.6136, 0.4117, 0.8700, 0.5167, 0.4188, 0.5083, 0.5147, 0.5495],
  17753. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6038, 0.6167],
  17754. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5837, 0.5500],
  17755. [0.6207, 0.4081, 0.7662, 0.2067, 0.3962, 0.3200, 0.6312, 0.5300],
  17756. [0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431],
  17757. [ nan, nan, 0.7268, 0.2333, 0.4125, 0.1933, 0.5112, 0.5383],
  17758. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  17759. [0.6099, 0.4030, 0.8638, 0.5117, 0.4983, 0.4965, 0.5086, 0.5388]],
  17760. device='cuda:0', dtype=torch.float64)
  17761. predictions are: tensor([[0.6511, 0.4257, 0.8758, 0.5555, 0.3867, 0.4742, 0.5685, 0.5145],
  17762. [0.6889, 0.4593, 0.7985, 0.3512, 0.3446, 0.3632, 0.6308, 0.5870],
  17763. [0.6273, 0.4013, 0.8833, 0.4678, 0.3919, 0.5013, 0.6203, 0.5304],
  17764. [0.5450, 0.3481, 0.7664, 0.2362, 0.3711, 0.2955, 0.6387, 0.5304],
  17765. [0.6303, 0.4170, 0.8735, 0.5181, 0.4061, 0.4715, 0.5485, 0.4984],
  17766. [0.0494, 0.0384, 0.7164, 0.2535, 0.3838, 0.1804, 0.5373, 0.5331],
  17767. [0.6281, 0.4168, 0.8878, 0.5096, 0.3740, 0.4657, 0.5807, 0.5154],
  17768. [0.6367, 0.4216, 0.8746, 0.5444, 0.4802, 0.4629, 0.5561, 0.5284]],
  17769. device='cuda:0', grad_fn=<AddmmBackward>)
  17770. landmarks are: tensor([[[0.6136, 0.4117, 0.8700, 0.5167, 0.4187, 0.5083, 0.5147, 0.5495],
  17771. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6037, 0.6167],
  17772. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5838, 0.5500],
  17773. [0.6207, 0.4081, 0.7663, 0.2067, 0.3963, 0.3200, 0.6313, 0.5300],
  17774. [0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431],
  17775. [0.0000, 0.0000, 0.7268, 0.2333, 0.4125, 0.1933, 0.5113, 0.5383],
  17776. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  17777. [0.6098, 0.4030, 0.8637, 0.5117, 0.4983, 0.4965, 0.5086, 0.5388]]],
  17778. device='cuda:0')
  17779. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  17780. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  17781. loss_train: 0.06771578977350146
  17782. step: 68
  17783. running loss: 0.0009958204378456097
  17784. Train Steps: 68/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17785. torch.Size([8, 8])
  17786. tensor([[0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  17787. [0.6160, 0.4093, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  17788. [0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  17789. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  17790. [0.6053, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  17791. [0.6114, 0.4018, 0.7213, 0.1967, 0.3763, 0.2700, 0.5875, 0.5533],
  17792. [0.6164, 0.4102, 0.8850, 0.4867, 0.4213, 0.5983, 0.5609, 0.5038],
  17793. [0.6201, 0.4151, 0.8588, 0.5467, 0.3700, 0.3950, 0.5637, 0.5933]],
  17794. device='cuda:0', dtype=torch.float64)
  17795. predictions are: tensor([[0.5082, 0.3207, 0.9109, 0.3394, 0.5157, 0.2867, 0.7239, 0.5230],
  17796. [0.5908, 0.3902, 0.8460, 0.4669, 0.3775, 0.4759, 0.5441, 0.5613],
  17797. [0.6483, 0.4339, 0.8936, 0.5021, 0.3721, 0.4222, 0.5393, 0.5726],
  17798. [0.5676, 0.3806, 0.8563, 0.3392, 0.3757, 0.3146, 0.5798, 0.5350],
  17799. [0.5690, 0.3795, 0.6955, 0.2171, 0.4206, 0.2327, 0.5407, 0.5215],
  17800. [0.5241, 0.3457, 0.7194, 0.2207, 0.3837, 0.2938, 0.5650, 0.5470],
  17801. [0.5168, 0.3465, 0.8857, 0.5028, 0.4674, 0.6062, 0.6085, 0.5349],
  17802. [0.6198, 0.4226, 0.8568, 0.5667, 0.3677, 0.4148, 0.5821, 0.5672]],
  17803. device='cuda:0', grad_fn=<AddmmBackward>)
  17804. landmarks are: tensor([[[0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  17805. [0.6160, 0.4092, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  17806. [0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  17807. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  17808. [0.6054, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  17809. [0.6114, 0.4018, 0.7212, 0.1967, 0.3762, 0.2700, 0.5875, 0.5533],
  17810. [0.6164, 0.4102, 0.8850, 0.4867, 0.4212, 0.5983, 0.5609, 0.5038],
  17811. [0.6202, 0.4151, 0.8587, 0.5467, 0.3700, 0.3950, 0.5638, 0.5933]]],
  17812. device='cuda:0')
  17813. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  17814. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  17815. loss_train: 0.06889769760891795
  17816. step: 69
  17817. running loss: 0.0009985173566509848
  17818. Train Steps: 69/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17819. torch.Size([8, 8])
  17820. tensor([[0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  17821. [0.6185, 0.4042, 0.7700, 0.2250, 0.4062, 0.2117, 0.5763, 0.5150],
  17822. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6063, 0.6083],
  17823. [0.6250, 0.4146, 0.8838, 0.3933, 0.3588, 0.4283, 0.6162, 0.5367],
  17824. [ nan, nan, 0.7192, 0.2346, 0.4037, 0.2050, 0.5138, 0.5650],
  17825. [0.6163, 0.4006, 0.8788, 0.4683, 0.3663, 0.4883, 0.5887, 0.5017],
  17826. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  17827. [0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367]],
  17828. device='cuda:0', dtype=torch.float64)
  17829. predictions are: tensor([[0.5883, 0.3851, 0.8589, 0.3032, 0.5168, 0.2417, 0.6563, 0.5515],
  17830. [0.5297, 0.3508, 0.7547, 0.2512, 0.4262, 0.2454, 0.5404, 0.5347],
  17831. [0.6445, 0.4419, 0.8571, 0.3767, 0.3818, 0.4626, 0.6121, 0.5993],
  17832. [0.6430, 0.4185, 0.8573, 0.4084, 0.3759, 0.4354, 0.6020, 0.5308],
  17833. [0.0117, 0.0080, 0.7005, 0.2403, 0.4421, 0.2541, 0.4979, 0.5550],
  17834. [0.6239, 0.4059, 0.8724, 0.4794, 0.3975, 0.5280, 0.5794, 0.5216],
  17835. [0.6547, 0.4375, 0.7912, 0.3228, 0.3815, 0.4045, 0.5427, 0.5234],
  17836. [0.6379, 0.4056, 0.8605, 0.3202, 0.4096, 0.3092, 0.6162, 0.5334]],
  17837. device='cuda:0', grad_fn=<AddmmBackward>)
  17838. landmarks are: tensor([[[0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  17839. [0.6184, 0.4042, 0.7700, 0.2250, 0.4062, 0.2117, 0.5763, 0.5150],
  17840. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6062, 0.6083],
  17841. [0.6250, 0.4146, 0.8838, 0.3933, 0.3587, 0.4283, 0.6162, 0.5367],
  17842. [0.0000, 0.0000, 0.7192, 0.2346, 0.4038, 0.2050, 0.5138, 0.5650],
  17843. [0.6163, 0.4006, 0.8788, 0.4683, 0.3663, 0.4883, 0.5888, 0.5017],
  17844. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  17845. [0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367]]],
  17846. device='cuda:0')
  17847. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  17848. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  17849. loss_train: 0.06951731926528737
  17850. step: 70
  17851. running loss: 0.0009931045609326767
  17852.  
  17853. Train Steps: 70/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17854. torch.Size([8, 8])
  17855. tensor([[ nan, nan, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  17856. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  17857. [0.6127, 0.4084, 0.8700, 0.4467, 0.3987, 0.4317, 0.5013, 0.5471],
  17858. [0.6276, 0.4235, 0.8888, 0.5333, 0.3800, 0.3117, 0.5427, 0.6164],
  17859. [ nan, nan, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  17860. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  17861. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  17862. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841]],
  17863. device='cuda:0', dtype=torch.float64)
  17864. predictions are: tensor([[0.1622, 0.1066, 0.7874, 0.3014, 0.3781, 0.3310, 0.5317, 0.5481],
  17865. [0.6755, 0.4376, 0.8550, 0.4936, 0.4995, 0.5143, 0.4808, 0.5582],
  17866. [0.6905, 0.4685, 0.8566, 0.4401, 0.3896, 0.4463, 0.4683, 0.5373],
  17867. [0.7136, 0.4800, 0.8401, 0.5447, 0.4031, 0.3674, 0.5622, 0.5945],
  17868. [0.0378, 0.0286, 0.8293, 0.2527, 0.5296, 0.2630, 0.6928, 0.5580],
  17869. [0.6766, 0.4412, 0.8797, 0.4461, 0.4491, 0.5980, 0.6137, 0.5414],
  17870. [0.6949, 0.4499, 0.8717, 0.4481, 0.4151, 0.5993, 0.6216, 0.5185],
  17871. [0.6602, 0.4530, 0.9019, 0.4224, 0.4236, 0.3748, 0.6915, 0.5616]],
  17872. device='cuda:0', grad_fn=<AddmmBackward>)
  17873. landmarks are: tensor([[[0.0000, 0.0000, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  17874. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  17875. [0.6127, 0.4084, 0.8700, 0.4467, 0.3988, 0.4317, 0.5013, 0.5471],
  17876. [0.6276, 0.4235, 0.8888, 0.5333, 0.3800, 0.3117, 0.5427, 0.6164],
  17877. [0.0000, 0.0000, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  17878. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  17879. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  17880. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841]]],
  17881. device='cuda:0')
  17882. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  17883. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  17884. loss_train: 0.07116618793224916
  17885. step: 71
  17886. running loss: 0.0010023406751021007
  17887. Train Steps: 71/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17888. torch.Size([8, 8])
  17889. tensor([[0.6201, 0.4082, 0.8827, 0.3715, 0.3825, 0.2712, 0.5845, 0.5412],
  17890. [0.6224, 0.4061, 0.8988, 0.4300, 0.3838, 0.4750, 0.6112, 0.5483],
  17891. [0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500],
  17892. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  17893. [0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  17894. [0.6260, 0.4153, 0.9000, 0.4533, 0.4025, 0.2633, 0.6223, 0.4967],
  17895. [0.6102, 0.3999, 0.8750, 0.5133, 0.3825, 0.4750, 0.5637, 0.5083],
  17896. [0.6199, 0.4071, 0.7600, 0.2117, 0.4037, 0.2767, 0.6138, 0.5550]],
  17897. device='cuda:0', dtype=torch.float64)
  17898. predictions are: tensor([[0.6020, 0.3793, 0.8858, 0.3604, 0.3895, 0.2816, 0.5661, 0.5512],
  17899. [0.5487, 0.3592, 0.8974, 0.4165, 0.3968, 0.4994, 0.6111, 0.5781],
  17900. [0.5426, 0.3551, 0.7878, 0.2664, 0.3672, 0.4156, 0.6022, 0.5935],
  17901. [0.5907, 0.3904, 0.8766, 0.3900, 0.3722, 0.4003, 0.5070, 0.5364],
  17902. [0.5848, 0.3734, 0.8911, 0.3657, 0.3688, 0.3688, 0.6141, 0.5162],
  17903. [0.5836, 0.3809, 0.8993, 0.4411, 0.4129, 0.2957, 0.6122, 0.5266],
  17904. [0.5586, 0.3550, 0.8771, 0.5170, 0.4175, 0.4976, 0.5237, 0.5390],
  17905. [0.6279, 0.4028, 0.7860, 0.2222, 0.4409, 0.3015, 0.6051, 0.5704]],
  17906. device='cuda:0', grad_fn=<AddmmBackward>)
  17907. landmarks are: tensor([[[0.6201, 0.4082, 0.8827, 0.3715, 0.3825, 0.2712, 0.5845, 0.5412],
  17908. [0.6224, 0.4061, 0.8988, 0.4300, 0.3837, 0.4750, 0.6112, 0.5483],
  17909. [0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500],
  17910. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  17911. [0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  17912. [0.6260, 0.4153, 0.9000, 0.4533, 0.4025, 0.2633, 0.6223, 0.4967],
  17913. [0.6102, 0.3999, 0.8750, 0.5133, 0.3825, 0.4750, 0.5638, 0.5083],
  17914. [0.6199, 0.4071, 0.7600, 0.2117, 0.4038, 0.2767, 0.6137, 0.5550]]],
  17915. device='cuda:0')
  17916. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  17917. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  17918. loss_train: 0.07192522671539336
  17919. step: 72
  17920. running loss: 0.000998961482158241
  17921. Train Steps: 72/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17922. torch.Size([8, 8])
  17923. tensor([[0.6182, 0.4099, 0.7812, 0.3000, 0.3937, 0.2367, 0.5325, 0.5750],
  17924. [0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  17925. [0.6224, 0.4061, 0.8988, 0.4300, 0.3838, 0.4750, 0.6112, 0.5483],
  17926. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  17927. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  17928. [0.6208, 0.4082, 0.8538, 0.3067, 0.3588, 0.3717, 0.6112, 0.5517],
  17929. [0.6132, 0.4066, 0.7259, 0.2402, 0.3588, 0.3300, 0.6000, 0.5600],
  17930. [0.6212, 0.4033, 0.8938, 0.4167, 0.3813, 0.4267, 0.5613, 0.5583]],
  17931. device='cuda:0', dtype=torch.float64)
  17932. predictions are: tensor([[0.5656, 0.3701, 0.8129, 0.3049, 0.4055, 0.2572, 0.5526, 0.5638],
  17933. [0.5737, 0.3879, 0.9114, 0.5173, 0.3807, 0.4555, 0.5800, 0.5659],
  17934. [0.5693, 0.3739, 0.9095, 0.4160, 0.3864, 0.4838, 0.6312, 0.5468],
  17935. [0.5836, 0.3736, 0.8785, 0.4841, 0.4992, 0.5042, 0.5113, 0.5590],
  17936. [0.5832, 0.3820, 0.8750, 0.5604, 0.4101, 0.4439, 0.5613, 0.5650],
  17937. [0.5447, 0.3638, 0.8757, 0.3092, 0.3734, 0.3732, 0.6159, 0.5459],
  17938. [0.5780, 0.3907, 0.7475, 0.2412, 0.3644, 0.3376, 0.6089, 0.5725],
  17939. [0.6014, 0.3843, 0.9054, 0.4120, 0.3852, 0.4092, 0.5635, 0.5452]],
  17940. device='cuda:0', grad_fn=<AddmmBackward>)
  17941. landmarks are: tensor([[[0.6182, 0.4099, 0.7812, 0.3000, 0.3938, 0.2367, 0.5325, 0.5750],
  17942. [0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  17943. [0.6224, 0.4061, 0.8988, 0.4300, 0.3837, 0.4750, 0.6112, 0.5483],
  17944. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  17945. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  17946. [0.6208, 0.4082, 0.8537, 0.3067, 0.3587, 0.3717, 0.6112, 0.5517],
  17947. [0.6132, 0.4066, 0.7259, 0.2402, 0.3587, 0.3300, 0.6000, 0.5600],
  17948. [0.6212, 0.4033, 0.8938, 0.4167, 0.3812, 0.4267, 0.5612, 0.5583]]],
  17949. device='cuda:0')
  17950. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  17951. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  17952. loss_train: 0.07242716127075255
  17953. step: 73
  17954. running loss: 0.000992152894119898
  17955. Train Steps: 73/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17956. torch.Size([8, 8])
  17957. tensor([[0.6199, 0.4071, 0.7600, 0.2117, 0.4037, 0.2767, 0.6138, 0.5550],
  17958. [0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6138, 0.5400],
  17959. [0.6264, 0.4248, 0.8938, 0.4183, 0.3875, 0.4100, 0.5400, 0.5350],
  17960. [0.6267, 0.4065, 0.8313, 0.2467, 0.4788, 0.1733, 0.6312, 0.5133],
  17961. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  17962. [0.6114, 0.4018, 0.7213, 0.1967, 0.3763, 0.2700, 0.5875, 0.5533],
  17963. [ nan, nan, 0.6793, 0.2110, 0.4012, 0.2167, 0.5112, 0.5583],
  17964. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098]],
  17965. device='cuda:0', dtype=torch.float64)
  17966. predictions are: tensor([[ 0.6857, 0.4347, 0.8096, 0.2185, 0.4308, 0.2854, 0.6269, 0.5558],
  17967. [ 0.6357, 0.4159, 0.9101, 0.4135, 0.3712, 0.4606, 0.6289, 0.5638],
  17968. [ 0.6290, 0.4377, 0.9140, 0.4249, 0.3806, 0.4106, 0.5434, 0.5583],
  17969. [ 0.5919, 0.3861, 0.8664, 0.2422, 0.4864, 0.1704, 0.6385, 0.5383],
  17970. [ 0.5907, 0.3928, 0.7141, 0.2533, 0.3866, 0.2558, 0.5609, 0.5914],
  17971. [ 0.6507, 0.4208, 0.7482, 0.1980, 0.3761, 0.2632, 0.5899, 0.5728],
  17972. [-0.0138, -0.0199, 0.7143, 0.2253, 0.4229, 0.2101, 0.5316, 0.5606],
  17973. [ 0.6328, 0.3930, 0.9018, 0.4740, 0.3764, 0.5228, 0.6135, 0.5193]],
  17974. device='cuda:0', grad_fn=<AddmmBackward>)
  17975. landmarks are: tensor([[[0.6199, 0.4071, 0.7600, 0.2117, 0.4038, 0.2767, 0.6137, 0.5550],
  17976. [0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6137, 0.5400],
  17977. [0.6264, 0.4248, 0.8938, 0.4183, 0.3875, 0.4100, 0.5400, 0.5350],
  17978. [0.6266, 0.4065, 0.8313, 0.2467, 0.4787, 0.1733, 0.6313, 0.5133],
  17979. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  17980. [0.6114, 0.4018, 0.7212, 0.1967, 0.3762, 0.2700, 0.5875, 0.5533],
  17981. [0.0000, 0.0000, 0.6793, 0.2110, 0.4013, 0.2167, 0.5113, 0.5583],
  17982. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098]]],
  17983. device='cuda:0')
  17984. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  17985. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  17986. loss_train: 0.07284825743408874
  17987. step: 74
  17988. running loss: 0.0009844359112714694
  17989.  
  17990. Train Steps: 74/90 Loss: 0.0010 torch.Size([8, 600, 800])
  17991. torch.Size([8, 8])
  17992. tensor([[0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921],
  17993. [0.6212, 0.4033, 0.8938, 0.4167, 0.3813, 0.4267, 0.5613, 0.5583],
  17994. [0.6289, 0.4032, 0.8419, 0.5446, 0.4075, 0.5017, 0.6312, 0.5117],
  17995. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  17996. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  17997. [0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  17998. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  17999. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533]],
  18000. device='cuda:0', dtype=torch.float64)
  18001. predictions are: tensor([[0.6102, 0.3897, 0.8551, 0.2125, 0.3776, 0.2651, 0.6252, 0.5094],
  18002. [0.6052, 0.3984, 0.9014, 0.4016, 0.3618, 0.3818, 0.5351, 0.5663],
  18003. [0.6176, 0.4032, 0.8555, 0.5277, 0.3850, 0.4669, 0.6429, 0.5278],
  18004. [0.5970, 0.3958, 0.8931, 0.4394, 0.4216, 0.5530, 0.6155, 0.5644],
  18005. [0.5029, 0.3541, 0.7187, 0.2376, 0.3972, 0.2121, 0.5850, 0.5618],
  18006. [0.5635, 0.3682, 0.8465, 0.2279, 0.4235, 0.1639, 0.6123, 0.5153],
  18007. [0.5826, 0.3937, 0.8821, 0.4890, 0.4012, 0.4987, 0.5960, 0.5400],
  18008. [0.5946, 0.4059, 0.8974, 0.4093, 0.4117, 0.4826, 0.5654, 0.5658]],
  18009. device='cuda:0', grad_fn=<AddmmBackward>)
  18010. landmarks are: tensor([[[0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921],
  18011. [0.6212, 0.4033, 0.8938, 0.4167, 0.3812, 0.4267, 0.5612, 0.5583],
  18012. [0.6289, 0.4031, 0.8419, 0.5446, 0.4075, 0.5017, 0.6313, 0.5117],
  18013. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  18014. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  18015. [0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  18016. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  18017. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533]]],
  18018. device='cuda:0')
  18019. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  18020. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  18021. loss_train: 0.0736210917821154
  18022. step: 75
  18023. running loss: 0.000981614557094872
  18024. Train Steps: 75/90 Loss: 0.0010 torch.Size([8, 600, 800])
  18025. torch.Size([8, 8])
  18026. tensor([[0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  18027. [0.6241, 0.4143, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  18028. [0.6202, 0.4079, 0.8025, 0.2500, 0.3763, 0.3217, 0.6125, 0.5533],
  18029. [0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  18030. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5363, 0.5550],
  18031. [0.6102, 0.4005, 0.8688, 0.5100, 0.4813, 0.5400, 0.5404, 0.5064],
  18032. [0.6332, 0.4118, 0.9238, 0.4267, 0.4012, 0.4733, 0.7525, 0.5436],
  18033. [0.6255, 0.4017, 0.8688, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901]],
  18034. device='cuda:0', dtype=torch.float64)
  18035. predictions are: tensor([[0.6487, 0.4125, 0.8374, 0.5678, 0.3638, 0.4479, 0.6072, 0.4584],
  18036. [0.5668, 0.3660, 0.8805, 0.4362, 0.3959, 0.5112, 0.6293, 0.5503],
  18037. [0.5917, 0.3874, 0.7891, 0.2344, 0.3570, 0.2849, 0.6054, 0.5305],
  18038. [0.5951, 0.3889, 0.8693, 0.4499, 0.3407, 0.3709, 0.5155, 0.5738],
  18039. [0.6648, 0.4452, 0.7050, 0.2302, 0.3687, 0.1957, 0.5489, 0.5349],
  18040. [0.5661, 0.3774, 0.8431, 0.4716, 0.4545, 0.4936, 0.5221, 0.5009],
  18041. [0.6120, 0.4011, 0.9029, 0.4008, 0.3748, 0.4457, 0.7348, 0.5422],
  18042. [0.6432, 0.4102, 0.8724, 0.3048, 0.3337, 0.3145, 0.6142, 0.4937]],
  18043. device='cuda:0', grad_fn=<AddmmBackward>)
  18044. landmarks are: tensor([[[0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  18045. [0.6241, 0.4142, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  18046. [0.6202, 0.4079, 0.8025, 0.2500, 0.3762, 0.3217, 0.6125, 0.5533],
  18047. [0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  18048. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5362, 0.5550],
  18049. [0.6102, 0.4005, 0.8687, 0.5100, 0.4812, 0.5400, 0.5404, 0.5064],
  18050. [0.6332, 0.4118, 0.9237, 0.4267, 0.4013, 0.4733, 0.7525, 0.5436],
  18051. [0.6255, 0.4017, 0.8687, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901]]],
  18052. device='cuda:0')
  18053. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  18054. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  18055. loss_train: 0.07427169638685882
  18056. step: 76
  18057. running loss: 0.0009772591629849845
  18058. Train Steps: 76/90 Loss: 0.0010 torch.Size([8, 600, 800])
  18059. torch.Size([8, 8])
  18060. tensor([[0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  18061. [0.6125, 0.4035, 0.7825, 0.3100, 0.3463, 0.4900, 0.5832, 0.5637],
  18062. [0.6186, 0.4060, 0.8750, 0.5050, 0.3538, 0.4367, 0.5813, 0.6083],
  18063. [0.6275, 0.4048, 0.8488, 0.2883, 0.4463, 0.2033, 0.6321, 0.5155],
  18064. [0.6259, 0.4133, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  18065. [0.6260, 0.4214, 0.8538, 0.5500, 0.3663, 0.3767, 0.5587, 0.5983],
  18066. [0.6255, 0.4017, 0.8688, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  18067. [0.6216, 0.4167, 0.8588, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667]],
  18068. device='cuda:0', dtype=torch.float64)
  18069. predictions are: tensor([[0.6554, 0.4331, 0.7440, 0.2241, 0.3653, 0.3191, 0.6177, 0.5078],
  18070. [0.5756, 0.3785, 0.7637, 0.3179, 0.3499, 0.4842, 0.5989, 0.5165],
  18071. [0.6087, 0.3967, 0.8508, 0.5047, 0.3572, 0.4255, 0.5865, 0.5805],
  18072. [0.6427, 0.4099, 0.8366, 0.2696, 0.4381, 0.2028, 0.6484, 0.4932],
  18073. [0.6039, 0.3984, 0.8158, 0.2238, 0.4992, 0.1749, 0.6391, 0.4684],
  18074. [0.6422, 0.4182, 0.8247, 0.5406, 0.3554, 0.3843, 0.5606, 0.5869],
  18075. [0.6590, 0.4183, 0.8570, 0.3301, 0.3534, 0.3471, 0.6224, 0.4780],
  18076. [0.6645, 0.4232, 0.8241, 0.5647, 0.3989, 0.5272, 0.6056, 0.5310]],
  18077. device='cuda:0', grad_fn=<AddmmBackward>)
  18078. landmarks are: tensor([[[0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  18079. [0.6125, 0.4035, 0.7825, 0.3100, 0.3462, 0.4900, 0.5832, 0.5637],
  18080. [0.6186, 0.4060, 0.8750, 0.5050, 0.3537, 0.4367, 0.5813, 0.6083],
  18081. [0.6275, 0.4048, 0.8487, 0.2883, 0.4462, 0.2033, 0.6321, 0.5155],
  18082. [0.6259, 0.4132, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  18083. [0.6260, 0.4214, 0.8537, 0.5500, 0.3663, 0.3767, 0.5587, 0.5983],
  18084. [0.6255, 0.4017, 0.8687, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  18085. [0.6216, 0.4167, 0.8587, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667]]],
  18086. device='cuda:0')
  18087. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  18088. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  18089. loss_train: 0.07463139062747359
  18090. step: 77
  18091. running loss: 0.0009692388393178388
  18092. Train Steps: 77/90 Loss: 0.0010 torch.Size([8, 600, 800])
  18093. torch.Size([8, 8])
  18094. tensor([[0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  18095. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  18096. [0.6214, 0.4112, 0.7838, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  18097. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  18098. [0.6274, 0.4270, 0.8938, 0.4967, 0.3550, 0.4283, 0.5700, 0.5733],
  18099. [0.6199, 0.4065, 0.7598, 0.2385, 0.4317, 0.1981, 0.5933, 0.5221],
  18100. [0.6266, 0.4070, 0.8712, 0.5600, 0.3713, 0.4783, 0.5775, 0.6100],
  18101. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167]],
  18102. device='cuda:0', dtype=torch.float64)
  18103. predictions are: tensor([[0.6023, 0.3914, 0.8483, 0.4306, 0.3720, 0.3634, 0.5348, 0.5365],
  18104. [0.6116, 0.4021, 0.8334, 0.4468, 0.4512, 0.2645, 0.5703, 0.5711],
  18105. [0.6597, 0.4230, 0.7794, 0.1988, 0.3646, 0.2922, 0.6104, 0.4579],
  18106. [0.5780, 0.3787, 0.7765, 0.3631, 0.3513, 0.4057, 0.5427, 0.5166],
  18107. [0.6164, 0.3919, 0.8649, 0.4805, 0.3788, 0.4427, 0.6133, 0.5166],
  18108. [0.6889, 0.4362, 0.7362, 0.2125, 0.4312, 0.2173, 0.6049, 0.4742],
  18109. [0.5940, 0.3791, 0.8297, 0.5513, 0.3694, 0.4709, 0.6086, 0.5311],
  18110. [0.6342, 0.4165, 0.7445, 0.2720, 0.4020, 0.2731, 0.6232, 0.5852]],
  18111. device='cuda:0', grad_fn=<AddmmBackward>)
  18112. landmarks are: tensor([[[0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  18113. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  18114. [0.6214, 0.4112, 0.7837, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  18115. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  18116. [0.6274, 0.4270, 0.8938, 0.4967, 0.3550, 0.4283, 0.5700, 0.5733],
  18117. [0.6199, 0.4065, 0.7598, 0.2385, 0.4317, 0.1981, 0.5933, 0.5221],
  18118. [0.6266, 0.4070, 0.8712, 0.5600, 0.3713, 0.4783, 0.5775, 0.6100],
  18119. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167]]],
  18120. device='cuda:0')
  18121. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  18122. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  18123. loss_train: 0.07542235916480422
  18124. step: 78
  18125. running loss: 0.0009669533226256951
  18126.  
  18127. Train Steps: 78/90 Loss: 0.0010 torch.Size([8, 600, 800])
  18128. torch.Size([8, 8])
  18129. tensor([[0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  18130. [0.6275, 0.4024, 0.8500, 0.5383, 0.3912, 0.4883, 0.6288, 0.5100],
  18131. [0.6226, 0.4098, 0.8912, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  18132. [0.6197, 0.4091, 0.8800, 0.4783, 0.3538, 0.4767, 0.5950, 0.5550],
  18133. [0.6346, 0.4086, 0.7938, 0.5500, 0.3962, 0.4867, 0.7343, 0.5702],
  18134. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  18135. [0.6350, 0.4043, 0.8738, 0.5650, 0.3850, 0.4750, 0.6401, 0.4950],
  18136. [0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377]],
  18137. device='cuda:0', dtype=torch.float64)
  18138. predictions are: tensor([[0.6322, 0.4329, 0.8176, 0.2788, 0.4258, 0.2343, 0.5572, 0.5423],
  18139. [0.6476, 0.4297, 0.8261, 0.5259, 0.4003, 0.5108, 0.5924, 0.4821],
  18140. [0.6028, 0.4122, 0.8890, 0.4161, 0.4090, 0.2621, 0.5544, 0.5396],
  18141. [0.6388, 0.4294, 0.8767, 0.4735, 0.3778, 0.4834, 0.5753, 0.5458],
  18142. [0.6263, 0.4230, 0.8026, 0.5230, 0.3972, 0.5063, 0.6677, 0.5665],
  18143. [0.6590, 0.4498, 0.7379, 0.1787, 0.4037, 0.2600, 0.6136, 0.5496],
  18144. [0.6252, 0.4112, 0.8502, 0.5726, 0.3917, 0.4984, 0.6056, 0.4986],
  18145. [0.6329, 0.4150, 0.8400, 0.5292, 0.4007, 0.4738, 0.5812, 0.5399]],
  18146. device='cuda:0', grad_fn=<AddmmBackward>)
  18147. landmarks are: tensor([[[0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  18148. [0.6275, 0.4024, 0.8500, 0.5383, 0.3913, 0.4883, 0.6288, 0.5100],
  18149. [0.6226, 0.4098, 0.8913, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  18150. [0.6197, 0.4091, 0.8800, 0.4783, 0.3537, 0.4767, 0.5950, 0.5550],
  18151. [0.6346, 0.4086, 0.7937, 0.5500, 0.3963, 0.4867, 0.7343, 0.5702],
  18152. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  18153. [0.6350, 0.4043, 0.8737, 0.5650, 0.3850, 0.4750, 0.6401, 0.4950],
  18154. [0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377]]],
  18155. device='cuda:0')
  18156. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  18157. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  18158. loss_train: 0.07581857420154847
  18159. step: 79
  18160. running loss: 0.000959728787361373
  18161. Train Steps: 79/90 Loss: 0.0010 torch.Size([8, 600, 800])
  18162. torch.Size([8, 8])
  18163. tensor([[0.6293, 0.4024, 0.8750, 0.5000, 0.4012, 0.5733, 0.7121, 0.5633],
  18164. [0.6276, 0.4120, 0.8738, 0.3133, 0.4225, 0.2217, 0.6203, 0.4892],
  18165. [0.6353, 0.4128, 0.9138, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  18166. [0.6114, 0.4018, 0.7213, 0.1967, 0.3763, 0.2700, 0.5875, 0.5533],
  18167. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  18168. [0.6299, 0.4303, 0.7963, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  18169. [0.6265, 0.4088, 0.8025, 0.1850, 0.4163, 0.2500, 0.6290, 0.4947],
  18170. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426]],
  18171. device='cuda:0', dtype=torch.float64)
  18172. predictions are: tensor([[0.6582, 0.4217, 0.8559, 0.5431, 0.3977, 0.5892, 0.6823, 0.5507],
  18173. [0.5421, 0.3751, 0.8435, 0.3560, 0.4198, 0.2311, 0.5940, 0.4788],
  18174. [0.5280, 0.3593, 0.8839, 0.3768, 0.4569, 0.3085, 0.7006, 0.6056],
  18175. [0.6138, 0.4281, 0.6970, 0.2340, 0.3655, 0.2705, 0.5795, 0.5592],
  18176. [0.6089, 0.4218, 0.8466, 0.4879, 0.4248, 0.5155, 0.5631, 0.5667],
  18177. [0.6077, 0.4128, 0.7871, 0.4091, 0.4669, 0.2647, 0.5149, 0.6082],
  18178. [0.6323, 0.4149, 0.7712, 0.2223, 0.4029, 0.2477, 0.6125, 0.5030],
  18179. [0.6510, 0.4502, 0.7419, 0.2712, 0.4474, 0.1928, 0.5697, 0.5589]],
  18180. device='cuda:0', grad_fn=<AddmmBackward>)
  18181. landmarks are: tensor([[[0.6293, 0.4024, 0.8750, 0.5000, 0.4013, 0.5733, 0.7121, 0.5633],
  18182. [0.6276, 0.4120, 0.8737, 0.3133, 0.4225, 0.2217, 0.6203, 0.4892],
  18183. [0.6353, 0.4128, 0.9137, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  18184. [0.6114, 0.4018, 0.7212, 0.1967, 0.3762, 0.2700, 0.5875, 0.5533],
  18185. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  18186. [0.6299, 0.4303, 0.7962, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  18187. [0.6265, 0.4088, 0.8025, 0.1850, 0.4162, 0.2500, 0.6290, 0.4947],
  18188. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426]]],
  18189. device='cuda:0')
  18190. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  18191. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  18192. loss_train: 0.07664816980832256
  18193. step: 80
  18194. running loss: 0.000958102122604032
  18195. Train Steps: 80/90 Loss: 0.0010 torch.Size([8, 600, 800])
  18196. torch.Size([8, 8])
  18197. tensor([[0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  18198. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  18199. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  18200. [0.6132, 0.4118, 0.8200, 0.3633, 0.3563, 0.5400, 0.5787, 0.5136],
  18201. [0.6113, 0.4088, 0.6859, 0.2208, 0.4363, 0.1700, 0.5188, 0.5533],
  18202. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  18203. [0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274],
  18204. [0.6166, 0.4008, 0.8563, 0.5667, 0.4388, 0.4933, 0.5575, 0.5567]],
  18205. device='cuda:0', dtype=torch.float64)
  18206. predictions are: tensor([[0.6689, 0.4372, 0.9093, 0.4226, 0.4324, 0.3600, 0.7135, 0.6139],
  18207. [0.6498, 0.4387, 0.8441, 0.5649, 0.3814, 0.4846, 0.7158, 0.5975],
  18208. [0.6646, 0.4442, 0.8671, 0.4075, 0.3482, 0.3767, 0.5877, 0.5507],
  18209. [0.6307, 0.4162, 0.8190, 0.3802, 0.3695, 0.5152, 0.5928, 0.5423],
  18210. [0.6437, 0.4392, 0.6882, 0.2361, 0.4377, 0.1644, 0.5355, 0.5713],
  18211. [0.6287, 0.4257, 0.8331, 0.5270, 0.4388, 0.5462, 0.5233, 0.5611],
  18212. [0.6295, 0.4227, 0.8565, 0.3296, 0.4674, 0.2036, 0.6249, 0.5424],
  18213. [0.5897, 0.3983, 0.8491, 0.5866, 0.4412, 0.4888, 0.5794, 0.5939]],
  18214. device='cuda:0', grad_fn=<AddmmBackward>)
  18215. landmarks are: tensor([[[0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  18216. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  18217. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  18218. [0.6132, 0.4118, 0.8200, 0.3633, 0.3562, 0.5400, 0.5787, 0.5136],
  18219. [0.6113, 0.4088, 0.6859, 0.2208, 0.4363, 0.1700, 0.5188, 0.5533],
  18220. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  18221. [0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274],
  18222. [0.6166, 0.4008, 0.8562, 0.5667, 0.4387, 0.4933, 0.5575, 0.5567]]],
  18223. device='cuda:0')
  18224. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  18225. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  18226. loss_train: 0.07702205292298459
  18227. step: 81
  18228. running loss: 0.0009508895422590689
  18229. Train Steps: 81/90 Loss: 0.0010 torch.Size([8, 600, 800])
  18230. torch.Size([8, 8])
  18231. tensor([[0.6263, 0.4030, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  18232. [ nan, nan, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  18233. [0.6226, 0.4098, 0.8912, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  18234. [0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  18235. [0.6286, 0.4086, 0.8408, 0.2801, 0.4163, 0.2800, 0.6725, 0.5393],
  18236. [0.6168, 0.4029, 0.8523, 0.3417, 0.3588, 0.5000, 0.6125, 0.5400],
  18237. [0.6250, 0.4106, 0.8700, 0.3717, 0.3588, 0.4967, 0.6038, 0.5167],
  18238. [ nan, nan, 0.7525, 0.2291, 0.3838, 0.3017, 0.6050, 0.5667]],
  18239. device='cuda:0', dtype=torch.float64)
  18240. predictions are: tensor([[0.7931, 0.5002, 0.9014, 0.4972, 0.3986, 0.5360, 0.6156, 0.5124],
  18241. [0.2200, 0.1395, 0.9035, 0.3557, 0.5113, 0.2155, 0.6968, 0.6014],
  18242. [0.7179, 0.4789, 0.8840, 0.4460, 0.4199, 0.2446, 0.5541, 0.5812],
  18243. [0.7196, 0.4776, 0.8282, 0.5575, 0.4153, 0.4038, 0.6555, 0.5852],
  18244. [0.7195, 0.4702, 0.8308, 0.2887, 0.4164, 0.2733, 0.6438, 0.5775],
  18245. [0.7504, 0.4889, 0.8567, 0.3720, 0.3693, 0.4779, 0.6205, 0.5722],
  18246. [0.7627, 0.4916, 0.8883, 0.4007, 0.3814, 0.4807, 0.6480, 0.5714],
  18247. [0.1268, 0.0987, 0.7463, 0.2425, 0.3991, 0.2815, 0.5947, 0.6059]],
  18248. device='cuda:0', grad_fn=<AddmmBackward>)
  18249. landmarks are: tensor([[[0.6263, 0.4029, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  18250. [0.0000, 0.0000, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  18251. [0.6226, 0.4098, 0.8913, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  18252. [0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  18253. [0.6286, 0.4086, 0.8408, 0.2801, 0.4162, 0.2800, 0.6725, 0.5393],
  18254. [0.6168, 0.4029, 0.8523, 0.3417, 0.3587, 0.5000, 0.6125, 0.5400],
  18255. [0.6250, 0.4105, 0.8700, 0.3717, 0.3587, 0.4967, 0.6037, 0.5167],
  18256. [0.0000, 0.0000, 0.7525, 0.2291, 0.3837, 0.3017, 0.6050, 0.5667]]],
  18257. device='cuda:0')
  18258. loss_train_step before backward: tensor(0.0039, device='cuda:0', grad_fn=<MseLossBackward>)
  18259. loss_train_step after backward: tensor(0.0039, device='cuda:0', grad_fn=<MseLossBackward>)
  18260. loss_train: 0.08094305740087293
  18261. step: 82
  18262. running loss: 0.0009871104561082066
  18263.  
  18264. Train Steps: 82/90 Loss: 0.0010 torch.Size([8, 600, 800])
  18265. torch.Size([8, 8])
  18266. tensor([[0.6182, 0.4099, 0.7812, 0.3000, 0.3937, 0.2367, 0.5325, 0.5750],
  18267. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  18268. [0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  18269. [0.6333, 0.4037, 0.8638, 0.5733, 0.4012, 0.4717, 0.6369, 0.4938],
  18270. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6038, 0.6167],
  18271. [ nan, nan, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600],
  18272. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  18273. [0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133]],
  18274. device='cuda:0', dtype=torch.float64)
  18275. predictions are: tensor([[0.6775, 0.4409, 0.8038, 0.2923, 0.4102, 0.2379, 0.5431, 0.5955],
  18276. [0.7007, 0.4416, 0.8367, 0.5775, 0.4175, 0.4591, 0.5865, 0.5586],
  18277. [0.6926, 0.4455, 0.8810, 0.4902, 0.4014, 0.5026, 0.7283, 0.5516],
  18278. [0.6904, 0.4220, 0.8995, 0.5639, 0.4153, 0.4616, 0.6263, 0.5324],
  18279. [0.6587, 0.4276, 0.8119, 0.3081, 0.3876, 0.3840, 0.6036, 0.6393],
  18280. [0.1182, 0.0820, 0.8003, 0.2678, 0.4247, 0.2427, 0.5935, 0.6027],
  18281. [0.6364, 0.4076, 0.9295, 0.4042, 0.3742, 0.3750, 0.6706, 0.5503],
  18282. [0.6407, 0.4136, 0.8899, 0.5239, 0.3889, 0.4709, 0.6358, 0.5303]],
  18283. device='cuda:0', grad_fn=<AddmmBackward>)
  18284. landmarks are: tensor([[[0.6182, 0.4099, 0.7812, 0.3000, 0.3938, 0.2367, 0.5325, 0.5750],
  18285. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  18286. [0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  18287. [0.6334, 0.4037, 0.8637, 0.5733, 0.4013, 0.4717, 0.6369, 0.4938],
  18288. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6037, 0.6167],
  18289. [0.0000, 0.0000, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600],
  18290. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  18291. [0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133]]],
  18292. device='cuda:0')
  18293. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  18294. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  18295. loss_train: 0.0819609404716175
  18296. step: 83
  18297. running loss: 0.0009874812105014157
  18298. Train Steps: 83/90 Loss: 0.0010 torch.Size([8, 600, 800])
  18299. torch.Size([8, 8])
  18300. tensor([[0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  18301. [0.6179, 0.4082, 0.6688, 0.2667, 0.3588, 0.3317, 0.5750, 0.5783],
  18302. [0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  18303. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200],
  18304. [0.6200, 0.4059, 0.8700, 0.4900, 0.4163, 0.5000, 0.6162, 0.5467],
  18305. [0.6332, 0.4165, 0.9100, 0.3350, 0.4188, 0.3683, 0.7438, 0.5528],
  18306. [0.6100, 0.4016, 0.8600, 0.5067, 0.4612, 0.5233, 0.5086, 0.5519],
  18307. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817]],
  18308. device='cuda:0', dtype=torch.float64)
  18309. predictions are: tensor([[0.6161, 0.3941, 0.7660, 0.3402, 0.4910, 0.1819, 0.5801, 0.6160],
  18310. [0.6550, 0.4167, 0.7156, 0.2698, 0.3404, 0.3261, 0.6140, 0.5784],
  18311. [0.5628, 0.3644, 0.9160, 0.4328, 0.3729, 0.5294, 0.5967, 0.5198],
  18312. [0.5325, 0.3280, 0.8888, 0.3057, 0.4375, 0.2048, 0.6688, 0.5315],
  18313. [0.5760, 0.3584, 0.9019, 0.4981, 0.4276, 0.5066, 0.6589, 0.5600],
  18314. [0.6087, 0.3825, 0.9356, 0.3558, 0.4076, 0.3620, 0.7728, 0.5539],
  18315. [0.5557, 0.3471, 0.8858, 0.5116, 0.4644, 0.5172, 0.5522, 0.5472],
  18316. [0.5297, 0.3323, 0.8645, 0.5822, 0.4484, 0.5120, 0.5551, 0.5111]],
  18317. device='cuda:0', grad_fn=<AddmmBackward>)
  18318. landmarks are: tensor([[[0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  18319. [0.6179, 0.4082, 0.6687, 0.2667, 0.3587, 0.3317, 0.5750, 0.5783],
  18320. [0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  18321. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200],
  18322. [0.6199, 0.4059, 0.8700, 0.4900, 0.4162, 0.5000, 0.6162, 0.5467],
  18323. [0.6332, 0.4165, 0.9100, 0.3350, 0.4187, 0.3683, 0.7438, 0.5528],
  18324. [0.6100, 0.4016, 0.8600, 0.5067, 0.4613, 0.5233, 0.5086, 0.5519],
  18325. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817]]],
  18326. device='cuda:0')
  18327. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  18328. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  18329. loss_train: 0.08303061101469211
  18330. step: 84
  18331. running loss: 0.0009884596549368109
  18332. Train Steps: 84/90 Loss: 0.0010 torch.Size([8, 600, 800])
  18333. torch.Size([8, 8])
  18334. tensor([[0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6138, 0.5400],
  18335. [0.6336, 0.4191, 0.8938, 0.5167, 0.3937, 0.3517, 0.7343, 0.5748],
  18336. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  18337. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  18338. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  18339. [0.6307, 0.4045, 0.8025, 0.5833, 0.3775, 0.4867, 0.6892, 0.5459],
  18340. [0.6219, 0.4114, 0.8175, 0.2817, 0.3925, 0.2783, 0.5900, 0.5350],
  18341. [0.6201, 0.4151, 0.8588, 0.5467, 0.3700, 0.3950, 0.5637, 0.5933]],
  18342. device='cuda:0', dtype=torch.float64)
  18343. predictions are: tensor([[0.5600, 0.3587, 0.8868, 0.3793, 0.3792, 0.4846, 0.6134, 0.5291],
  18344. [0.6271, 0.4055, 0.9044, 0.4859, 0.4086, 0.3777, 0.7089, 0.5524],
  18345. [0.5926, 0.3717, 0.8923, 0.5424, 0.3704, 0.4701, 0.6231, 0.5157],
  18346. [0.6262, 0.3905, 0.9014, 0.4869, 0.3972, 0.5699, 0.7311, 0.5323],
  18347. [0.5637, 0.3538, 0.9013, 0.3600, 0.4607, 0.2612, 0.6273, 0.5327],
  18348. [0.5475, 0.3581, 0.8491, 0.5177, 0.4012, 0.4949, 0.6715, 0.5227],
  18349. [0.4972, 0.3167, 0.8166, 0.2413, 0.4067, 0.2753, 0.5941, 0.5354],
  18350. [0.5164, 0.3416, 0.8693, 0.5236, 0.3773, 0.4062, 0.5634, 0.5922]],
  18351. device='cuda:0', grad_fn=<AddmmBackward>)
  18352. landmarks are: tensor([[[0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6137, 0.5400],
  18353. [0.6336, 0.4191, 0.8938, 0.5167, 0.3938, 0.3517, 0.7343, 0.5748],
  18354. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  18355. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  18356. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  18357. [0.6307, 0.4045, 0.8025, 0.5833, 0.3775, 0.4867, 0.6892, 0.5459],
  18358. [0.6219, 0.4114, 0.8175, 0.2817, 0.3925, 0.2783, 0.5900, 0.5350],
  18359. [0.6202, 0.4151, 0.8587, 0.5467, 0.3700, 0.3950, 0.5638, 0.5933]]],
  18360. device='cuda:0')
  18361. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  18362. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  18363. loss_train: 0.08437361448886804
  18364. step: 85
  18365. running loss: 0.0009926307586925651
  18366. Train Steps: 85/90 Loss: 0.0010 torch.Size([8, 600, 800])
  18367. torch.Size([8, 8])
  18368. tensor([[ nan, nan, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733],
  18369. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  18370. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  18371. [ nan, nan, 0.6688, 0.2513, 0.4113, 0.2117, 0.5193, 0.5933],
  18372. [0.6196, 0.4094, 0.7562, 0.2817, 0.3937, 0.3183, 0.6013, 0.6183],
  18373. [0.6222, 0.4169, 0.8638, 0.5650, 0.4313, 0.4783, 0.5637, 0.5633],
  18374. [0.6185, 0.4042, 0.7700, 0.2250, 0.4062, 0.2117, 0.5763, 0.5150],
  18375. [0.6176, 0.3911, 0.8738, 0.4217, 0.3488, 0.4033, 0.6025, 0.4817]],
  18376. device='cuda:0', dtype=torch.float64)
  18377. predictions are: tensor([[0.0196, 0.0066, 0.7102, 0.2163, 0.4524, 0.2058, 0.6032, 0.5641],
  18378. [0.6906, 0.4361, 0.8928, 0.5351, 0.4137, 0.5438, 0.6143, 0.5024],
  18379. [0.6659, 0.4039, 0.9054, 0.3828, 0.3644, 0.4686, 0.7225, 0.4885],
  18380. [0.0653, 0.0219, 0.6876, 0.2563, 0.4154, 0.2180, 0.5575, 0.5544],
  18381. [0.6370, 0.3989, 0.7716, 0.3058, 0.3992, 0.3338, 0.6395, 0.5891],
  18382. [0.6990, 0.4554, 0.8908, 0.5944, 0.4268, 0.5075, 0.5973, 0.5411],
  18383. [0.6624, 0.4194, 0.7643, 0.2401, 0.4091, 0.2334, 0.6098, 0.5052],
  18384. [0.6629, 0.4042, 0.8942, 0.4275, 0.3569, 0.4162, 0.6152, 0.4653]],
  18385. device='cuda:0', grad_fn=<AddmmBackward>)
  18386. landmarks are: tensor([[[0.0000, 0.0000, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733],
  18387. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  18388. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  18389. [0.0000, 0.0000, 0.6688, 0.2513, 0.4112, 0.2117, 0.5193, 0.5933],
  18390. [0.6196, 0.4094, 0.7563, 0.2817, 0.3938, 0.3183, 0.6012, 0.6183],
  18391. [0.6222, 0.4169, 0.8637, 0.5650, 0.4313, 0.4783, 0.5638, 0.5633],
  18392. [0.6184, 0.4042, 0.7700, 0.2250, 0.4062, 0.2117, 0.5763, 0.5150],
  18393. [0.6176, 0.3911, 0.8737, 0.4217, 0.3487, 0.4033, 0.6025, 0.4817]]],
  18394. device='cuda:0')
  18395. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  18396. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  18397. loss_train: 0.08524726060568355
  18398. step: 86
  18399. running loss: 0.0009912472163451575
  18400.  
  18401. Train Steps: 86/90 Loss: 0.0010 torch.Size([8, 600, 800])
  18402. torch.Size([8, 8])
  18403. tensor([[0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  18404. [0.6273, 0.4105, 0.8988, 0.4517, 0.3912, 0.2550, 0.5894, 0.4811],
  18405. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  18406. [0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510],
  18407. [0.6102, 0.4005, 0.8688, 0.5100, 0.4813, 0.5400, 0.5404, 0.5064],
  18408. [0.6133, 0.4066, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358],
  18409. [0.6126, 0.3954, 0.8538, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  18410. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373]],
  18411. device='cuda:0', dtype=torch.float64)
  18412. predictions are: tensor([[0.5647, 0.3489, 0.9124, 0.4385, 0.4083, 0.3114, 0.7018, 0.5291],
  18413. [0.5665, 0.3577, 0.9090, 0.4379, 0.3979, 0.2554, 0.6149, 0.4734],
  18414. [0.5564, 0.3416, 0.8504, 0.4353, 0.3480, 0.4655, 0.5663, 0.5339],
  18415. [0.4692, 0.2868, 0.6970, 0.2790, 0.3601, 0.3093, 0.5470, 0.5174],
  18416. [0.5477, 0.3552, 0.8623, 0.5110, 0.4692, 0.5417, 0.5568, 0.4848],
  18417. [0.5594, 0.3586, 0.6936, 0.2703, 0.3643, 0.2353, 0.5577, 0.4833],
  18418. [0.6098, 0.3628, 0.8656, 0.5068, 0.4168, 0.4914, 0.5540, 0.5095],
  18419. [0.5030, 0.3005, 0.8339, 0.5956, 0.3856, 0.4719, 0.5939, 0.5110]],
  18420. device='cuda:0', grad_fn=<AddmmBackward>)
  18421. landmarks are: tensor([[[0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  18422. [0.6273, 0.4105, 0.8988, 0.4517, 0.3913, 0.2550, 0.5894, 0.4811],
  18423. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  18424. [0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510],
  18425. [0.6102, 0.4005, 0.8687, 0.5100, 0.4812, 0.5400, 0.5404, 0.5064],
  18426. [0.6133, 0.4065, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358],
  18427. [0.6126, 0.3954, 0.8537, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  18428. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373]]],
  18429. device='cuda:0')
  18430. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  18431. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  18432. loss_train: 0.08683439713786356
  18433. step: 87
  18434. running loss: 0.000998096518826018
  18435. Train Steps: 87/90 Loss: 0.0010 torch.Size([8, 600, 800])
  18436. torch.Size([8, 8])
  18437. tensor([[0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  18438. [0.6198, 0.4115, 0.7762, 0.2717, 0.3713, 0.3200, 0.5837, 0.5683],
  18439. [0.6104, 0.4029, 0.8738, 0.4900, 0.4088, 0.4533, 0.5070, 0.5510],
  18440. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  18441. [0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090],
  18442. [0.6256, 0.4199, 0.8638, 0.5800, 0.3987, 0.4383, 0.5600, 0.5950],
  18443. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864],
  18444. [0.6308, 0.3990, 0.8688, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133]],
  18445. device='cuda:0', dtype=torch.float64)
  18446. predictions are: tensor([[0.5745, 0.3609, 0.8605, 0.4861, 0.3773, 0.5457, 0.6321, 0.4751],
  18447. [0.6006, 0.3970, 0.7590, 0.2772, 0.3567, 0.3082, 0.5580, 0.5534],
  18448. [0.5465, 0.3792, 0.8374, 0.4923, 0.3945, 0.4651, 0.4942, 0.5230],
  18449. [0.5758, 0.3915, 0.8660, 0.4992, 0.3439, 0.4264, 0.6091, 0.4961],
  18450. [0.5833, 0.3887, 0.8603, 0.3826, 0.3814, 0.2886, 0.6262, 0.4685],
  18451. [0.5718, 0.3936, 0.8292, 0.5865, 0.3843, 0.4386, 0.5741, 0.5793],
  18452. [0.5968, 0.3960, 0.8556, 0.5090, 0.3664, 0.4894, 0.6865, 0.5578],
  18453. [0.5435, 0.3496, 0.8422, 0.5232, 0.3791, 0.5010, 0.6271, 0.4993]],
  18454. device='cuda:0', grad_fn=<AddmmBackward>)
  18455. landmarks are: tensor([[[0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  18456. [0.6198, 0.4115, 0.7763, 0.2717, 0.3713, 0.3200, 0.5838, 0.5683],
  18457. [0.6104, 0.4029, 0.8737, 0.4900, 0.4087, 0.4533, 0.5070, 0.5510],
  18458. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  18459. [0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090],
  18460. [0.6256, 0.4199, 0.8637, 0.5800, 0.3988, 0.4383, 0.5600, 0.5950],
  18461. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864],
  18462. [0.6308, 0.3990, 0.8687, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133]]],
  18463. device='cuda:0')
  18464. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  18465. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  18466. loss_train: 0.0875556360988412
  18467. step: 88
  18468. running loss: 0.0009949504102141045
  18469. Train Steps: 88/90 Loss: 0.0010 torch.Size([8, 600, 800])
  18470. torch.Size([8, 8])
  18471. tensor([[0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  18472. [0.6117, 0.4019, 0.8538, 0.4067, 0.3513, 0.3583, 0.5663, 0.5133],
  18473. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  18474. [0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090],
  18475. [0.6098, 0.3991, 0.8638, 0.4717, 0.4263, 0.4967, 0.5212, 0.5650],
  18476. [0.6277, 0.4029, 0.8250, 0.2433, 0.4325, 0.2100, 0.6366, 0.5207],
  18477. [0.6275, 0.4111, 0.8463, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  18478. [0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767]],
  18479. device='cuda:0', dtype=torch.float64)
  18480. predictions are: tensor([[0.5855, 0.3954, 0.6825, 0.2724, 0.3647, 0.2915, 0.5967, 0.5634],
  18481. [0.5919, 0.3891, 0.8399, 0.4297, 0.3236, 0.3514, 0.5381, 0.5206],
  18482. [0.5814, 0.3986, 0.8560, 0.4971, 0.4336, 0.5638, 0.6024, 0.5450],
  18483. [0.5850, 0.3971, 0.8536, 0.3945, 0.3774, 0.2862, 0.6237, 0.4889],
  18484. [0.5602, 0.3785, 0.8340, 0.4917, 0.4045, 0.5007, 0.5293, 0.5463],
  18485. [0.5367, 0.3529, 0.7761, 0.2756, 0.4170, 0.2268, 0.6246, 0.5053],
  18486. [0.6315, 0.4231, 0.8110, 0.2677, 0.4568, 0.2181, 0.6119, 0.4950],
  18487. [0.5476, 0.3665, 0.8404, 0.5231, 0.3812, 0.4663, 0.5407, 0.5821]],
  18488. device='cuda:0', grad_fn=<AddmmBackward>)
  18489. landmarks are: tensor([[[0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  18490. [0.6116, 0.4019, 0.8537, 0.4067, 0.3512, 0.3583, 0.5663, 0.5133],
  18491. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  18492. [0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090],
  18493. [0.6098, 0.3991, 0.8637, 0.4717, 0.4263, 0.4967, 0.5213, 0.5650],
  18494. [0.6277, 0.4029, 0.8250, 0.2433, 0.4325, 0.2100, 0.6366, 0.5207],
  18495. [0.6275, 0.4111, 0.8462, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  18496. [0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767]]],
  18497. device='cuda:0')
  18498. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  18499. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  18500. loss_train: 0.08827467539231293
  18501. step: 89
  18502. running loss: 0.000991850285306887
  18503. Train Steps: 89/90 Loss: 0.0010 torch.Size([8, 600, 800])
  18504. torch.Size([8, 8])
  18505. tensor([[0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  18506. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  18507. [0.6284, 0.4093, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  18508. [0.6038, 0.3946, 0.8413, 0.4883, 0.3563, 0.4550, 0.5266, 0.4693],
  18509. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284],
  18510. [0.6185, 0.4042, 0.7700, 0.2250, 0.4062, 0.2117, 0.5763, 0.5150],
  18511. [0.6276, 0.4095, 0.8237, 0.2250, 0.4662, 0.1783, 0.6171, 0.4869],
  18512. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947]],
  18513. device='cuda:0', dtype=torch.float64)
  18514. predictions are: tensor([[0.5839, 0.3950, 0.8307, 0.5161, 0.3785, 0.4998, 0.6994, 0.5390],
  18515. [0.5677, 0.3877, 0.7637, 0.2760, 0.3595, 0.3222, 0.5726, 0.5647],
  18516. [0.6210, 0.4307, 0.8743, 0.4635, 0.3645, 0.3866, 0.5997, 0.5272],
  18517. [0.5621, 0.3761, 0.8384, 0.4889, 0.3639, 0.4513, 0.4973, 0.5228],
  18518. [0.5775, 0.3924, 0.7376, 0.2344, 0.4520, 0.1773, 0.5464, 0.5549],
  18519. [0.5597, 0.3808, 0.7354, 0.2337, 0.4123, 0.2073, 0.5410, 0.5483],
  18520. [0.6110, 0.4146, 0.8000, 0.2184, 0.4810, 0.1910, 0.5711, 0.5012],
  18521. [0.5807, 0.3846, 0.8845, 0.4835, 0.3652, 0.4919, 0.6218, 0.5216]],
  18522. device='cuda:0', grad_fn=<AddmmBackward>)
  18523. landmarks are: tensor([[[0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  18524. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  18525. [0.6284, 0.4092, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  18526. [0.6038, 0.3946, 0.8413, 0.4883, 0.3562, 0.4550, 0.5266, 0.4693],
  18527. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284],
  18528. [0.6184, 0.4042, 0.7700, 0.2250, 0.4062, 0.2117, 0.5763, 0.5150],
  18529. [0.6276, 0.4095, 0.8238, 0.2250, 0.4663, 0.1783, 0.6171, 0.4869],
  18530. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947]]],
  18531. device='cuda:0')
  18532. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  18533. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  18534. loss_train: 0.08886227456969209
  18535. step: 90
  18536. running loss: 0.000987358606329912
  18537.  
  18538. Valid Steps: 10/10 Loss: nan 9.3309
  18539. --------------------------------------------------
  18540. Epoch: 6 Train Loss: 0.0010 Valid Loss: nan
  18541. --------------------------------------------------
  18542. size of train loader is: 90
  18543. torch.Size([8, 600, 800])
  18544. torch.Size([8, 8])
  18545. tensor([[0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090],
  18546. [0.6275, 0.4081, 0.8063, 0.2017, 0.4825, 0.1583, 0.6156, 0.4869],
  18547. [0.6145, 0.4007, 0.8775, 0.4533, 0.4562, 0.5533, 0.6088, 0.5533],
  18548. [ nan, nan, 0.6512, 0.1717, 0.4100, 0.1983, 0.5253, 0.5240],
  18549. [0.6200, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913],
  18550. [0.6196, 0.4094, 0.7562, 0.2817, 0.3937, 0.3183, 0.6013, 0.6183],
  18551. [0.6277, 0.4083, 0.8350, 0.2717, 0.4562, 0.1800, 0.5918, 0.4878],
  18552. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261]],
  18553. device='cuda:0', dtype=torch.float64)
  18554. predictions are: tensor([[ 0.6453, 0.4394, 0.8678, 0.3829, 0.3862, 0.2651, 0.6238, 0.5143],
  18555. [ 0.6626, 0.4516, 0.7857, 0.2041, 0.4790, 0.1266, 0.6018, 0.5084],
  18556. [ 0.5989, 0.4005, 0.8487, 0.4353, 0.4402, 0.5201, 0.5815, 0.5510],
  18557. [-0.0018, -0.0064, 0.6547, 0.1877, 0.4051, 0.1848, 0.5274, 0.5574],
  18558. [ 0.6113, 0.3946, 0.8557, 0.5194, 0.3778, 0.4875, 0.5826, 0.4969],
  18559. [ 0.6051, 0.4128, 0.7403, 0.2913, 0.3818, 0.2822, 0.5789, 0.6328],
  18560. [ 0.6751, 0.4404, 0.8000, 0.2561, 0.4524, 0.1953, 0.5601, 0.5146],
  18561. [ 0.6628, 0.4574, 0.8744, 0.4788, 0.3909, 0.4544, 0.5270, 0.5527]],
  18562. device='cuda:0', grad_fn=<AddmmBackward>)
  18563. landmarks are: tensor([[[0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090],
  18564. [0.6275, 0.4081, 0.8062, 0.2017, 0.4825, 0.1583, 0.6156, 0.4869],
  18565. [0.6145, 0.4007, 0.8775, 0.4533, 0.4563, 0.5533, 0.6087, 0.5533],
  18566. [0.0000, 0.0000, 0.6513, 0.1717, 0.4100, 0.1983, 0.5253, 0.5240],
  18567. [0.6199, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913],
  18568. [0.6196, 0.4094, 0.7563, 0.2817, 0.3938, 0.3183, 0.6012, 0.6183],
  18569. [0.6277, 0.4083, 0.8350, 0.2717, 0.4563, 0.1800, 0.5918, 0.4878],
  18570. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261]]],
  18571. device='cuda:0')
  18572. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  18573. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  18574. loss_train: 0.0004884650697931647
  18575. step: 1
  18576. running loss: 0.0004884650697931647
  18577. Train Steps: 1/90 Loss: 0.0005 torch.Size([8, 600, 800])
  18578. torch.Size([8, 8])
  18579. tensor([[0.6261, 0.4066, 0.8325, 0.2150, 0.4763, 0.2667, 0.7002, 0.5633],
  18580. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  18581. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  18582. [0.6154, 0.4112, 0.7037, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  18583. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864],
  18584. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  18585. [0.6276, 0.4120, 0.8738, 0.3133, 0.4225, 0.2217, 0.6203, 0.4892],
  18586. [0.6286, 0.4055, 0.9000, 0.4717, 0.3763, 0.4683, 0.7018, 0.5494]],
  18587. device='cuda:0', dtype=torch.float64)
  18588. predictions are: tensor([[0.6640, 0.4397, 0.8230, 0.2081, 0.4813, 0.2426, 0.6619, 0.5524],
  18589. [0.5629, 0.3849, 0.7947, 0.1877, 0.4817, 0.1585, 0.5632, 0.4943],
  18590. [0.6622, 0.4429, 0.8487, 0.2691, 0.4924, 0.1942, 0.6363, 0.5429],
  18591. [0.5763, 0.4041, 0.6908, 0.2247, 0.4289, 0.1865, 0.4956, 0.5708],
  18592. [0.6442, 0.4362, 0.8877, 0.5054, 0.3816, 0.4920, 0.6745, 0.5856],
  18593. [0.5904, 0.4002, 0.8679, 0.4434, 0.3575, 0.4028, 0.5257, 0.5667],
  18594. [0.6361, 0.4374, 0.8812, 0.3180, 0.4441, 0.2118, 0.5942, 0.4950],
  18595. [0.6376, 0.4275, 0.9151, 0.4626, 0.3637, 0.4694, 0.6836, 0.5516]],
  18596. device='cuda:0', grad_fn=<AddmmBackward>)
  18597. landmarks are: tensor([[[0.6261, 0.4066, 0.8325, 0.2150, 0.4762, 0.2667, 0.7002, 0.5633],
  18598. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  18599. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  18600. [0.6154, 0.4112, 0.7038, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  18601. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864],
  18602. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  18603. [0.6276, 0.4120, 0.8737, 0.3133, 0.4225, 0.2217, 0.6203, 0.4892],
  18604. [0.6286, 0.4055, 0.9000, 0.4717, 0.3762, 0.4683, 0.7018, 0.5494]]],
  18605. device='cuda:0')
  18606. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  18607. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  18608. loss_train: 0.000938043842324987
  18609. step: 2
  18610. running loss: 0.0004690219211624935
  18611. Train Steps: 2/90 Loss: 0.0005 torch.Size([8, 600, 800])
  18612. torch.Size([8, 8])
  18613. tensor([[0.6193, 0.4050, 0.7313, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  18614. [0.6200, 0.3993, 0.8519, 0.4923, 0.3962, 0.4717, 0.6013, 0.5433],
  18615. [0.6245, 0.4100, 0.7762, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  18616. [0.6250, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6088, 0.5183],
  18617. [0.6284, 0.4093, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  18618. [0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6088, 0.5583],
  18619. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  18620. [0.6145, 0.4008, 0.8750, 0.5383, 0.3975, 0.4650, 0.5563, 0.5533]],
  18621. device='cuda:0', dtype=torch.float64)
  18622. predictions are: tensor([[0.5811, 0.3845, 0.7509, 0.2356, 0.4348, 0.2075, 0.5513, 0.5607],
  18623. [0.6369, 0.4152, 0.8912, 0.4861, 0.3891, 0.4627, 0.6156, 0.5446],
  18624. [0.7136, 0.4810, 0.7813, 0.2562, 0.4903, 0.1406, 0.6018, 0.5508],
  18625. [0.6665, 0.4429, 0.9152, 0.4684, 0.4712, 0.5621, 0.6198, 0.5209],
  18626. [0.7010, 0.4706, 0.9237, 0.4485, 0.3770, 0.3864, 0.6339, 0.5018],
  18627. [0.6253, 0.4285, 0.7250, 0.2029, 0.3956, 0.2703, 0.5921, 0.5407],
  18628. [0.6760, 0.4365, 0.9321, 0.4231, 0.3687, 0.3669, 0.6178, 0.5479],
  18629. [0.6130, 0.4115, 0.8999, 0.5236, 0.4055, 0.4553, 0.5712, 0.5576]],
  18630. device='cuda:0', grad_fn=<AddmmBackward>)
  18631. landmarks are: tensor([[[0.6193, 0.4050, 0.7312, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  18632. [0.6200, 0.3993, 0.8519, 0.4923, 0.3963, 0.4717, 0.6012, 0.5433],
  18633. [0.6245, 0.4100, 0.7763, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  18634. [0.6251, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6087, 0.5183],
  18635. [0.6284, 0.4092, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  18636. [0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6087, 0.5583],
  18637. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  18638. [0.6145, 0.4008, 0.8750, 0.5383, 0.3975, 0.4650, 0.5562, 0.5533]]],
  18639. device='cuda:0')
  18640. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  18641. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  18642. loss_train: 0.001644053525524214
  18643. step: 3
  18644. running loss: 0.0005480178418414047
  18645. Train Steps: 3/90 Loss: 0.0005 torch.Size([8, 600, 800])
  18646. torch.Size([8, 8])
  18647. tensor([[0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  18648. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  18649. [0.6229, 0.4066, 0.8513, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  18650. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  18651. [0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  18652. [0.6202, 0.4053, 0.8638, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  18653. [0.6260, 0.4153, 0.9000, 0.4533, 0.4025, 0.2633, 0.6223, 0.4967],
  18654. [0.6261, 0.4029, 0.8720, 0.3364, 0.3665, 0.3753, 0.6531, 0.5183]],
  18655. device='cuda:0', dtype=torch.float64)
  18656. predictions are: tensor([[0.6809, 0.4596, 0.8557, 0.4188, 0.4484, 0.2379, 0.5578, 0.6116],
  18657. [0.6355, 0.4162, 0.8816, 0.5361, 0.4015, 0.4226, 0.5648, 0.5624],
  18658. [0.6737, 0.4352, 0.8612, 0.5365, 0.4400, 0.4808, 0.6048, 0.5333],
  18659. [0.6598, 0.4225, 0.8887, 0.4296, 0.3856, 0.4134, 0.6230, 0.5433],
  18660. [0.6424, 0.3978, 0.7910, 0.1386, 0.4325, 0.2133, 0.6732, 0.5020],
  18661. [0.6612, 0.4249, 0.8707, 0.4846, 0.4547, 0.4941, 0.5900, 0.5281],
  18662. [0.6900, 0.4489, 0.9203, 0.4105, 0.4064, 0.2570, 0.6465, 0.5050],
  18663. [0.6496, 0.4127, 0.8628, 0.3158, 0.3697, 0.3593, 0.6484, 0.5083]],
  18664. device='cuda:0', grad_fn=<AddmmBackward>)
  18665. landmarks are: tensor([[[0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  18666. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  18667. [0.6229, 0.4066, 0.8512, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  18668. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  18669. [0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  18670. [0.6202, 0.4053, 0.8637, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  18671. [0.6260, 0.4153, 0.9000, 0.4533, 0.4025, 0.2633, 0.6223, 0.4967],
  18672. [0.6261, 0.4029, 0.8720, 0.3364, 0.3665, 0.3753, 0.6531, 0.5183]]],
  18673. device='cuda:0')
  18674. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  18675. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  18676. loss_train: 0.0021857938554603606
  18677. step: 4
  18678. running loss: 0.0005464484638650902
  18679.  
  18680. Train Steps: 4/90 Loss: 0.0005 torch.Size([8, 600, 800])
  18681. torch.Size([8, 8])
  18682. tensor([[ nan, nan, 0.7335, 0.2569, 0.3788, 0.2667, 0.5066, 0.5578],
  18683. [0.6296, 0.4076, 0.8400, 0.5583, 0.3700, 0.4367, 0.6876, 0.5494],
  18684. [0.6251, 0.4163, 0.8662, 0.4467, 0.3625, 0.3567, 0.6038, 0.5533],
  18685. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  18686. [0.6193, 0.4079, 0.7288, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  18687. [0.6346, 0.4086, 0.7938, 0.5500, 0.3962, 0.4867, 0.7343, 0.5702],
  18688. [0.6263, 0.4030, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  18689. [0.6310, 0.4017, 0.8563, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006]],
  18690. device='cuda:0', dtype=torch.float64)
  18691. predictions are: tensor([[0.1280, 0.0777, 0.7659, 0.2281, 0.4004, 0.2490, 0.5271, 0.5553],
  18692. [0.7083, 0.4421, 0.8876, 0.5065, 0.4055, 0.4180, 0.7057, 0.5312],
  18693. [0.7266, 0.4623, 0.9024, 0.4195, 0.3830, 0.3276, 0.6195, 0.5389],
  18694. [0.7253, 0.4551, 0.9119, 0.4579, 0.4334, 0.4596, 0.5622, 0.5170],
  18695. [0.6726, 0.4309, 0.7633, 0.2221, 0.4598, 0.2383, 0.6116, 0.6030],
  18696. [0.7279, 0.4571, 0.8371, 0.5180, 0.4143, 0.4637, 0.7164, 0.5570],
  18697. [0.7040, 0.4440, 0.9334, 0.4408, 0.3999, 0.5113, 0.6502, 0.4685],
  18698. [0.7033, 0.4410, 0.8890, 0.5672, 0.3969, 0.4621, 0.6445, 0.4977]],
  18699. device='cuda:0', grad_fn=<AddmmBackward>)
  18700. landmarks are: tensor([[[0.0000, 0.0000, 0.7335, 0.2569, 0.3787, 0.2667, 0.5066, 0.5578],
  18701. [0.6296, 0.4076, 0.8400, 0.5583, 0.3700, 0.4367, 0.6876, 0.5494],
  18702. [0.6252, 0.4162, 0.8662, 0.4467, 0.3625, 0.3567, 0.6037, 0.5533],
  18703. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  18704. [0.6193, 0.4078, 0.7287, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  18705. [0.6346, 0.4086, 0.7937, 0.5500, 0.3963, 0.4867, 0.7343, 0.5702],
  18706. [0.6263, 0.4029, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  18707. [0.6310, 0.4017, 0.8562, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006]]],
  18708. device='cuda:0')
  18709. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  18710. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  18711. loss_train: 0.004074420110555366
  18712. step: 5
  18713. running loss: 0.0008148840221110732
  18714. Train Steps: 5/90 Loss: 0.0008 torch.Size([8, 600, 800])
  18715. torch.Size([8, 8])
  18716. tensor([[0.6296, 0.4045, 0.9138, 0.4100, 0.4232, 0.4242, 0.7422, 0.5297],
  18717. [0.6126, 0.4039, 0.8237, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138],
  18718. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  18719. [0.6164, 0.4102, 0.8850, 0.4867, 0.4213, 0.5983, 0.5609, 0.5038],
  18720. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  18721. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  18722. [0.6144, 0.4032, 0.8563, 0.3283, 0.3525, 0.4200, 0.5775, 0.5583],
  18723. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131]],
  18724. device='cuda:0', dtype=torch.float64)
  18725. predictions are: tensor([[0.7257, 0.4661, 0.8940, 0.4321, 0.4246, 0.4201, 0.7156, 0.5389],
  18726. [0.6664, 0.4251, 0.8172, 0.4186, 0.3816, 0.3503, 0.6246, 0.6174],
  18727. [0.6524, 0.4117, 0.8400, 0.3153, 0.4190, 0.2505, 0.6523, 0.4660],
  18728. [0.6420, 0.4179, 0.8749, 0.5114, 0.4585, 0.5876, 0.6144, 0.5262],
  18729. [0.6775, 0.4307, 0.8955, 0.3628, 0.3980, 0.2863, 0.6482, 0.5436],
  18730. [0.6652, 0.4166, 0.8540, 0.5090, 0.4253, 0.5043, 0.6642, 0.5239],
  18731. [0.6302, 0.3965, 0.8610, 0.3333, 0.3732, 0.4108, 0.5898, 0.5453],
  18732. [0.6933, 0.4378, 0.8574, 0.4011, 0.3623, 0.3861, 0.6253, 0.5153]],
  18733. device='cuda:0', grad_fn=<AddmmBackward>)
  18734. landmarks are: tensor([[[0.6296, 0.4045, 0.9137, 0.4100, 0.4232, 0.4242, 0.7422, 0.5297],
  18735. [0.6126, 0.4038, 0.8238, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138],
  18736. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  18737. [0.6164, 0.4102, 0.8850, 0.4867, 0.4212, 0.5983, 0.5609, 0.5038],
  18738. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  18739. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  18740. [0.6144, 0.4032, 0.8562, 0.3283, 0.3525, 0.4200, 0.5775, 0.5583],
  18741. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131]]],
  18742. device='cuda:0')
  18743. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  18744. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  18745. loss_train: 0.004839719942538068
  18746. step: 6
  18747. running loss: 0.0008066199904230112
  18748. Train Steps: 6/90 Loss: 0.0008 torch.Size([8, 600, 800])
  18749. torch.Size([8, 8])
  18750. tensor([[0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5988, 0.5700],
  18751. [0.6132, 0.4066, 0.7259, 0.2402, 0.3588, 0.3300, 0.6000, 0.5600],
  18752. [0.6216, 0.4167, 0.8588, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667],
  18753. [0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  18754. [0.6127, 0.4066, 0.8550, 0.5567, 0.4662, 0.5141, 0.5070, 0.5412],
  18755. [0.6083, 0.3957, 0.8638, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  18756. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  18757. [0.6245, 0.4115, 0.8700, 0.4883, 0.4625, 0.5517, 0.6100, 0.5217]],
  18758. device='cuda:0', dtype=torch.float64)
  18759. predictions are: tensor([[0.5816, 0.3644, 0.7977, 0.2874, 0.3627, 0.3423, 0.6396, 0.5712],
  18760. [0.6387, 0.4050, 0.7193, 0.2647, 0.3624, 0.3305, 0.6420, 0.5596],
  18761. [0.6548, 0.4103, 0.8759, 0.5895, 0.4044, 0.4977, 0.6551, 0.5572],
  18762. [0.6460, 0.4036, 0.8800, 0.5074, 0.4045, 0.5317, 0.6356, 0.5149],
  18763. [0.6181, 0.4096, 0.8668, 0.5840, 0.4459, 0.4698, 0.5891, 0.5323],
  18764. [0.6339, 0.3948, 0.8712, 0.4996, 0.4313, 0.5057, 0.5855, 0.4824],
  18765. [0.6447, 0.4080, 0.8645, 0.4941, 0.4178, 0.4753, 0.5918, 0.5320],
  18766. [0.6328, 0.4041, 0.8715, 0.4879, 0.4285, 0.5262, 0.6690, 0.5281]],
  18767. device='cuda:0', grad_fn=<AddmmBackward>)
  18768. landmarks are: tensor([[[0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5987, 0.5700],
  18769. [0.6132, 0.4066, 0.7259, 0.2402, 0.3587, 0.3300, 0.6000, 0.5600],
  18770. [0.6216, 0.4167, 0.8587, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667],
  18771. [0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  18772. [0.6127, 0.4066, 0.8550, 0.5567, 0.4662, 0.5141, 0.5070, 0.5412],
  18773. [0.6083, 0.3957, 0.8637, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  18774. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  18775. [0.6245, 0.4115, 0.8700, 0.4883, 0.4625, 0.5517, 0.6100, 0.5217]]],
  18776. device='cuda:0')
  18777. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  18778. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  18779. loss_train: 0.005535328615223989
  18780. step: 7
  18781. running loss: 0.0007907612307462841
  18782. Train Steps: 7/90 Loss: 0.0008 torch.Size([8, 600, 800])
  18783. torch.Size([8, 8])
  18784. tensor([[ nan, nan, 0.8850, 0.2817, 0.5112, 0.2183, 0.7184, 0.5436],
  18785. [ nan, nan, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  18786. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  18787. [0.6122, 0.3993, 0.8738, 0.4667, 0.4517, 0.4879, 0.5155, 0.4927],
  18788. [0.6186, 0.4060, 0.8750, 0.5050, 0.3538, 0.4367, 0.5813, 0.6083],
  18789. [0.6361, 0.4076, 0.8862, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297],
  18790. [0.6289, 0.4019, 0.8113, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332],
  18791. [ nan, nan, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600]],
  18792. device='cuda:0', dtype=torch.float64)
  18793. predictions are: tensor([[0.2315, 0.1313, 0.8693, 0.2834, 0.5037, 0.2446, 0.6931, 0.5315],
  18794. [0.1525, 0.0825, 0.9006, 0.3641, 0.5002, 0.2595, 0.7082, 0.5658],
  18795. [0.6859, 0.4555, 0.8358, 0.3334, 0.3620, 0.3177, 0.5499, 0.5456],
  18796. [0.7301, 0.4657, 0.8544, 0.5141, 0.4441, 0.5523, 0.5109, 0.4873],
  18797. [0.7582, 0.4908, 0.8610, 0.5299, 0.3585, 0.4687, 0.5936, 0.6015],
  18798. [0.7582, 0.4922, 0.8566, 0.5872, 0.3775, 0.5032, 0.6848, 0.5286],
  18799. [0.7481, 0.4672, 0.7994, 0.5595, 0.3788, 0.5282, 0.7132, 0.5261],
  18800. [0.1357, 0.0802, 0.7567, 0.2895, 0.4016, 0.2786, 0.5403, 0.5677]],
  18801. device='cuda:0', grad_fn=<AddmmBackward>)
  18802. landmarks are: tensor([[[0.0000, 0.0000, 0.8850, 0.2817, 0.5113, 0.2183, 0.7184, 0.5436],
  18803. [0.0000, 0.0000, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  18804. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  18805. [0.6122, 0.3993, 0.8737, 0.4667, 0.4517, 0.4879, 0.5155, 0.4927],
  18806. [0.6186, 0.4060, 0.8750, 0.5050, 0.3537, 0.4367, 0.5813, 0.6083],
  18807. [0.6361, 0.4076, 0.8863, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297],
  18808. [0.6289, 0.4019, 0.8112, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332],
  18809. [0.0000, 0.0000, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600]]],
  18810. device='cuda:0')
  18811. loss_train_step before backward: tensor(0.0038, device='cuda:0', grad_fn=<MseLossBackward>)
  18812. loss_train_step after backward: tensor(0.0038, device='cuda:0', grad_fn=<MseLossBackward>)
  18813. loss_train: 0.009357340022688732
  18814. step: 8
  18815. running loss: 0.0011696675028360914
  18816.  
  18817. Train Steps: 8/90 Loss: 0.0012 torch.Size([8, 600, 800])
  18818. torch.Size([8, 8])
  18819. tensor([[0.6268, 0.4102, 0.8938, 0.3667, 0.4025, 0.2833, 0.6275, 0.5183],
  18820. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6063, 0.5617],
  18821. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  18822. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  18823. [0.6214, 0.4112, 0.7838, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  18824. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  18825. [0.6113, 0.4006, 0.8700, 0.5350, 0.3638, 0.3767, 0.5097, 0.4882],
  18826. [0.6148, 0.4053, 0.8750, 0.4550, 0.4850, 0.5218, 0.5863, 0.5567]],
  18827. device='cuda:0', dtype=torch.float64)
  18828. predictions are: tensor([[0.5513, 0.3499, 0.8754, 0.3885, 0.4092, 0.2980, 0.6123, 0.5509],
  18829. [0.5079, 0.3261, 0.8825, 0.5033, 0.3551, 0.4708, 0.5870, 0.5716],
  18830. [0.5486, 0.3592, 0.8733, 0.5777, 0.3856, 0.4378, 0.5609, 0.5814],
  18831. [0.5862, 0.3672, 0.8805, 0.3626, 0.4499, 0.3485, 0.7259, 0.5326],
  18832. [0.5220, 0.3297, 0.7695, 0.2367, 0.3591, 0.3185, 0.5644, 0.5184],
  18833. [0.5648, 0.3485, 0.8498, 0.4890, 0.3768, 0.5428, 0.5952, 0.5136],
  18834. [0.5259, 0.3511, 0.8384, 0.5438, 0.3694, 0.4067, 0.5304, 0.5357],
  18835. [0.5168, 0.3309, 0.8624, 0.4739, 0.4763, 0.5202, 0.5775, 0.5674]],
  18836. device='cuda:0', grad_fn=<AddmmBackward>)
  18837. landmarks are: tensor([[[0.6268, 0.4102, 0.8938, 0.3667, 0.4025, 0.2833, 0.6275, 0.5183],
  18838. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6062, 0.5617],
  18839. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  18840. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  18841. [0.6214, 0.4112, 0.7837, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  18842. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  18843. [0.6113, 0.4006, 0.8700, 0.5350, 0.3638, 0.3767, 0.5097, 0.4882],
  18844. [0.6148, 0.4053, 0.8750, 0.4550, 0.4850, 0.5218, 0.5863, 0.5567]]],
  18845. device='cuda:0')
  18846. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  18847. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  18848. loss_train: 0.01096438625245355
  18849. step: 9
  18850. running loss: 0.0012182651391615057
  18851. Train Steps: 9/90 Loss: 0.0012 torch.Size([8, 600, 800])
  18852. torch.Size([8, 8])
  18853. tensor([[ nan, nan, 0.6900, 0.1917, 0.3937, 0.2367, 0.5240, 0.5246],
  18854. [0.6207, 0.4081, 0.7662, 0.2067, 0.3962, 0.3200, 0.6312, 0.5300],
  18855. [0.6082, 0.4024, 0.8738, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  18856. [0.6124, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  18857. [0.6275, 0.4024, 0.8500, 0.5383, 0.3912, 0.4883, 0.6288, 0.5100],
  18858. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  18859. [0.6255, 0.4017, 0.8688, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  18860. [0.6085, 0.4008, 0.8588, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283]],
  18861. device='cuda:0', dtype=torch.float64)
  18862. predictions are: tensor([[-0.1154, -0.0725, 0.7042, 0.2203, 0.4047, 0.2516, 0.5119, 0.5439],
  18863. [ 0.6155, 0.3898, 0.7578, 0.2185, 0.3942, 0.3477, 0.6270, 0.5514],
  18864. [ 0.5186, 0.3507, 0.8849, 0.4107, 0.3595, 0.4197, 0.5105, 0.5245],
  18865. [ 0.5673, 0.3583, 0.8317, 0.5371, 0.3788, 0.4971, 0.5513, 0.5711],
  18866. [ 0.5857, 0.3667, 0.8425, 0.5602, 0.4127, 0.5175, 0.6205, 0.5132],
  18867. [ 0.5985, 0.3863, 0.7517, 0.3606, 0.5034, 0.2179, 0.5441, 0.6325],
  18868. [ 0.5683, 0.3477, 0.8672, 0.3361, 0.3666, 0.3888, 0.6350, 0.5105],
  18869. [ 0.5255, 0.3333, 0.8658, 0.5461, 0.4960, 0.5110, 0.5175, 0.5537]],
  18870. device='cuda:0', grad_fn=<AddmmBackward>)
  18871. landmarks are: tensor([[[0.0000, 0.0000, 0.6900, 0.1917, 0.3938, 0.2367, 0.5240, 0.5246],
  18872. [0.6207, 0.4081, 0.7663, 0.2067, 0.3963, 0.3200, 0.6313, 0.5300],
  18873. [0.6082, 0.4024, 0.8737, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  18874. [0.6123, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  18875. [0.6275, 0.4024, 0.8500, 0.5383, 0.3913, 0.4883, 0.6288, 0.5100],
  18876. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  18877. [0.6255, 0.4017, 0.8687, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  18878. [0.6084, 0.4008, 0.8587, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283]]],
  18879. device='cuda:0')
  18880. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  18881. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  18882. loss_train: 0.0121056787611451
  18883. step: 10
  18884. running loss: 0.00121056787611451
  18885. Train Steps: 10/90 Loss: 0.0012 torch.Size([8, 600, 800])
  18886. torch.Size([8, 8])
  18887. tensor([[0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  18888. [0.6226, 0.4098, 0.8912, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  18889. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  18890. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  18891. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  18892. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  18893. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317],
  18894. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791]],
  18895. device='cuda:0', dtype=torch.float64)
  18896. predictions are: tensor([[0.4749, 0.3249, 0.6890, 0.2968, 0.3606, 0.3122, 0.4926, 0.5845],
  18897. [0.5149, 0.3481, 0.9194, 0.4068, 0.4093, 0.2657, 0.5357, 0.5565],
  18898. [0.5039, 0.3380, 0.8925, 0.4838, 0.3632, 0.4430, 0.5818, 0.5308],
  18899. [0.5325, 0.3546, 0.7897, 0.2229, 0.4606, 0.2118, 0.5699, 0.5353],
  18900. [0.5349, 0.3446, 0.8689, 0.5037, 0.4380, 0.5839, 0.5417, 0.5008],
  18901. [0.5464, 0.3667, 0.8537, 0.5286, 0.3979, 0.4906, 0.6903, 0.5944],
  18902. [0.4846, 0.3226, 0.8902, 0.5299, 0.3904, 0.3915, 0.5313, 0.5475],
  18903. [0.4769, 0.3094, 0.8922, 0.4655, 0.4012, 0.5251, 0.5598, 0.4930]],
  18904. device='cuda:0', grad_fn=<AddmmBackward>)
  18905. landmarks are: tensor([[[0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  18906. [0.6226, 0.4098, 0.8913, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  18907. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  18908. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  18909. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  18910. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  18911. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317],
  18912. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791]]],
  18913. device='cuda:0')
  18914. loss_train_step before backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  18915. loss_train_step after backward: tensor(0.0028, device='cuda:0', grad_fn=<MseLossBackward>)
  18916. loss_train: 0.0149084952136036
  18917. step: 11
  18918. running loss: 0.0013553177466912364
  18919. Train Steps: 11/90 Loss: 0.0014 torch.Size([8, 600, 800])
  18920. torch.Size([8, 8])
  18921. tensor([[0.6152, 0.4131, 0.6863, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  18922. [0.6112, 0.4029, 0.8638, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  18923. [ nan, nan, 0.7648, 0.2722, 0.3962, 0.2183, 0.5060, 0.5422],
  18924. [0.6346, 0.4165, 0.9138, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471],
  18925. [0.6175, 0.4093, 0.8800, 0.4433, 0.4075, 0.4367, 0.5128, 0.4901],
  18926. [0.6293, 0.4024, 0.8750, 0.5000, 0.4012, 0.5733, 0.7121, 0.5633],
  18927. [0.6336, 0.4191, 0.8938, 0.5167, 0.3937, 0.3517, 0.7343, 0.5748],
  18928. [ nan, nan, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729]],
  18929. device='cuda:0', dtype=torch.float64)
  18930. predictions are: tensor([[0.5501, 0.3820, 0.7020, 0.2429, 0.3641, 0.3376, 0.5606, 0.5479],
  18931. [0.6051, 0.4123, 0.8787, 0.4856, 0.4827, 0.4924, 0.5378, 0.5501],
  18932. [0.0207, 0.0367, 0.7621, 0.2604, 0.3706, 0.2205, 0.4753, 0.5109],
  18933. [0.6254, 0.4297, 0.9316, 0.4102, 0.3995, 0.4381, 0.7016, 0.5453],
  18934. [0.5953, 0.4058, 0.8951, 0.4503, 0.3976, 0.4549, 0.4712, 0.5083],
  18935. [0.6086, 0.3952, 0.8804, 0.4871, 0.3942, 0.5839, 0.6646, 0.5367],
  18936. [0.6103, 0.4185, 0.8988, 0.5163, 0.3965, 0.3625, 0.7024, 0.5593],
  18937. [0.0516, 0.0381, 0.7138, 0.2252, 0.4206, 0.1837, 0.4850, 0.5502]],
  18938. device='cuda:0', grad_fn=<AddmmBackward>)
  18939. landmarks are: tensor([[[0.6152, 0.4131, 0.6862, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  18940. [0.6112, 0.4029, 0.8637, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  18941. [0.0000, 0.0000, 0.7648, 0.2722, 0.3963, 0.2183, 0.5060, 0.5422],
  18942. [0.6346, 0.4165, 0.9137, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471],
  18943. [0.6175, 0.4093, 0.8800, 0.4433, 0.4075, 0.4367, 0.5128, 0.4901],
  18944. [0.6293, 0.4024, 0.8750, 0.5000, 0.4013, 0.5733, 0.7121, 0.5633],
  18945. [0.6336, 0.4191, 0.8938, 0.5167, 0.3938, 0.3517, 0.7343, 0.5748],
  18946. [0.0000, 0.0000, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729]]],
  18947. device='cuda:0')
  18948. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  18949. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  18950. loss_train: 0.015397945273434743
  18951. step: 12
  18952. running loss: 0.001283162106119562
  18953.  
  18954. Train Steps: 12/90 Loss: 0.0013 torch.Size([8, 600, 800])
  18955. torch.Size([8, 8])
  18956. tensor([[0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  18957. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  18958. [0.6197, 0.3986, 0.8800, 0.4617, 0.4188, 0.4783, 0.5687, 0.5550],
  18959. [0.6226, 0.4103, 0.8575, 0.3450, 0.4388, 0.2067, 0.5787, 0.5383],
  18960. [ nan, nan, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  18961. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092],
  18962. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  18963. [0.6127, 0.4084, 0.8700, 0.4467, 0.3987, 0.4317, 0.5013, 0.5471]],
  18964. device='cuda:0', dtype=torch.float64)
  18965. predictions are: tensor([[0.2244, 0.1580, 0.7232, 0.2552, 0.4145, 0.2532, 0.5267, 0.5611],
  18966. [0.6021, 0.4020, 0.8655, 0.5411, 0.3562, 0.4406, 0.6404, 0.4828],
  18967. [0.5888, 0.3883, 0.8618, 0.4325, 0.3983, 0.4777, 0.5589, 0.5230],
  18968. [0.6022, 0.4384, 0.8692, 0.3581, 0.4352, 0.2404, 0.5637, 0.5412],
  18969. [0.0772, 0.0743, 0.8547, 0.2651, 0.5129, 0.2014, 0.6702, 0.5527],
  18970. [0.5748, 0.3815, 0.8642, 0.5016, 0.3821, 0.5142, 0.5752, 0.4859],
  18971. [0.6070, 0.4108, 0.8904, 0.4143, 0.3503, 0.3365, 0.5993, 0.5210],
  18972. [0.5782, 0.4060, 0.8511, 0.4246, 0.3702, 0.4451, 0.4878, 0.5240]],
  18973. device='cuda:0', grad_fn=<AddmmBackward>)
  18974. landmarks are: tensor([[[0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  18975. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  18976. [0.6197, 0.3986, 0.8800, 0.4617, 0.4187, 0.4783, 0.5688, 0.5550],
  18977. [0.6226, 0.4103, 0.8575, 0.3450, 0.4387, 0.2067, 0.5788, 0.5383],
  18978. [0.0000, 0.0000, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  18979. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092],
  18980. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  18981. [0.6127, 0.4084, 0.8700, 0.4467, 0.3988, 0.4317, 0.5013, 0.5471]]],
  18982. device='cuda:0')
  18983. loss_train_step before backward: tensor(0.0039, device='cuda:0', grad_fn=<MseLossBackward>)
  18984. loss_train_step after backward: tensor(0.0039, device='cuda:0', grad_fn=<MseLossBackward>)
  18985. loss_train: 0.019262570276623592
  18986. step: 13
  18987. running loss: 0.0014817361751248916
  18988. Train Steps: 13/90 Loss: 0.0015 torch.Size([8, 600, 800])
  18989. torch.Size([8, 8])
  18990. tensor([[0.6043, 0.4022, 0.6887, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136],
  18991. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  18992. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  18993. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  18994. [0.6100, 0.4016, 0.8600, 0.5067, 0.4612, 0.5233, 0.5086, 0.5519],
  18995. [0.6197, 0.4091, 0.8800, 0.4783, 0.3538, 0.4767, 0.5950, 0.5550],
  18996. [0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726],
  18997. [0.6210, 0.4164, 0.7202, 0.2930, 0.4025, 0.2483, 0.5687, 0.5567]],
  18998. device='cuda:0', dtype=torch.float64)
  18999. predictions are: tensor([[0.5710, 0.3831, 0.7021, 0.1966, 0.3722, 0.2206, 0.5852, 0.5107],
  19000. [0.5573, 0.3777, 0.8761, 0.3221, 0.3587, 0.2827, 0.6023, 0.4967],
  19001. [0.5786, 0.3829, 0.8698, 0.4437, 0.3681, 0.3612, 0.5885, 0.5433],
  19002. [0.5425, 0.3700, 0.9020, 0.4852, 0.3674, 0.4227, 0.6205, 0.5100],
  19003. [0.5838, 0.3936, 0.8620, 0.5194, 0.4552, 0.4883, 0.5314, 0.5250],
  19004. [0.5672, 0.3824, 0.8751, 0.4618, 0.3647, 0.4477, 0.6069, 0.5339],
  19005. [0.5511, 0.3646, 0.7593, 0.2733, 0.3573, 0.2556, 0.5577, 0.4689],
  19006. [0.4920, 0.3405, 0.7663, 0.2715, 0.4157, 0.2283, 0.5999, 0.5700]],
  19007. device='cuda:0', grad_fn=<AddmmBackward>)
  19008. landmarks are: tensor([[[0.6043, 0.4022, 0.6888, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136],
  19009. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  19010. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  19011. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  19012. [0.6100, 0.4016, 0.8600, 0.5067, 0.4613, 0.5233, 0.5086, 0.5519],
  19013. [0.6197, 0.4091, 0.8800, 0.4783, 0.3537, 0.4767, 0.5950, 0.5550],
  19014. [0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726],
  19015. [0.6210, 0.4164, 0.7202, 0.2930, 0.4025, 0.2483, 0.5688, 0.5567]]],
  19016. device='cuda:0')
  19017. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  19018. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  19019. loss_train: 0.020266804698621854
  19020. step: 14
  19021. running loss: 0.0014476289070444182
  19022. Train Steps: 14/90 Loss: 0.0014 torch.Size([8, 600, 800])
  19023. torch.Size([8, 8])
  19024. tensor([[0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  19025. [0.6250, 0.4131, 0.8688, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  19026. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  19027. [0.6213, 0.4001, 0.7712, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183],
  19028. [0.6149, 0.4054, 0.6713, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  19029. [0.6203, 0.4096, 0.8862, 0.4267, 0.3538, 0.4117, 0.6025, 0.5650],
  19030. [0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  19031. [0.6205, 0.4016, 0.8350, 0.2717, 0.3987, 0.2550, 0.5787, 0.5133]],
  19032. device='cuda:0', dtype=torch.float64)
  19033. predictions are: tensor([[0.5906, 0.4013, 0.8613, 0.4550, 0.4265, 0.5389, 0.6329, 0.5143],
  19034. [0.5986, 0.4023, 0.8560, 0.2970, 0.4190, 0.2048, 0.6304, 0.5300],
  19035. [0.5941, 0.4060, 0.7585, 0.2281, 0.4382, 0.1613, 0.6238, 0.5178],
  19036. [0.5804, 0.3693, 0.7585, 0.2230, 0.4211, 0.1526, 0.6025, 0.5110],
  19037. [0.5636, 0.3875, 0.6671, 0.2289, 0.3848, 0.1903, 0.5313, 0.5479],
  19038. [0.5481, 0.3683, 0.8799, 0.4000, 0.3423, 0.3808, 0.6015, 0.5648],
  19039. [0.5518, 0.3761, 0.8546, 0.4256, 0.3682, 0.4922, 0.5651, 0.4971],
  19040. [0.5714, 0.3834, 0.8223, 0.2560, 0.3741, 0.2333, 0.5960, 0.5108]],
  19041. device='cuda:0', grad_fn=<AddmmBackward>)
  19042. landmarks are: tensor([[[0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  19043. [0.6250, 0.4131, 0.8687, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  19044. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  19045. [0.6213, 0.4001, 0.7713, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183],
  19046. [0.6149, 0.4054, 0.6712, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  19047. [0.6203, 0.4096, 0.8863, 0.4267, 0.3537, 0.4117, 0.6025, 0.5650],
  19048. [0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  19049. [0.6205, 0.4015, 0.8350, 0.2717, 0.3988, 0.2550, 0.5788, 0.5133]]],
  19050. device='cuda:0')
  19051. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  19052. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  19053. loss_train: 0.020893012144370005
  19054. step: 15
  19055. running loss: 0.0013928674762913337
  19056. Train Steps: 15/90 Loss: 0.0014 torch.Size([8, 600, 800])
  19057. torch.Size([8, 8])
  19058. tensor([[0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  19059. [0.6168, 0.4029, 0.8523, 0.3417, 0.3588, 0.5000, 0.6125, 0.5400],
  19060. [0.6353, 0.4128, 0.8488, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  19061. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  19062. [0.6272, 0.4045, 0.8538, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  19063. [0.6218, 0.4098, 0.7238, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  19064. [0.6307, 0.4029, 0.8988, 0.4817, 0.3937, 0.3500, 0.7311, 0.5378],
  19065. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355]],
  19066. device='cuda:0', dtype=torch.float64)
  19067. predictions are: tensor([[0.6354, 0.4278, 0.7517, 0.2153, 0.4245, 0.1671, 0.5917, 0.5210],
  19068. [0.5721, 0.3818, 0.8357, 0.3290, 0.3228, 0.4648, 0.6028, 0.5294],
  19069. [0.6360, 0.4356, 0.8251, 0.2208, 0.5036, 0.1596, 0.6466, 0.5631],
  19070. [0.5783, 0.3829, 0.8533, 0.4649, 0.3959, 0.4808, 0.5663, 0.5193],
  19071. [0.6522, 0.4228, 0.8267, 0.5780, 0.3619, 0.4118, 0.5654, 0.4704],
  19072. [0.6159, 0.4050, 0.7258, 0.1741, 0.4098, 0.2255, 0.6184, 0.5360],
  19073. [0.6335, 0.4194, 0.8854, 0.4637, 0.3663, 0.3228, 0.6904, 0.5412],
  19074. [0.5684, 0.3690, 0.8838, 0.3206, 0.4286, 0.3061, 0.7137, 0.5155]],
  19075. device='cuda:0', grad_fn=<AddmmBackward>)
  19076. landmarks are: tensor([[[0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  19077. [0.6168, 0.4029, 0.8523, 0.3417, 0.3587, 0.5000, 0.6125, 0.5400],
  19078. [0.6353, 0.4128, 0.8487, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  19079. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  19080. [0.6271, 0.4045, 0.8537, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  19081. [0.6218, 0.4098, 0.7237, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  19082. [0.6307, 0.4029, 0.8988, 0.4817, 0.3938, 0.3500, 0.7311, 0.5378],
  19083. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355]]],
  19084. device='cuda:0')
  19085. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  19086. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  19087. loss_train: 0.02157996202004142
  19088. step: 16
  19089. running loss: 0.0013487476262525888
  19090.  
  19091. Train Steps: 16/90 Loss: 0.0013 torch.Size([8, 600, 800])
  19092. torch.Size([8, 8])
  19093. tensor([[0.6151, 0.4125, 0.8738, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  19094. [0.6277, 0.4083, 0.8350, 0.2717, 0.4562, 0.1800, 0.5918, 0.4878],
  19095. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  19096. [0.6186, 0.4060, 0.8750, 0.5050, 0.3538, 0.4367, 0.5813, 0.6083],
  19097. [0.6087, 0.3976, 0.8337, 0.3867, 0.3713, 0.3117, 0.5938, 0.5300],
  19098. [ nan, nan, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  19099. [0.6198, 0.4101, 0.8838, 0.5283, 0.3763, 0.5267, 0.5913, 0.5567],
  19100. [0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148]],
  19101. device='cuda:0', dtype=torch.float64)
  19102. predictions are: tensor([[0.7122, 0.4747, 0.8445, 0.4124, 0.3622, 0.3767, 0.5372, 0.5509],
  19103. [0.7166, 0.4586, 0.8083, 0.2608, 0.4620, 0.2070, 0.5982, 0.4951],
  19104. [0.5601, 0.3597, 0.7098, 0.2516, 0.4317, 0.2329, 0.5504, 0.5738],
  19105. [0.6958, 0.4550, 0.8660, 0.4839, 0.3600, 0.4225, 0.6120, 0.5877],
  19106. [0.6833, 0.4521, 0.8362, 0.3760, 0.3806, 0.3086, 0.6095, 0.5258],
  19107. [0.0834, 0.0515, 0.8465, 0.2422, 0.5094, 0.2428, 0.7514, 0.5423],
  19108. [0.7125, 0.4723, 0.8479, 0.5150, 0.3856, 0.5142, 0.6055, 0.5365],
  19109. [0.7223, 0.4807, 0.8598, 0.5261, 0.3942, 0.4611, 0.6115, 0.5069]],
  19110. device='cuda:0', grad_fn=<AddmmBackward>)
  19111. landmarks are: tensor([[[0.6151, 0.4125, 0.8737, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  19112. [0.6277, 0.4083, 0.8350, 0.2717, 0.4563, 0.1800, 0.5918, 0.4878],
  19113. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  19114. [0.6186, 0.4060, 0.8750, 0.5050, 0.3537, 0.4367, 0.5813, 0.6083],
  19115. [0.6087, 0.3976, 0.8338, 0.3867, 0.3713, 0.3117, 0.5938, 0.5300],
  19116. [0.0000, 0.0000, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  19117. [0.6198, 0.4101, 0.8838, 0.5283, 0.3762, 0.5267, 0.5913, 0.5567],
  19118. [0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148]]],
  19119. device='cuda:0')
  19120. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  19121. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  19122. loss_train: 0.0230987589166034
  19123. step: 17
  19124. running loss: 0.0013587505245060824
  19125. Train Steps: 17/90 Loss: 0.0014 torch.Size([8, 600, 800])
  19126. torch.Size([8, 8])
  19127. tensor([[0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  19128. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5513, 0.5750],
  19129. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  19130. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  19131. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  19132. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5837, 0.5583],
  19133. [ nan, nan, 0.8850, 0.3000, 0.5363, 0.2250, 0.7343, 0.5771],
  19134. [0.6197, 0.4091, 0.8800, 0.4783, 0.3538, 0.4767, 0.5950, 0.5550]],
  19135. device='cuda:0', dtype=torch.float64)
  19136. predictions are: tensor([[0.6948, 0.4342, 0.8274, 0.3955, 0.3516, 0.4189, 0.5985, 0.5181],
  19137. [0.6616, 0.4353, 0.6757, 0.2311, 0.4291, 0.1741, 0.5695, 0.5668],
  19138. [0.7638, 0.4901, 0.8623, 0.5660, 0.4073, 0.4236, 0.5623, 0.5521],
  19139. [0.7171, 0.4646, 0.8997, 0.4317, 0.4204, 0.2874, 0.6812, 0.5608],
  19140. [0.7199, 0.4767, 0.8423, 0.4943, 0.4293, 0.5077, 0.5502, 0.5317],
  19141. [0.7093, 0.4723, 0.8679, 0.4347, 0.4737, 0.5264, 0.5791, 0.5443],
  19142. [0.1111, 0.0722, 0.8653, 0.2842, 0.5132, 0.2146, 0.7645, 0.5478],
  19143. [0.7002, 0.4582, 0.8673, 0.4589, 0.3723, 0.4649, 0.6140, 0.5413]],
  19144. device='cuda:0', grad_fn=<AddmmBackward>)
  19145. landmarks are: tensor([[[0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  19146. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5512, 0.5750],
  19147. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  19148. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  19149. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  19150. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5838, 0.5583],
  19151. [0.0000, 0.0000, 0.8850, 0.3000, 0.5362, 0.2250, 0.7343, 0.5771],
  19152. [0.6197, 0.4091, 0.8800, 0.4783, 0.3537, 0.4767, 0.5950, 0.5550]]],
  19153. device='cuda:0')
  19154. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  19155. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  19156. loss_train: 0.02501470936113037
  19157. step: 18
  19158. running loss: 0.001389706075618354
  19159. Train Steps: 18/90 Loss: 0.0014 torch.Size([8, 600, 800])
  19160. torch.Size([8, 8])
  19161. tensor([[0.6102, 0.4001, 0.7738, 0.3583, 0.3463, 0.3800, 0.5524, 0.5689],
  19162. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  19163. [0.6310, 0.4017, 0.8563, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006],
  19164. [0.6083, 0.3957, 0.8638, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  19165. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  19166. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  19167. [0.6171, 0.4127, 0.8900, 0.4800, 0.4325, 0.5783, 0.5769, 0.5090],
  19168. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284]],
  19169. device='cuda:0', dtype=torch.float64)
  19170. predictions are: tensor([[0.6704, 0.4291, 0.7901, 0.3458, 0.3656, 0.3822, 0.5583, 0.5874],
  19171. [0.6621, 0.4222, 0.8653, 0.3725, 0.3861, 0.3233, 0.6159, 0.5628],
  19172. [0.7010, 0.4444, 0.8489, 0.5905, 0.3913, 0.4867, 0.6440, 0.5469],
  19173. [0.6631, 0.4274, 0.8679, 0.4871, 0.4668, 0.5191, 0.5627, 0.5301],
  19174. [0.6619, 0.4045, 0.9055, 0.4651, 0.3786, 0.3689, 0.6459, 0.5426],
  19175. [0.7109, 0.4619, 0.8841, 0.5022, 0.3762, 0.4304, 0.6520, 0.5508],
  19176. [0.6863, 0.4545, 0.8942, 0.4778, 0.4615, 0.5887, 0.5857, 0.5413],
  19177. [0.7541, 0.4829, 0.7754, 0.2297, 0.4743, 0.1854, 0.6195, 0.5590]],
  19178. device='cuda:0', grad_fn=<AddmmBackward>)
  19179. landmarks are: tensor([[[0.6102, 0.4001, 0.7738, 0.3583, 0.3462, 0.3800, 0.5524, 0.5689],
  19180. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  19181. [0.6310, 0.4017, 0.8562, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006],
  19182. [0.6083, 0.3957, 0.8637, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  19183. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  19184. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  19185. [0.6171, 0.4127, 0.8900, 0.4800, 0.4325, 0.5783, 0.5769, 0.5090],
  19186. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284]]],
  19187. device='cuda:0')
  19188. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  19189. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  19190. loss_train: 0.026208211231278256
  19191. step: 19
  19192. running loss: 0.0013793795384883293
  19193. Train Steps: 19/90 Loss: 0.0014 torch.Size([8, 600, 800])
  19194. torch.Size([8, 8])
  19195. tensor([[0.6275, 0.4003, 0.9100, 0.3783, 0.4388, 0.3133, 0.7058, 0.5343],
  19196. [0.6122, 0.4006, 0.8850, 0.4217, 0.4088, 0.5517, 0.6063, 0.5517],
  19197. [0.6364, 0.4165, 0.9088, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  19198. [0.6299, 0.4303, 0.7963, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  19199. [0.6256, 0.4199, 0.8638, 0.5800, 0.3987, 0.4383, 0.5600, 0.5950],
  19200. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  19201. [0.6147, 0.4107, 0.8137, 0.3333, 0.3750, 0.2683, 0.5006, 0.5412],
  19202. [ nan, nan, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552]],
  19203. device='cuda:0', dtype=torch.float64)
  19204. predictions are: tensor([[0.6235, 0.3860, 0.8834, 0.3867, 0.4348, 0.3253, 0.7166, 0.5325],
  19205. [0.6541, 0.4193, 0.8757, 0.4302, 0.4235, 0.5741, 0.5998, 0.5390],
  19206. [0.6558, 0.4154, 0.8989, 0.4283, 0.4241, 0.3357, 0.6506, 0.5496],
  19207. [0.7330, 0.4808, 0.7978, 0.3939, 0.4796, 0.2567, 0.5489, 0.6241],
  19208. [0.7382, 0.4679, 0.8532, 0.5744, 0.3992, 0.4674, 0.5695, 0.5939],
  19209. [0.6836, 0.4243, 0.8089, 0.2868, 0.4234, 0.2646, 0.6214, 0.5413],
  19210. [0.6716, 0.4293, 0.7856, 0.3307, 0.3658, 0.2910, 0.4913, 0.5589],
  19211. [0.0623, 0.0216, 0.8508, 0.2625, 0.5168, 0.2670, 0.7309, 0.5608]],
  19212. device='cuda:0', grad_fn=<AddmmBackward>)
  19213. landmarks are: tensor([[[0.6275, 0.4003, 0.9100, 0.3783, 0.4387, 0.3133, 0.7058, 0.5343],
  19214. [0.6122, 0.4006, 0.8850, 0.4217, 0.4087, 0.5517, 0.6062, 0.5517],
  19215. [0.6364, 0.4165, 0.9087, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  19216. [0.6299, 0.4303, 0.7962, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  19217. [0.6256, 0.4199, 0.8637, 0.5800, 0.3988, 0.4383, 0.5600, 0.5950],
  19218. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  19219. [0.6147, 0.4107, 0.8138, 0.3333, 0.3750, 0.2683, 0.5006, 0.5412],
  19220. [0.0000, 0.0000, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552]]],
  19221. device='cuda:0')
  19222. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  19223. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  19224. loss_train: 0.027047252719057724
  19225. step: 20
  19226. running loss: 0.0013523626359528862
  19227.  
  19228. Train Steps: 20/90 Loss: 0.0014 torch.Size([8, 600, 800])
  19229. torch.Size([8, 8])
  19230. tensor([[ nan, nan, 0.8525, 0.2217, 0.5413, 0.2367, 0.7367, 0.5482],
  19231. [0.6160, 0.4093, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  19232. [0.6202, 0.4066, 0.8398, 0.2648, 0.3925, 0.2627, 0.5845, 0.5124],
  19233. [0.6148, 0.3996, 0.8488, 0.3867, 0.3488, 0.4067, 0.5863, 0.5000],
  19234. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  19235. [ nan, nan, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  19236. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  19237. [0.6275, 0.4048, 0.8488, 0.2883, 0.4463, 0.2033, 0.6321, 0.5155]],
  19238. device='cuda:0', dtype=torch.float64)
  19239. predictions are: tensor([[0.1991, 0.1185, 0.8436, 0.2488, 0.5410, 0.2496, 0.7251, 0.5584],
  19240. [0.6871, 0.4371, 0.8549, 0.5069, 0.3750, 0.4871, 0.5453, 0.5909],
  19241. [0.7278, 0.4729, 0.8267, 0.3110, 0.4081, 0.2709, 0.5883, 0.5380],
  19242. [0.6961, 0.4302, 0.8692, 0.4282, 0.3687, 0.4330, 0.5951, 0.5323],
  19243. [0.6967, 0.4432, 0.7616, 0.2722, 0.3858, 0.3439, 0.6094, 0.5486],
  19244. [0.0834, 0.0346, 0.8687, 0.2925, 0.5354, 0.2652, 0.7239, 0.5716],
  19245. [0.7481, 0.4821, 0.8982, 0.5283, 0.3879, 0.4561, 0.6644, 0.5748],
  19246. [0.7045, 0.4447, 0.8591, 0.3253, 0.4729, 0.2292, 0.6392, 0.5443]],
  19247. device='cuda:0', grad_fn=<AddmmBackward>)
  19248. landmarks are: tensor([[[0.0000, 0.0000, 0.8525, 0.2217, 0.5412, 0.2367, 0.7367, 0.5482],
  19249. [0.6160, 0.4092, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  19250. [0.6202, 0.4066, 0.8398, 0.2648, 0.3925, 0.2627, 0.5845, 0.5124],
  19251. [0.6148, 0.3996, 0.8487, 0.3867, 0.3487, 0.4067, 0.5863, 0.5000],
  19252. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  19253. [0.0000, 0.0000, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  19254. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  19255. [0.6275, 0.4048, 0.8487, 0.2883, 0.4462, 0.2033, 0.6321, 0.5155]]],
  19256. device='cuda:0')
  19257. loss_train_step before backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  19258. loss_train_step after backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  19259. loss_train: 0.029495812923414633
  19260. step: 21
  19261. running loss: 0.0014045625201626016
  19262. Train Steps: 21/90 Loss: 0.0014 torch.Size([8, 600, 800])
  19263. torch.Size([8, 8])
  19264. tensor([[0.6165, 0.4106, 0.7575, 0.1733, 0.3838, 0.2650, 0.5680, 0.5116],
  19265. [ nan, nan, 0.8488, 0.2300, 0.5563, 0.2100, 0.7390, 0.5679],
  19266. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  19267. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  19268. [0.6204, 0.4110, 0.7913, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367],
  19269. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  19270. [0.6133, 0.4094, 0.8495, 0.4028, 0.3588, 0.3200, 0.5003, 0.5407],
  19271. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5837, 0.5583]],
  19272. device='cuda:0', dtype=torch.float64)
  19273. predictions are: tensor([[0.6522, 0.4116, 0.7930, 0.2094, 0.4086, 0.2687, 0.5724, 0.5104],
  19274. [0.1545, 0.0771, 0.8851, 0.2805, 0.5526, 0.2638, 0.7352, 0.5587],
  19275. [0.5950, 0.3730, 0.9092, 0.4708, 0.4053, 0.4660, 0.5343, 0.5407],
  19276. [0.6627, 0.4341, 0.7680, 0.3104, 0.4542, 0.2209, 0.5618, 0.5739],
  19277. [0.6252, 0.3891, 0.8365, 0.2923, 0.4193, 0.2608, 0.6167, 0.5603],
  19278. [0.5939, 0.3655, 0.8846, 0.5923, 0.4072, 0.5317, 0.6605, 0.5606],
  19279. [0.6327, 0.4021, 0.8764, 0.4260, 0.3743, 0.3276, 0.5130, 0.5581],
  19280. [0.5707, 0.3681, 0.9099, 0.4746, 0.4918, 0.5689, 0.5726, 0.5683]],
  19281. device='cuda:0', grad_fn=<AddmmBackward>)
  19282. landmarks are: tensor([[[0.6165, 0.4106, 0.7575, 0.1733, 0.3837, 0.2650, 0.5680, 0.5116],
  19283. [0.0000, 0.0000, 0.8487, 0.2300, 0.5562, 0.2100, 0.7390, 0.5679],
  19284. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  19285. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  19286. [0.6204, 0.4110, 0.7912, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367],
  19287. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  19288. [0.6133, 0.4094, 0.8495, 0.4028, 0.3587, 0.3200, 0.5003, 0.5407],
  19289. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5838, 0.5583]]],
  19290. device='cuda:0')
  19291. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  19292. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  19293. loss_train: 0.030567696288926527
  19294. step: 22
  19295. running loss: 0.0013894407404057513
  19296. Train Steps: 22/90 Loss: 0.0014 torch.Size([8, 600, 800])
  19297. torch.Size([8, 8])
  19298. tensor([[0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  19299. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  19300. [0.6336, 0.4086, 0.8900, 0.3950, 0.3900, 0.2950, 0.6504, 0.5066],
  19301. [ nan, nan, 0.9088, 0.3783, 0.4562, 0.2617, 0.6741, 0.5575],
  19302. [ nan, nan, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  19303. [ nan, nan, 0.7725, 0.2611, 0.3675, 0.2733, 0.5413, 0.5167],
  19304. [0.6259, 0.4156, 0.8812, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960],
  19305. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167]],
  19306. device='cuda:0', dtype=torch.float64)
  19307. predictions are: tensor([[0.6392, 0.4052, 0.8825, 0.5237, 0.4866, 0.5350, 0.5361, 0.5516],
  19308. [0.7080, 0.4577, 0.7620, 0.3528, 0.5123, 0.1878, 0.5503, 0.6081],
  19309. [0.6843, 0.4237, 0.9446, 0.3873, 0.4054, 0.3337, 0.6932, 0.5064],
  19310. [0.1743, 0.1102, 0.9388, 0.3837, 0.4724, 0.2862, 0.6806, 0.5682],
  19311. [0.1335, 0.0885, 0.7111, 0.2123, 0.4566, 0.2587, 0.5505, 0.5482],
  19312. [0.1014, 0.0485, 0.7536, 0.2699, 0.3843, 0.2945, 0.5392, 0.5497],
  19313. [0.5646, 0.3630, 0.9077, 0.3126, 0.4761, 0.2239, 0.6187, 0.5062],
  19314. [0.6201, 0.4074, 0.8433, 0.3348, 0.3664, 0.3087, 0.5544, 0.5099]],
  19315. device='cuda:0', grad_fn=<AddmmBackward>)
  19316. landmarks are: tensor([[[0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  19317. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  19318. [0.6336, 0.4086, 0.8900, 0.3950, 0.3900, 0.2950, 0.6504, 0.5066],
  19319. [0.0000, 0.0000, 0.9087, 0.3783, 0.4563, 0.2617, 0.6741, 0.5575],
  19320. [0.0000, 0.0000, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  19321. [0.0000, 0.0000, 0.7725, 0.2611, 0.3675, 0.2733, 0.5412, 0.5167],
  19322. [0.6259, 0.4156, 0.8813, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960],
  19323. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167]]],
  19324. device='cuda:0')
  19325. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  19326. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  19327. loss_train: 0.03241958349826746
  19328. step: 23
  19329. running loss: 0.0014095471086203243
  19330. Train Steps: 23/90 Loss: 0.0014 torch.Size([8, 600, 800])
  19331. torch.Size([8, 8])
  19332. tensor([[0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332],
  19333. [0.6092, 0.4001, 0.8638, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  19334. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  19335. [0.6109, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  19336. [0.6171, 0.4127, 0.8900, 0.4800, 0.4325, 0.5783, 0.5769, 0.5090],
  19337. [0.6153, 0.4119, 0.8463, 0.3833, 0.3600, 0.3200, 0.5106, 0.5563],
  19338. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  19339. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600]],
  19340. device='cuda:0', dtype=torch.float64)
  19341. predictions are: tensor([[0.5196, 0.3046, 0.9327, 0.4675, 0.3851, 0.3874, 0.6833, 0.5238],
  19342. [0.5047, 0.3282, 0.8866, 0.4704, 0.4349, 0.5383, 0.5316, 0.5102],
  19343. [0.4951, 0.3208, 0.9140, 0.3828, 0.4615, 0.2056, 0.6317, 0.5085],
  19344. [0.5170, 0.3447, 0.8880, 0.4490, 0.3800, 0.3987, 0.5487, 0.5116],
  19345. [0.4440, 0.2850, 0.9048, 0.4702, 0.4556, 0.5799, 0.5876, 0.5062],
  19346. [0.4837, 0.3263, 0.8488, 0.3854, 0.3605, 0.3197, 0.5185, 0.5563],
  19347. [0.4881, 0.3055, 0.8886, 0.5065, 0.4360, 0.5546, 0.6133, 0.5081],
  19348. [0.5354, 0.3664, 0.8931, 0.5447, 0.4107, 0.4721, 0.5752, 0.5636]],
  19349. device='cuda:0', grad_fn=<AddmmBackward>)
  19350. landmarks are: tensor([[[0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332],
  19351. [0.6092, 0.4001, 0.8637, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  19352. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  19353. [0.6108, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  19354. [0.6171, 0.4127, 0.8900, 0.4800, 0.4325, 0.5783, 0.5769, 0.5090],
  19355. [0.6153, 0.4119, 0.8462, 0.3833, 0.3600, 0.3200, 0.5106, 0.5563],
  19356. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  19357. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600]]],
  19358. device='cuda:0')
  19359. loss_train_step before backward: tensor(0.0030, device='cuda:0', grad_fn=<MseLossBackward>)
  19360. loss_train_step after backward: tensor(0.0030, device='cuda:0', grad_fn=<MseLossBackward>)
  19361. loss_train: 0.035387905983952805
  19362. step: 24
  19363. running loss: 0.0014744960826647002
  19364.  
  19365. Train Steps: 24/90 Loss: 0.0015 torch.Size([8, 600, 800])
  19366. torch.Size([8, 8])
  19367. tensor([[0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  19368. [0.6250, 0.4131, 0.8688, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  19369. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  19370. [0.6270, 0.4267, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  19371. [ nan, nan, 0.7097, 0.2346, 0.4250, 0.1850, 0.5175, 0.5583],
  19372. [0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050],
  19373. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  19374. [0.6064, 0.4019, 0.8650, 0.4517, 0.4037, 0.5367, 0.5703, 0.5609]],
  19375. device='cuda:0', dtype=torch.float64)
  19376. predictions are: tensor([[ 0.5617, 0.3751, 0.8700, 0.3838, 0.3548, 0.3857, 0.5578, 0.5258],
  19377. [ 0.5048, 0.3418, 0.8775, 0.3221, 0.4436, 0.2397, 0.6185, 0.4976],
  19378. [ 0.5276, 0.3512, 0.9314, 0.4296, 0.4197, 0.3713, 0.7200, 0.5474],
  19379. [ 0.5901, 0.4103, 0.7223, 0.2898, 0.4692, 0.1786, 0.5616, 0.5901],
  19380. [-0.0880, -0.0387, 0.7203, 0.2246, 0.4522, 0.1743, 0.5261, 0.5415],
  19381. [ 0.4960, 0.3342, 0.8938, 0.4103, 0.3636, 0.5101, 0.6043, 0.5562],
  19382. [ 0.4895, 0.3157, 0.9209, 0.5233, 0.3707, 0.3770, 0.6103, 0.4452],
  19383. [ 0.4501, 0.3161, 0.8708, 0.4482, 0.4058, 0.5794, 0.5658, 0.5021]],
  19384. device='cuda:0', grad_fn=<AddmmBackward>)
  19385. landmarks are: tensor([[[0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  19386. [0.6250, 0.4131, 0.8687, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  19387. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  19388. [0.6270, 0.4266, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  19389. [0.0000, 0.0000, 0.7097, 0.2346, 0.4250, 0.1850, 0.5175, 0.5583],
  19390. [0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050],
  19391. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  19392. [0.6064, 0.4019, 0.8650, 0.4517, 0.4038, 0.5367, 0.5703, 0.5609]]],
  19393. device='cuda:0')
  19394. loss_train_step before backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  19395. loss_train_step after backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  19396. loss_train: 0.03780612416449003
  19397. step: 25
  19398. running loss: 0.0015122449665796011
  19399. Train Steps: 25/90 Loss: 0.0015 torch.Size([8, 600, 800])
  19400. torch.Size([8, 8])
  19401. tensor([[0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6188, 0.5283],
  19402. [ nan, nan, 0.7192, 0.2346, 0.4037, 0.2050, 0.5138, 0.5650],
  19403. [0.6219, 0.4097, 0.8738, 0.3400, 0.3563, 0.4117, 0.5975, 0.5683],
  19404. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  19405. [0.6204, 0.4007, 0.7838, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  19406. [0.6226, 0.4103, 0.8575, 0.3450, 0.4388, 0.2067, 0.5787, 0.5383],
  19407. [0.6222, 0.3957, 0.8838, 0.5017, 0.3937, 0.4600, 0.5900, 0.5017],
  19408. [0.6346, 0.4165, 0.9138, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471]],
  19409. device='cuda:0', dtype=torch.float64)
  19410. predictions are: tensor([[ 0.4933, 0.3324, 0.9031, 0.3576, 0.3816, 0.2557, 0.6110, 0.5010],
  19411. [-0.1040, -0.0676, 0.7089, 0.2206, 0.4172, 0.1880, 0.5342, 0.5466],
  19412. [ 0.5285, 0.3772, 0.8592, 0.3481, 0.3227, 0.3968, 0.5869, 0.5273],
  19413. [ 0.5383, 0.3741, 0.7581, 0.3609, 0.3283, 0.4108, 0.5078, 0.5228],
  19414. [ 0.5499, 0.3637, 0.7490, 0.2379, 0.4342, 0.1393, 0.5775, 0.4974],
  19415. [ 0.4884, 0.3450, 0.8489, 0.3635, 0.4307, 0.2122, 0.5531, 0.5142],
  19416. [ 0.4173, 0.2685, 0.8786, 0.4987, 0.3810, 0.4578, 0.5729, 0.4654],
  19417. [ 0.4876, 0.3254, 0.9143, 0.3999, 0.3792, 0.4312, 0.7038, 0.5270]],
  19418. device='cuda:0', grad_fn=<AddmmBackward>)
  19419. landmarks are: tensor([[[0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6187, 0.5283],
  19420. [0.0000, 0.0000, 0.7192, 0.2346, 0.4038, 0.2050, 0.5138, 0.5650],
  19421. [0.6219, 0.4097, 0.8737, 0.3400, 0.3562, 0.4117, 0.5975, 0.5683],
  19422. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  19423. [0.6204, 0.4007, 0.7837, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  19424. [0.6226, 0.4103, 0.8575, 0.3450, 0.4387, 0.2067, 0.5788, 0.5383],
  19425. [0.6222, 0.3957, 0.8838, 0.5017, 0.3938, 0.4600, 0.5900, 0.5017],
  19426. [0.6346, 0.4165, 0.9137, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471]]],
  19427. device='cuda:0')
  19428. loss_train_step before backward: tensor(0.0029, device='cuda:0', grad_fn=<MseLossBackward>)
  19429. loss_train_step after backward: tensor(0.0029, device='cuda:0', grad_fn=<MseLossBackward>)
  19430. loss_train: 0.040753726119874045
  19431. step: 26
  19432. running loss: 0.0015674510046105401
  19433. Train Steps: 26/90 Loss: 0.0016 torch.Size([8, 600, 800])
  19434. torch.Size([8, 8])
  19435. tensor([[0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  19436. [0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  19437. [0.6151, 0.4125, 0.8738, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  19438. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  19439. [0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  19440. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  19441. [0.6277, 0.4118, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  19442. [0.6095, 0.3970, 0.8688, 0.4767, 0.4860, 0.4879, 0.5191, 0.4940]],
  19443. device='cuda:0', dtype=torch.float64)
  19444. predictions are: tensor([[0.4985, 0.3262, 0.8613, 0.4730, 0.4176, 0.4879, 0.6350, 0.5325],
  19445. [0.5425, 0.3775, 0.8711, 0.4228, 0.3583, 0.5446, 0.5575, 0.5107],
  19446. [0.5347, 0.3735, 0.8484, 0.4161, 0.3385, 0.3609, 0.5067, 0.5492],
  19447. [0.5321, 0.3510, 0.8759, 0.4189, 0.3166, 0.3310, 0.6180, 0.4916],
  19448. [0.5305, 0.3616, 0.8636, 0.4562, 0.3478, 0.3919, 0.5319, 0.5672],
  19449. [0.5022, 0.3362, 0.8598, 0.4538, 0.3929, 0.4576, 0.5421, 0.5218],
  19450. [0.5116, 0.3473, 0.8895, 0.3635, 0.3684, 0.2352, 0.6363, 0.5174],
  19451. [0.4949, 0.3282, 0.8744, 0.4593, 0.4581, 0.4634, 0.5353, 0.5341]],
  19452. device='cuda:0', grad_fn=<AddmmBackward>)
  19453. landmarks are: tensor([[[0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  19454. [0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  19455. [0.6151, 0.4125, 0.8737, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  19456. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  19457. [0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  19458. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  19459. [0.6277, 0.4117, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  19460. [0.6095, 0.3970, 0.8687, 0.4767, 0.4860, 0.4879, 0.5191, 0.4940]]],
  19461. device='cuda:0')
  19462. loss_train_step before backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  19463. loss_train_step after backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  19464. loss_train: 0.042814259621081874
  19465. step: 27
  19466. running loss: 0.0015857133192993286
  19467. Train Steps: 27/90 Loss: 0.0016 torch.Size([8, 600, 800])
  19468. torch.Size([8, 8])
  19469. tensor([[0.6185, 0.4129, 0.8900, 0.4567, 0.3937, 0.5417, 0.5734, 0.5110],
  19470. [0.6250, 0.4013, 0.8525, 0.5417, 0.4037, 0.5117, 0.6325, 0.5017],
  19471. [0.6254, 0.4076, 0.8700, 0.3267, 0.4150, 0.3083, 0.7050, 0.5609],
  19472. [ nan, nan, 0.7335, 0.2569, 0.3788, 0.2667, 0.5066, 0.5578],
  19473. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  19474. [0.6173, 0.4114, 0.7325, 0.2500, 0.4213, 0.1917, 0.5338, 0.5700],
  19475. [0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  19476. [0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950]],
  19477. device='cuda:0', dtype=torch.float64)
  19478. predictions are: tensor([[ 0.5983, 0.4167, 0.8870, 0.4386, 0.3617, 0.5374, 0.5467, 0.5090],
  19479. [ 0.6104, 0.4011, 0.8272, 0.5336, 0.3790, 0.4975, 0.5823, 0.4976],
  19480. [ 0.5249, 0.3588, 0.8783, 0.3062, 0.3926, 0.2847, 0.6929, 0.5565],
  19481. [-0.1273, -0.0658, 0.7397, 0.2481, 0.3711, 0.2330, 0.5285, 0.5775],
  19482. [ 0.5797, 0.4041, 0.7436, 0.2373, 0.4315, 0.1334, 0.5688, 0.5258],
  19483. [ 0.6172, 0.4352, 0.7226, 0.2292, 0.3858, 0.1790, 0.5193, 0.5609],
  19484. [ 0.6064, 0.3998, 0.8377, 0.5035, 0.3568, 0.5040, 0.6954, 0.5307],
  19485. [ 0.6284, 0.4079, 0.8532, 0.4978, 0.3948, 0.4959, 0.5576, 0.4846]],
  19486. device='cuda:0', grad_fn=<AddmmBackward>)
  19487. landmarks are: tensor([[[0.6186, 0.4129, 0.8900, 0.4567, 0.3938, 0.5417, 0.5734, 0.5110],
  19488. [0.6250, 0.4013, 0.8525, 0.5417, 0.4038, 0.5117, 0.6325, 0.5017],
  19489. [0.6254, 0.4076, 0.8700, 0.3267, 0.4150, 0.3083, 0.7050, 0.5609],
  19490. [0.0000, 0.0000, 0.7335, 0.2569, 0.3787, 0.2667, 0.5066, 0.5578],
  19491. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  19492. [0.6173, 0.4114, 0.7325, 0.2500, 0.4212, 0.1917, 0.5337, 0.5700],
  19493. [0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  19494. [0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950]]],
  19495. device='cuda:0')
  19496. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  19497. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  19498. loss_train: 0.043800648621981964
  19499. step: 28
  19500. running loss: 0.0015643088793564988
  19501.  
  19502. Train Steps: 28/90 Loss: 0.0016 torch.Size([8, 600, 800])
  19503. torch.Size([8, 8])
  19504. tensor([[0.6147, 0.4107, 0.8137, 0.3333, 0.3750, 0.2683, 0.5006, 0.5412],
  19505. [0.6185, 0.4080, 0.8625, 0.3483, 0.3788, 0.2650, 0.5320, 0.5272],
  19506. [0.6251, 0.4163, 0.8662, 0.4467, 0.3625, 0.3567, 0.6038, 0.5533],
  19507. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  19508. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  19509. [0.6276, 0.4235, 0.8888, 0.5333, 0.3800, 0.3117, 0.5427, 0.6164],
  19510. [0.6111, 0.4033, 0.8300, 0.3267, 0.3588, 0.3333, 0.5444, 0.5637],
  19511. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892]],
  19512. device='cuda:0', dtype=torch.float64)
  19513. predictions are: tensor([[0.5070, 0.3404, 0.7796, 0.3037, 0.3528, 0.2542, 0.5178, 0.5303],
  19514. [0.5961, 0.3965, 0.8401, 0.3414, 0.3790, 0.2841, 0.5402, 0.5174],
  19515. [0.6057, 0.3977, 0.8470, 0.4220, 0.3571, 0.3361, 0.6071, 0.5284],
  19516. [0.5779, 0.3969, 0.7766, 0.3082, 0.3443, 0.3543, 0.5923, 0.5151],
  19517. [0.6144, 0.4063, 0.8500, 0.4293, 0.3710, 0.4007, 0.5790, 0.5648],
  19518. [0.5379, 0.3636, 0.8434, 0.5175, 0.3964, 0.3437, 0.5801, 0.5920],
  19519. [0.6287, 0.4280, 0.8185, 0.3094, 0.3471, 0.3301, 0.5489, 0.5375],
  19520. [0.6084, 0.4077, 0.8355, 0.4366, 0.3513, 0.3734, 0.5412, 0.5617]],
  19521. device='cuda:0', grad_fn=<AddmmBackward>)
  19522. landmarks are: tensor([[[0.6147, 0.4107, 0.8138, 0.3333, 0.3750, 0.2683, 0.5006, 0.5412],
  19523. [0.6186, 0.4080, 0.8625, 0.3483, 0.3787, 0.2650, 0.5320, 0.5272],
  19524. [0.6252, 0.4162, 0.8662, 0.4467, 0.3625, 0.3567, 0.6037, 0.5533],
  19525. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  19526. [0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  19527. [0.6276, 0.4235, 0.8888, 0.5333, 0.3800, 0.3117, 0.5427, 0.6164],
  19528. [0.6111, 0.4033, 0.8300, 0.3267, 0.3587, 0.3333, 0.5444, 0.5637],
  19529. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892]]],
  19530. device='cuda:0')
  19531. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  19532. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  19533. loss_train: 0.044581994734471664
  19534. step: 29
  19535. running loss: 0.0015373101632576436
  19536. Train Steps: 29/90 Loss: 0.0015 torch.Size([8, 600, 800])
  19537. torch.Size([8, 8])
  19538. tensor([[0.6148, 0.4053, 0.8750, 0.4550, 0.4850, 0.5218, 0.5863, 0.5567],
  19539. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  19540. [0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  19541. [0.6084, 0.3981, 0.8588, 0.5233, 0.4600, 0.5367, 0.5680, 0.5006],
  19542. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  19543. [0.6339, 0.4123, 0.8638, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  19544. [0.6040, 0.4002, 0.7338, 0.2267, 0.3975, 0.2100, 0.5231, 0.4778],
  19545. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426]],
  19546. device='cuda:0', dtype=torch.float64)
  19547. predictions are: tensor([[0.5954, 0.4118, 0.8543, 0.4456, 0.4525, 0.4905, 0.5458, 0.5619],
  19548. [0.6419, 0.4508, 0.8598, 0.4282, 0.4053, 0.5013, 0.5462, 0.5402],
  19549. [0.6981, 0.4758, 0.8232, 0.5364, 0.3724, 0.3886, 0.6626, 0.5528],
  19550. [0.6648, 0.4608, 0.8242, 0.5233, 0.4306, 0.5186, 0.5066, 0.5186],
  19551. [0.6749, 0.4485, 0.8571, 0.3060, 0.3899, 0.3924, 0.6725, 0.5735],
  19552. [0.6380, 0.4405, 0.8437, 0.5172, 0.3811, 0.5291, 0.7002, 0.5559],
  19553. [0.6293, 0.4254, 0.7027, 0.2040, 0.3756, 0.1766, 0.5020, 0.5047],
  19554. [0.6604, 0.4439, 0.8687, 0.4279, 0.3489, 0.3185, 0.5489, 0.5276]],
  19555. device='cuda:0', grad_fn=<AddmmBackward>)
  19556. landmarks are: tensor([[[0.6148, 0.4053, 0.8750, 0.4550, 0.4850, 0.5218, 0.5863, 0.5567],
  19557. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  19558. [0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  19559. [0.6084, 0.3981, 0.8587, 0.5233, 0.4600, 0.5367, 0.5680, 0.5006],
  19560. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  19561. [0.6339, 0.4123, 0.8637, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  19562. [0.6040, 0.4002, 0.7337, 0.2267, 0.3975, 0.2100, 0.5231, 0.4778],
  19563. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426]]],
  19564. device='cuda:0')
  19565. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  19566. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  19567. loss_train: 0.04559074839926325
  19568. step: 30
  19569. running loss: 0.001519691613308775
  19570. Train Steps: 30/90 Loss: 0.0015 torch.Size([8, 600, 800])
  19571. torch.Size([8, 8])
  19572. tensor([[0.6060, 0.3924, 0.8450, 0.5717, 0.4200, 0.5217, 0.5253, 0.4752],
  19573. [0.6305, 0.3983, 0.8950, 0.4833, 0.3688, 0.4683, 0.6375, 0.5117],
  19574. [0.6147, 0.4112, 0.7988, 0.3200, 0.3775, 0.2767, 0.5150, 0.5550],
  19575. [0.6167, 0.4048, 0.6831, 0.3639, 0.3763, 0.3017, 0.5700, 0.5883],
  19576. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092],
  19577. [0.6110, 0.3984, 0.8750, 0.4933, 0.4625, 0.4950, 0.5578, 0.5676],
  19578. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284],
  19579. [0.6332, 0.4118, 0.9238, 0.4267, 0.4012, 0.4733, 0.7525, 0.5436]],
  19580. device='cuda:0', dtype=torch.float64)
  19581. predictions are: tensor([[0.6383, 0.4051, 0.8426, 0.5858, 0.4593, 0.5259, 0.5491, 0.5228],
  19582. [0.7065, 0.4534, 0.8905, 0.4899, 0.3710, 0.4799, 0.6445, 0.5080],
  19583. [0.6552, 0.4527, 0.7933, 0.3352, 0.3711, 0.2825, 0.5235, 0.5575],
  19584. [0.7061, 0.4787, 0.7113, 0.3436, 0.3689, 0.3042, 0.5457, 0.5738],
  19585. [0.6846, 0.4498, 0.8709, 0.5279, 0.3860, 0.5291, 0.6037, 0.5104],
  19586. [0.6956, 0.4875, 0.8825, 0.5053, 0.4610, 0.5115, 0.5537, 0.5858],
  19587. [0.5936, 0.3968, 0.7494, 0.2417, 0.4437, 0.1887, 0.5902, 0.5428],
  19588. [0.7195, 0.4881, 0.8953, 0.4497, 0.4147, 0.4874, 0.7514, 0.5634]],
  19589. device='cuda:0', grad_fn=<AddmmBackward>)
  19590. landmarks are: tensor([[[0.6060, 0.3924, 0.8450, 0.5717, 0.4200, 0.5217, 0.5253, 0.4752],
  19591. [0.6305, 0.3983, 0.8950, 0.4833, 0.3688, 0.4683, 0.6375, 0.5117],
  19592. [0.6147, 0.4112, 0.7987, 0.3200, 0.3775, 0.2767, 0.5150, 0.5550],
  19593. [0.6167, 0.4048, 0.6831, 0.3639, 0.3762, 0.3017, 0.5700, 0.5883],
  19594. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092],
  19595. [0.6110, 0.3984, 0.8750, 0.4933, 0.4625, 0.4950, 0.5578, 0.5676],
  19596. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284],
  19597. [0.6332, 0.4118, 0.9237, 0.4267, 0.4013, 0.4733, 0.7525, 0.5436]]],
  19598. device='cuda:0')
  19599. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  19600. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  19601. loss_train: 0.04676109258434735
  19602. step: 31
  19603. running loss: 0.0015084223414305598
  19604. Train Steps: 31/90 Loss: 0.0015 torch.Size([8, 600, 800])
  19605. torch.Size([8, 8])
  19606. tensor([[0.6277, 0.4057, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  19607. [0.6197, 0.4090, 0.7825, 0.2500, 0.4200, 0.2483, 0.5988, 0.5667],
  19608. [0.6125, 0.3999, 0.8750, 0.4883, 0.4750, 0.4700, 0.5533, 0.5617],
  19609. [0.6222, 0.3957, 0.8838, 0.5017, 0.3937, 0.4600, 0.5900, 0.5017],
  19610. [0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5837, 0.5317],
  19611. [0.6307, 0.4029, 0.8988, 0.4817, 0.3937, 0.3500, 0.7311, 0.5378],
  19612. [0.6059, 0.4002, 0.7562, 0.2767, 0.3538, 0.3033, 0.5529, 0.5455],
  19613. [0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133]],
  19614. device='cuda:0', dtype=torch.float64)
  19615. predictions are: tensor([[0.6692, 0.4374, 0.8151, 0.2765, 0.4514, 0.2167, 0.6261, 0.5162],
  19616. [0.5933, 0.3883, 0.7637, 0.2549, 0.4119, 0.2749, 0.5830, 0.6016],
  19617. [0.6993, 0.4590, 0.8689, 0.5085, 0.4934, 0.4897, 0.5562, 0.5860],
  19618. [0.6885, 0.4309, 0.8752, 0.5305, 0.4021, 0.4740, 0.5839, 0.5138],
  19619. [0.7271, 0.4863, 0.8403, 0.3473, 0.3716, 0.3249, 0.5934, 0.5531],
  19620. [0.7110, 0.4639, 0.8905, 0.4920, 0.4118, 0.3469, 0.7221, 0.5454],
  19621. [0.7115, 0.4821, 0.7279, 0.3006, 0.3740, 0.3376, 0.5692, 0.5613],
  19622. [0.7465, 0.4876, 0.8552, 0.5547, 0.3968, 0.4857, 0.6460, 0.5347]],
  19623. device='cuda:0', grad_fn=<AddmmBackward>)
  19624. landmarks are: tensor([[[0.6277, 0.4056, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  19625. [0.6197, 0.4090, 0.7825, 0.2500, 0.4200, 0.2483, 0.5987, 0.5667],
  19626. [0.6125, 0.3999, 0.8750, 0.4883, 0.4750, 0.4700, 0.5533, 0.5617],
  19627. [0.6222, 0.3957, 0.8838, 0.5017, 0.3938, 0.4600, 0.5900, 0.5017],
  19628. [0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5838, 0.5317],
  19629. [0.6307, 0.4029, 0.8988, 0.4817, 0.3938, 0.3500, 0.7311, 0.5378],
  19630. [0.6059, 0.4002, 0.7563, 0.2767, 0.3537, 0.3033, 0.5529, 0.5455],
  19631. [0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133]]],
  19632. device='cuda:0')
  19633. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  19634. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  19635. loss_train: 0.048350871045840904
  19636. step: 32
  19637. running loss: 0.0015109647201825283
  19638.  
  19639. Train Steps: 32/90 Loss: 0.0015 torch.Size([8, 600, 800])
  19640. torch.Size([8, 8])
  19641. tensor([[0.6204, 0.4007, 0.7838, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  19642. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  19643. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  19644. [0.6140, 0.4070, 0.8700, 0.5000, 0.4612, 0.4900, 0.5260, 0.5852],
  19645. [0.6311, 0.4008, 0.7935, 0.5746, 0.3900, 0.5033, 0.6955, 0.5366],
  19646. [ nan, nan, 0.7515, 0.2708, 0.3987, 0.2267, 0.5162, 0.5567],
  19647. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  19648. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869]],
  19649. device='cuda:0', dtype=torch.float64)
  19650. predictions are: tensor([[0.7183, 0.4435, 0.7780, 0.2519, 0.4622, 0.1673, 0.6079, 0.5254],
  19651. [0.6890, 0.4253, 0.8122, 0.2988, 0.4185, 0.2408, 0.6214, 0.5200],
  19652. [0.7704, 0.4847, 0.8624, 0.4061, 0.3801, 0.4079, 0.6020, 0.5743],
  19653. [0.7668, 0.4961, 0.9040, 0.5122, 0.4858, 0.4915, 0.5727, 0.5860],
  19654. [0.7760, 0.4867, 0.8250, 0.5689, 0.4193, 0.4985, 0.7223, 0.5521],
  19655. [0.1626, 0.0791, 0.7773, 0.2831, 0.4128, 0.2344, 0.5432, 0.5718],
  19656. [0.7430, 0.4662, 0.8972, 0.3785, 0.4033, 0.4556, 0.7315, 0.5490],
  19657. [0.7862, 0.4972, 0.8749, 0.5722, 0.4226, 0.4457, 0.5695, 0.4978]],
  19658. device='cuda:0', grad_fn=<AddmmBackward>)
  19659. landmarks are: tensor([[[0.6204, 0.4007, 0.7837, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  19660. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  19661. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  19662. [0.6140, 0.4070, 0.8700, 0.5000, 0.4613, 0.4900, 0.5260, 0.5852],
  19663. [0.6311, 0.4008, 0.7935, 0.5746, 0.3900, 0.5033, 0.6955, 0.5366],
  19664. [0.0000, 0.0000, 0.7515, 0.2708, 0.3988, 0.2267, 0.5163, 0.5567],
  19665. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  19666. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869]]],
  19667. device='cuda:0')
  19668. loss_train_step before backward: tensor(0.0034, device='cuda:0', grad_fn=<MseLossBackward>)
  19669. loss_train_step after backward: tensor(0.0034, device='cuda:0', grad_fn=<MseLossBackward>)
  19670. loss_train: 0.0517900716222357
  19671. step: 33
  19672. running loss: 0.0015693961097647182
  19673. Train Steps: 33/90 Loss: 0.0016 torch.Size([8, 600, 800])
  19674. torch.Size([8, 8])
  19675. tensor([[0.6239, 0.4123, 0.8313, 0.2550, 0.4500, 0.2050, 0.6175, 0.5400],
  19676. [0.6200, 0.4049, 0.8638, 0.5617, 0.4125, 0.5100, 0.6013, 0.5317],
  19677. [0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5837, 0.5317],
  19678. [0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367],
  19679. [0.6224, 0.4097, 0.7438, 0.2267, 0.3850, 0.2850, 0.5988, 0.5250],
  19680. [0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  19681. [0.6129, 0.4063, 0.8738, 0.5250, 0.4313, 0.4733, 0.5230, 0.5874],
  19682. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938]],
  19683. device='cuda:0', dtype=torch.float64)
  19684. predictions are: tensor([[0.6262, 0.4034, 0.8379, 0.2548, 0.4757, 0.2024, 0.6381, 0.5267],
  19685. [0.6725, 0.4191, 0.8749, 0.5841, 0.4385, 0.5090, 0.6299, 0.5375],
  19686. [0.6988, 0.4461, 0.8549, 0.3476, 0.3818, 0.3153, 0.6010, 0.5361],
  19687. [0.6572, 0.4037, 0.8795, 0.3257, 0.4055, 0.2845, 0.6564, 0.5286],
  19688. [0.6318, 0.4080, 0.7481, 0.2524, 0.4223, 0.3030, 0.6058, 0.5469],
  19689. [0.6642, 0.4098, 0.8865, 0.4767, 0.4578, 0.4683, 0.5849, 0.5601],
  19690. [0.6951, 0.4315, 0.8775, 0.5485, 0.4589, 0.4654, 0.5631, 0.5894],
  19691. [0.6888, 0.4206, 0.9079, 0.5414, 0.3968, 0.3758, 0.6470, 0.4819]],
  19692. device='cuda:0', grad_fn=<AddmmBackward>)
  19693. landmarks are: tensor([[[0.6239, 0.4123, 0.8313, 0.2550, 0.4500, 0.2050, 0.6175, 0.5400],
  19694. [0.6199, 0.4049, 0.8637, 0.5617, 0.4125, 0.5100, 0.6012, 0.5317],
  19695. [0.6225, 0.4116, 0.8662, 0.3517, 0.3663, 0.3233, 0.5838, 0.5317],
  19696. [0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367],
  19697. [0.6224, 0.4097, 0.7437, 0.2267, 0.3850, 0.2850, 0.5987, 0.5250],
  19698. [0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  19699. [0.6130, 0.4063, 0.8737, 0.5250, 0.4313, 0.4733, 0.5230, 0.5874],
  19700. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938]]],
  19701. device='cuda:0')
  19702. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  19703. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  19704. loss_train: 0.052406656759558246
  19705. step: 34
  19706. running loss: 0.001541372257634066
  19707. Train Steps: 34/90 Loss: 0.0015 torch.Size([8, 600, 800])
  19708. torch.Size([8, 8])
  19709. tensor([[0.6229, 0.4066, 0.7612, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  19710. [0.6268, 0.4029, 0.8500, 0.2683, 0.3937, 0.3500, 0.6860, 0.5297],
  19711. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791],
  19712. [0.6329, 0.4196, 0.9238, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748],
  19713. [0.6185, 0.4129, 0.8900, 0.4567, 0.3937, 0.5417, 0.5734, 0.5110],
  19714. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  19715. [0.6250, 0.4106, 0.8700, 0.3717, 0.3588, 0.4967, 0.6038, 0.5167],
  19716. [0.6192, 0.4128, 0.8513, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633]],
  19717. device='cuda:0', dtype=torch.float64)
  19718. predictions are: tensor([[0.6202, 0.3780, 0.7768, 0.2914, 0.4400, 0.2355, 0.5699, 0.5551],
  19719. [0.6527, 0.4156, 0.8333, 0.2886, 0.3895, 0.3419, 0.6861, 0.5263],
  19720. [0.7031, 0.4200, 0.8915, 0.5075, 0.4197, 0.4965, 0.6249, 0.4885],
  19721. [0.5822, 0.3480, 0.9264, 0.4856, 0.4590, 0.2685, 0.7117, 0.5417],
  19722. [0.7206, 0.4528, 0.9131, 0.4676, 0.4242, 0.5323, 0.5750, 0.5227],
  19723. [0.6428, 0.3892, 0.8874, 0.4650, 0.4004, 0.4453, 0.5611, 0.5443],
  19724. [0.6559, 0.4030, 0.8732, 0.3825, 0.3912, 0.4849, 0.6456, 0.5597],
  19725. [0.6766, 0.4138, 0.8838, 0.5775, 0.4509, 0.5091, 0.5833, 0.5711]],
  19726. device='cuda:0', grad_fn=<AddmmBackward>)
  19727. landmarks are: tensor([[[0.6229, 0.4066, 0.7613, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  19728. [0.6268, 0.4029, 0.8500, 0.2683, 0.3938, 0.3500, 0.6860, 0.5297],
  19729. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791],
  19730. [0.6329, 0.4196, 0.9237, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748],
  19731. [0.6186, 0.4129, 0.8900, 0.4567, 0.3938, 0.5417, 0.5734, 0.5110],
  19732. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  19733. [0.6250, 0.4105, 0.8700, 0.3717, 0.3587, 0.4967, 0.6037, 0.5167],
  19734. [0.6192, 0.4128, 0.8512, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633]]],
  19735. device='cuda:0')
  19736. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  19737. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  19738. loss_train: 0.05321587107027881
  19739. step: 35
  19740. running loss: 0.001520453459150823
  19741. Train Steps: 35/90 Loss: 0.0015 torch.Size([8, 600, 800])
  19742. torch.Size([8, 8])
  19743. tensor([[0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  19744. [0.6199, 0.4060, 0.8888, 0.4667, 0.3800, 0.5050, 0.6188, 0.5433],
  19745. [0.6127, 0.4118, 0.8650, 0.5083, 0.4088, 0.5367, 0.5300, 0.5456],
  19746. [0.6261, 0.4066, 0.8325, 0.2150, 0.4763, 0.2667, 0.7002, 0.5633],
  19747. [ nan, nan, 0.8888, 0.3100, 0.5262, 0.2817, 0.7145, 0.6003],
  19748. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  19749. [0.6317, 0.4038, 0.8287, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  19750. [0.6072, 0.4029, 0.7037, 0.2150, 0.3912, 0.2267, 0.5516, 0.5507]],
  19751. device='cuda:0', dtype=torch.float64)
  19752. predictions are: tensor([[0.6875, 0.4320, 0.8564, 0.5744, 0.4058, 0.4555, 0.5670, 0.5928],
  19753. [0.6644, 0.4154, 0.9198, 0.4740, 0.3834, 0.5058, 0.6488, 0.5174],
  19754. [0.6649, 0.4253, 0.8915, 0.5074, 0.4305, 0.5249, 0.5541, 0.5414],
  19755. [0.6130, 0.3686, 0.8652, 0.2302, 0.4823, 0.2487, 0.7005, 0.5395],
  19756. [0.0716, 0.0118, 0.9327, 0.3118, 0.5047, 0.2719, 0.7061, 0.5590],
  19757. [0.7000, 0.4239, 0.8849, 0.3953, 0.3745, 0.4427, 0.6062, 0.5183],
  19758. [0.6816, 0.4226, 0.8694, 0.5829, 0.3913, 0.4535, 0.6265, 0.4815],
  19759. [0.7006, 0.4562, 0.7253, 0.2361, 0.4113, 0.2272, 0.5532, 0.5245]],
  19760. device='cuda:0', grad_fn=<AddmmBackward>)
  19761. landmarks are: tensor([[[0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  19762. [0.6199, 0.4060, 0.8888, 0.4667, 0.3800, 0.5050, 0.6187, 0.5433],
  19763. [0.6127, 0.4118, 0.8650, 0.5083, 0.4087, 0.5367, 0.5300, 0.5456],
  19764. [0.6261, 0.4066, 0.8325, 0.2150, 0.4762, 0.2667, 0.7002, 0.5633],
  19765. [0.0000, 0.0000, 0.8888, 0.3100, 0.5263, 0.2817, 0.7145, 0.6003],
  19766. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  19767. [0.6317, 0.4038, 0.8288, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  19768. [0.6072, 0.4029, 0.7038, 0.2150, 0.3913, 0.2267, 0.5516, 0.5507]]],
  19769. device='cuda:0')
  19770. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  19771. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  19772. loss_train: 0.05410699962521903
  19773. step: 36
  19774. running loss: 0.0015029722118116398
  19775.  
  19776. Train Steps: 36/90 Loss: 0.0015 torch.Size([8, 600, 800])
  19777. torch.Size([8, 8])
  19778. tensor([[ nan, nan, 0.7225, 0.2167, 0.3987, 0.2283, 0.5427, 0.5181],
  19779. [0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767],
  19780. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  19781. [0.6229, 0.4107, 0.8137, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  19782. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  19783. [0.6198, 0.4164, 0.8700, 0.5067, 0.4625, 0.5650, 0.5464, 0.5197],
  19784. [0.6274, 0.4270, 0.8938, 0.4967, 0.3550, 0.4283, 0.5700, 0.5733],
  19785. [0.6277, 0.4029, 0.8250, 0.2433, 0.4325, 0.2100, 0.6366, 0.5207]],
  19786. device='cuda:0', dtype=torch.float64)
  19787. predictions are: tensor([[0.1819, 0.1039, 0.7380, 0.1940, 0.3809, 0.2428, 0.5503, 0.5216],
  19788. [0.5895, 0.3619, 0.9011, 0.5202, 0.3940, 0.4769, 0.5646, 0.5651],
  19789. [0.6023, 0.3690, 0.8773, 0.3985, 0.3527, 0.3939, 0.6104, 0.5608],
  19790. [0.5871, 0.3863, 0.8467, 0.2964, 0.4725, 0.2073, 0.5912, 0.5331],
  19791. [0.5822, 0.3717, 0.7326, 0.2561, 0.4252, 0.2154, 0.5554, 0.5833],
  19792. [0.5647, 0.3731, 0.8998, 0.5127, 0.4519, 0.5512, 0.5964, 0.5058],
  19793. [0.6267, 0.3978, 0.9258, 0.5146, 0.3502, 0.4465, 0.6199, 0.5489],
  19794. [0.6408, 0.3909, 0.8375, 0.2682, 0.4460, 0.2356, 0.6727, 0.5104]],
  19795. device='cuda:0', grad_fn=<AddmmBackward>)
  19796. landmarks are: tensor([[[0.0000, 0.0000, 0.7225, 0.2167, 0.3988, 0.2283, 0.5427, 0.5181],
  19797. [0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767],
  19798. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  19799. [0.6229, 0.4107, 0.8138, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  19800. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  19801. [0.6198, 0.4164, 0.8700, 0.5067, 0.4625, 0.5650, 0.5464, 0.5197],
  19802. [0.6274, 0.4270, 0.8938, 0.4967, 0.3550, 0.4283, 0.5700, 0.5733],
  19803. [0.6277, 0.4029, 0.8250, 0.2433, 0.4325, 0.2100, 0.6366, 0.5207]]],
  19804. device='cuda:0')
  19805. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  19806. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  19807. loss_train: 0.05532921847770922
  19808. step: 37
  19809. running loss: 0.0014953842831813302
  19810. Train Steps: 37/90 Loss: 0.0015 torch.Size([8, 600, 800])
  19811. torch.Size([8, 8])
  19812. tensor([[0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  19813. [0.6200, 0.4071, 0.7338, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  19814. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  19815. [ nan, nan, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  19816. [0.6185, 0.4098, 0.8838, 0.4900, 0.4537, 0.5800, 0.6288, 0.5400],
  19817. [0.6268, 0.4029, 0.8500, 0.2683, 0.3937, 0.3500, 0.6860, 0.5297],
  19818. [0.6203, 0.4021, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  19819. [0.6207, 0.4110, 0.8738, 0.5000, 0.4800, 0.5633, 0.6300, 0.5433]],
  19820. device='cuda:0', dtype=torch.float64)
  19821. predictions are: tensor([[ 0.5584, 0.3814, 0.7396, 0.3220, 0.4831, 0.1942, 0.5172, 0.6068],
  19822. [ 0.5950, 0.3975, 0.7396, 0.1990, 0.3960, 0.2508, 0.5917, 0.5550],
  19823. [ 0.5619, 0.3668, 0.7915, 0.3295, 0.3272, 0.3418, 0.5757, 0.5112],
  19824. [-0.0712, -0.0459, 0.8614, 0.2210, 0.4886, 0.2268, 0.6629, 0.5561],
  19825. [ 0.5797, 0.3691, 0.8868, 0.4916, 0.4381, 0.5639, 0.5886, 0.5283],
  19826. [ 0.5762, 0.3851, 0.8199, 0.2775, 0.3516, 0.3389, 0.6678, 0.5093],
  19827. [ 0.6372, 0.4073, 0.8711, 0.5256, 0.3495, 0.3846, 0.5695, 0.5189],
  19828. [ 0.6062, 0.3832, 0.8912, 0.4883, 0.4486, 0.5551, 0.5764, 0.5444]],
  19829. device='cuda:0', grad_fn=<AddmmBackward>)
  19830. landmarks are: tensor([[[0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  19831. [0.6200, 0.4071, 0.7337, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  19832. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  19833. [0.0000, 0.0000, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  19834. [0.6185, 0.4098, 0.8838, 0.4900, 0.4538, 0.5800, 0.6288, 0.5400],
  19835. [0.6268, 0.4029, 0.8500, 0.2683, 0.3938, 0.3500, 0.6860, 0.5297],
  19836. [0.6203, 0.4020, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  19837. [0.6207, 0.4110, 0.8737, 0.5000, 0.4800, 0.5633, 0.6300, 0.5433]]],
  19838. device='cuda:0')
  19839. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  19840. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  19841. loss_train: 0.056172026408603415
  19842. step: 38
  19843. running loss: 0.0014782112212790373
  19844. Train Steps: 38/90 Loss: 0.0015 torch.Size([8, 600, 800])
  19845. torch.Size([8, 8])
  19846. tensor([[0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153],
  19847. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  19848. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  19849. [0.6205, 0.4016, 0.8350, 0.2717, 0.3987, 0.2550, 0.5787, 0.5133],
  19850. [0.6022, 0.3994, 0.8025, 0.3350, 0.3350, 0.4400, 0.5565, 0.5025],
  19851. [0.6115, 0.4081, 0.6725, 0.2433, 0.4088, 0.1933, 0.5167, 0.5544],
  19852. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  19853. [0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033]],
  19854. device='cuda:0', dtype=torch.float64)
  19855. predictions are: tensor([[0.4933, 0.3400, 0.8663, 0.5160, 0.4407, 0.5056, 0.5874, 0.5225],
  19856. [0.4956, 0.3527, 0.7297, 0.2796, 0.4278, 0.2104, 0.5411, 0.5783],
  19857. [0.4876, 0.3190, 0.8714, 0.2921, 0.3974, 0.3867, 0.7012, 0.5721],
  19858. [0.5360, 0.3603, 0.8380, 0.2608, 0.3976, 0.2755, 0.5699, 0.5331],
  19859. [0.4918, 0.3276, 0.7912, 0.3296, 0.3477, 0.4230, 0.5811, 0.5216],
  19860. [0.4919, 0.3439, 0.6967, 0.2405, 0.4097, 0.2051, 0.5022, 0.5646],
  19861. [0.4976, 0.3403, 0.8864, 0.4499, 0.4459, 0.5219, 0.5894, 0.5640],
  19862. [0.5277, 0.3468, 0.8753, 0.4974, 0.3895, 0.5026, 0.5835, 0.5277]],
  19863. device='cuda:0', grad_fn=<AddmmBackward>)
  19864. landmarks are: tensor([[[0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153],
  19865. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  19866. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  19867. [0.6205, 0.4015, 0.8350, 0.2717, 0.3988, 0.2550, 0.5788, 0.5133],
  19868. [0.6022, 0.3994, 0.8025, 0.3350, 0.3350, 0.4400, 0.5565, 0.5025],
  19869. [0.6115, 0.4081, 0.6725, 0.2433, 0.4087, 0.1933, 0.5167, 0.5544],
  19870. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  19871. [0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033]]],
  19872. device='cuda:0')
  19873. loss_train_step before backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  19874. loss_train_step after backward: tensor(0.0024, device='cuda:0', grad_fn=<MseLossBackward>)
  19875. loss_train: 0.05856394310831092
  19876. step: 39
  19877. running loss: 0.0015016395668797672
  19878. Train Steps: 39/90 Loss: 0.0015 torch.Size([8, 600, 800])
  19879. torch.Size([8, 8])
  19880. tensor([[0.6201, 0.3970, 0.8413, 0.4950, 0.4413, 0.5183, 0.6088, 0.5400],
  19881. [0.6198, 0.4101, 0.8838, 0.5283, 0.3763, 0.5267, 0.5913, 0.5567],
  19882. [0.6182, 0.3972, 0.8552, 0.5914, 0.3683, 0.4181, 0.5688, 0.5378],
  19883. [0.6098, 0.3991, 0.8638, 0.4717, 0.4263, 0.4967, 0.5212, 0.5650],
  19884. [0.6124, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  19885. [0.6250, 0.4106, 0.8700, 0.3717, 0.3588, 0.4967, 0.6038, 0.5167],
  19886. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5363, 0.5550],
  19887. [0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533]],
  19888. device='cuda:0', dtype=torch.float64)
  19889. predictions are: tensor([[0.4798, 0.3501, 0.8590, 0.4827, 0.4369, 0.5005, 0.6019, 0.5522],
  19890. [0.5323, 0.3763, 0.8803, 0.5342, 0.3938, 0.5222, 0.5869, 0.5534],
  19891. [0.5625, 0.3739, 0.8506, 0.5645, 0.3774, 0.4322, 0.5961, 0.5285],
  19892. [0.4954, 0.3495, 0.8718, 0.4683, 0.4258, 0.4900, 0.5436, 0.5614],
  19893. [0.5050, 0.3452, 0.7314, 0.2752, 0.3728, 0.3034, 0.5197, 0.5486],
  19894. [0.5076, 0.3545, 0.8631, 0.3498, 0.3658, 0.4916, 0.6479, 0.5467],
  19895. [0.5609, 0.3935, 0.7184, 0.2431, 0.4048, 0.2111, 0.5341, 0.5568],
  19896. [0.5422, 0.3743, 0.7302, 0.2188, 0.3692, 0.3188, 0.6069, 0.5689]],
  19897. device='cuda:0', grad_fn=<AddmmBackward>)
  19898. landmarks are: tensor([[[0.6201, 0.3970, 0.8413, 0.4950, 0.4412, 0.5183, 0.6087, 0.5400],
  19899. [0.6198, 0.4101, 0.8838, 0.5283, 0.3762, 0.5267, 0.5913, 0.5567],
  19900. [0.6182, 0.3972, 0.8552, 0.5914, 0.3683, 0.4181, 0.5688, 0.5378],
  19901. [0.6098, 0.3991, 0.8637, 0.4717, 0.4263, 0.4967, 0.5213, 0.5650],
  19902. [0.6123, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  19903. [0.6250, 0.4105, 0.8700, 0.3717, 0.3587, 0.4967, 0.6037, 0.5167],
  19904. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5362, 0.5550],
  19905. [0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533]]],
  19906. device='cuda:0')
  19907. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  19908. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  19909. loss_train: 0.060194525722181424
  19910. step: 40
  19911. running loss: 0.0015048631430545356
  19912.  
  19913. Train Steps: 40/90 Loss: 0.0015 torch.Size([8, 600, 800])
  19914. torch.Size([8, 8])
  19915. tensor([[0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5988, 0.5667],
  19916. [0.6289, 0.4019, 0.8113, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332],
  19917. [0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915],
  19918. [0.6118, 0.4052, 0.8463, 0.3917, 0.3538, 0.3450, 0.5053, 0.5593],
  19919. [0.6276, 0.4095, 0.8237, 0.2250, 0.4662, 0.1783, 0.6171, 0.4869],
  19920. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  19921. [0.6227, 0.4083, 0.8938, 0.4800, 0.3800, 0.2950, 0.5737, 0.5350],
  19922. [0.6114, 0.4018, 0.7213, 0.1967, 0.3763, 0.2700, 0.5875, 0.5533]],
  19923. device='cuda:0', dtype=torch.float64)
  19924. predictions are: tensor([[0.5218, 0.3743, 0.8242, 0.3354, 0.3818, 0.3548, 0.5813, 0.5783],
  19925. [0.5036, 0.3452, 0.7850, 0.4899, 0.3735, 0.4991, 0.6932, 0.5218],
  19926. [0.5335, 0.3687, 0.8543, 0.4282, 0.3633, 0.3501, 0.5961, 0.5151],
  19927. [0.5147, 0.3757, 0.8075, 0.3539, 0.3500, 0.3448, 0.4714, 0.5646],
  19928. [0.5429, 0.3846, 0.7967, 0.2209, 0.4756, 0.2256, 0.6027, 0.5078],
  19929. [0.5002, 0.3524, 0.8492, 0.4275, 0.4547, 0.5480, 0.5696, 0.5613],
  19930. [0.5504, 0.4018, 0.8314, 0.4504, 0.3838, 0.3229, 0.5412, 0.5547],
  19931. [0.5075, 0.3631, 0.7072, 0.2028, 0.3909, 0.2826, 0.5505, 0.5508]],
  19932. device='cuda:0', grad_fn=<AddmmBackward>)
  19933. landmarks are: tensor([[[0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5987, 0.5667],
  19934. [0.6289, 0.4019, 0.8112, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332],
  19935. [0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915],
  19936. [0.6118, 0.4052, 0.8462, 0.3917, 0.3537, 0.3450, 0.5053, 0.5593],
  19937. [0.6276, 0.4095, 0.8238, 0.2250, 0.4663, 0.1783, 0.6171, 0.4869],
  19938. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  19939. [0.6227, 0.4083, 0.8938, 0.4800, 0.3800, 0.2950, 0.5738, 0.5350],
  19940. [0.6114, 0.4018, 0.7212, 0.1967, 0.3762, 0.2700, 0.5875, 0.5533]]],
  19941. device='cuda:0')
  19942. loss_train_step before backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  19943. loss_train_step after backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  19944. loss_train: 0.062190487602492794
  19945. step: 41
  19946. running loss: 0.0015168411610364096
  19947. Train Steps: 41/90 Loss: 0.0015 torch.Size([8, 600, 800])
  19948. torch.Size([8, 8])
  19949. tensor([[0.6245, 0.4100, 0.7762, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  19950. [0.6167, 0.4048, 0.6831, 0.3639, 0.3763, 0.3017, 0.5700, 0.5883],
  19951. [0.6085, 0.4005, 0.8400, 0.4317, 0.3763, 0.4750, 0.5476, 0.5058],
  19952. [0.6229, 0.4066, 0.8513, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  19953. [0.6268, 0.4061, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  19954. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795],
  19955. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683],
  19956. [0.6166, 0.4008, 0.8563, 0.5667, 0.4388, 0.4933, 0.5575, 0.5567]],
  19957. device='cuda:0', dtype=torch.float64)
  19958. predictions are: tensor([[0.5534, 0.3963, 0.7433, 0.2502, 0.4639, 0.1631, 0.5861, 0.5406],
  19959. [0.4998, 0.3601, 0.7076, 0.3060, 0.3642, 0.3174, 0.5446, 0.5721],
  19960. [0.5338, 0.3740, 0.8254, 0.3796, 0.3466, 0.4556, 0.5302, 0.5044],
  19961. [0.5335, 0.3762, 0.8138, 0.5281, 0.4133, 0.4841, 0.5833, 0.5231],
  19962. [0.5301, 0.3713, 0.8326, 0.2195, 0.4565, 0.2315, 0.6110, 0.5118],
  19963. [0.5860, 0.4042, 0.8220, 0.4887, 0.3692, 0.5435, 0.6988, 0.5536],
  19964. [0.5799, 0.4024, 0.8479, 0.4923, 0.3590, 0.3807, 0.5321, 0.5592],
  19965. [0.5488, 0.3756, 0.8268, 0.5125, 0.4210, 0.4862, 0.5559, 0.5352]],
  19966. device='cuda:0', grad_fn=<AddmmBackward>)
  19967. landmarks are: tensor([[[0.6245, 0.4100, 0.7763, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  19968. [0.6167, 0.4048, 0.6831, 0.3639, 0.3762, 0.3017, 0.5700, 0.5883],
  19969. [0.6084, 0.4005, 0.8400, 0.4317, 0.3762, 0.4750, 0.5476, 0.5058],
  19970. [0.6229, 0.4066, 0.8512, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  19971. [0.6268, 0.4060, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  19972. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795],
  19973. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683],
  19974. [0.6166, 0.4008, 0.8562, 0.5667, 0.4387, 0.4933, 0.5575, 0.5567]]],
  19975. device='cuda:0')
  19976. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  19977. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  19978. loss_train: 0.06355495491879992
  19979. step: 42
  19980. running loss: 0.0015132132123523792
  19981. Train Steps: 42/90 Loss: 0.0015 torch.Size([8, 600, 800])
  19982. torch.Size([8, 8])
  19983. tensor([[0.6284, 0.4127, 0.8538, 0.5867, 0.4363, 0.5083, 0.6038, 0.5433],
  19984. [0.6201, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  19985. [0.6173, 0.4114, 0.7325, 0.2500, 0.4213, 0.1917, 0.5338, 0.5700],
  19986. [ nan, nan, 0.8888, 0.3100, 0.5262, 0.2817, 0.7145, 0.6003],
  19987. [0.6100, 0.4016, 0.8600, 0.5067, 0.4612, 0.5233, 0.5086, 0.5519],
  19988. [0.6293, 0.4097, 0.8800, 0.2517, 0.5262, 0.2600, 0.7430, 0.5378],
  19989. [0.6185, 0.4129, 0.8900, 0.4567, 0.3937, 0.5417, 0.5734, 0.5110],
  19990. [0.6264, 0.4071, 0.9038, 0.3867, 0.3663, 0.3917, 0.6338, 0.5283]],
  19991. device='cuda:0', dtype=torch.float64)
  19992. predictions are: tensor([[0.6403, 0.4367, 0.8090, 0.5827, 0.4012, 0.4859, 0.5567, 0.5294],
  19993. [0.6944, 0.4701, 0.7308, 0.2138, 0.4012, 0.2153, 0.5763, 0.5016],
  19994. [0.6507, 0.4582, 0.7047, 0.2664, 0.4054, 0.1817, 0.5152, 0.5649],
  19995. [0.0979, 0.0755, 0.8524, 0.3215, 0.4901, 0.2879, 0.6982, 0.5806],
  19996. [0.6130, 0.4266, 0.8024, 0.5167, 0.4311, 0.5133, 0.5184, 0.5435],
  19997. [0.5819, 0.3945, 0.8331, 0.2693, 0.4923, 0.2523, 0.7011, 0.5474],
  19998. [0.6537, 0.4636, 0.8480, 0.4701, 0.3679, 0.5577, 0.5616, 0.5122],
  19999. [0.5964, 0.4112, 0.8555, 0.3988, 0.3533, 0.4023, 0.6510, 0.5319]],
  20000. device='cuda:0', grad_fn=<AddmmBackward>)
  20001. landmarks are: tensor([[[0.6284, 0.4127, 0.8537, 0.5867, 0.4363, 0.5083, 0.6037, 0.5433],
  20002. [0.6202, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  20003. [0.6173, 0.4114, 0.7325, 0.2500, 0.4212, 0.1917, 0.5337, 0.5700],
  20004. [0.0000, 0.0000, 0.8888, 0.3100, 0.5263, 0.2817, 0.7145, 0.6003],
  20005. [0.6100, 0.4016, 0.8600, 0.5067, 0.4613, 0.5233, 0.5086, 0.5519],
  20006. [0.6293, 0.4097, 0.8800, 0.2517, 0.5263, 0.2600, 0.7430, 0.5378],
  20007. [0.6186, 0.4129, 0.8900, 0.4567, 0.3938, 0.5417, 0.5734, 0.5110],
  20008. [0.6264, 0.4071, 0.9038, 0.3867, 0.3663, 0.3917, 0.6338, 0.5283]]],
  20009. device='cuda:0')
  20010. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  20011. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  20012. loss_train: 0.0646070629300084
  20013. step: 43
  20014. running loss: 0.0015024898355815905
  20015. Train Steps: 43/90 Loss: 0.0015 torch.Size([8, 600, 800])
  20016. torch.Size([8, 8])
  20017. tensor([[0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6063, 0.6083],
  20018. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  20019. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  20020. [0.6240, 0.4217, 0.8150, 0.3133, 0.4425, 0.2650, 0.5650, 0.5817],
  20021. [0.6148, 0.4053, 0.8750, 0.4550, 0.4850, 0.5218, 0.5863, 0.5567],
  20022. [0.6357, 0.4118, 0.8400, 0.2500, 0.5413, 0.1633, 0.6725, 0.5586],
  20023. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  20024. [0.6250, 0.4054, 0.8770, 0.4723, 0.4662, 0.5367, 0.6162, 0.5433]],
  20025. device='cuda:0', dtype=torch.float64)
  20026. predictions are: tensor([[0.6535, 0.4311, 0.8405, 0.3824, 0.3650, 0.4421, 0.6013, 0.5688],
  20027. [0.6359, 0.4276, 0.8393, 0.3346, 0.3727, 0.3086, 0.5698, 0.5271],
  20028. [0.6506, 0.4239, 0.8290, 0.5575, 0.3890, 0.4996, 0.7215, 0.5566],
  20029. [0.5780, 0.3946, 0.8061, 0.3165, 0.4428, 0.2544, 0.5824, 0.5658],
  20030. [0.6421, 0.4263, 0.8572, 0.4694, 0.4742, 0.5226, 0.5898, 0.5518],
  20031. [0.5819, 0.3656, 0.8441, 0.2752, 0.5485, 0.1852, 0.6449, 0.5419],
  20032. [0.6521, 0.4286, 0.7027, 0.2786, 0.4204, 0.2309, 0.5696, 0.5364],
  20033. [0.6317, 0.4114, 0.8712, 0.4954, 0.4499, 0.5450, 0.6113, 0.5097]],
  20034. device='cuda:0', grad_fn=<AddmmBackward>)
  20035. landmarks are: tensor([[[0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6062, 0.6083],
  20036. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  20037. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  20038. [0.6240, 0.4217, 0.8150, 0.3133, 0.4425, 0.2650, 0.5650, 0.5817],
  20039. [0.6148, 0.4053, 0.8750, 0.4550, 0.4850, 0.5218, 0.5863, 0.5567],
  20040. [0.6357, 0.4118, 0.8400, 0.2500, 0.5412, 0.1633, 0.6725, 0.5586],
  20041. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  20042. [0.6250, 0.4054, 0.8770, 0.4723, 0.4663, 0.5367, 0.6162, 0.5433]]],
  20043. device='cuda:0')
  20044. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  20045. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  20046. loss_train: 0.06501127360388637
  20047. step: 44
  20048. running loss: 0.001477528945542872
  20049.  
  20050. Train Steps: 44/90 Loss: 0.0015 torch.Size([8, 600, 800])
  20051. torch.Size([8, 8])
  20052. tensor([[0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5787, 0.5117],
  20053. [ nan, nan, 0.8463, 0.2550, 0.5850, 0.2133, 0.7129, 0.6072],
  20054. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600],
  20055. [ nan, nan, 0.7335, 0.2569, 0.3788, 0.2667, 0.5066, 0.5578],
  20056. [0.6153, 0.4117, 0.8688, 0.5167, 0.4895, 0.5647, 0.5524, 0.5136],
  20057. [0.6245, 0.4100, 0.7762, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  20058. [0.6216, 0.4099, 0.7225, 0.2033, 0.4188, 0.2217, 0.5975, 0.5283],
  20059. [0.6268, 0.4061, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300]],
  20060. device='cuda:0', dtype=torch.float64)
  20061. predictions are: tensor([[0.7726, 0.4826, 0.7481, 0.2335, 0.4231, 0.2256, 0.5740, 0.5187],
  20062. [0.3963, 0.2433, 0.8701, 0.3023, 0.5606, 0.2781, 0.7299, 0.5774],
  20063. [0.7517, 0.4843, 0.8652, 0.5935, 0.4058, 0.5193, 0.5969, 0.5574],
  20064. [0.2212, 0.1395, 0.7287, 0.2680, 0.3992, 0.2929, 0.5313, 0.5597],
  20065. [0.6715, 0.4485, 0.8737, 0.5333, 0.4792, 0.5533, 0.5826, 0.5285],
  20066. [0.7009, 0.4446, 0.7609, 0.2963, 0.4872, 0.1567, 0.6087, 0.5468],
  20067. [0.7345, 0.4826, 0.7151, 0.2483, 0.4275, 0.2223, 0.5629, 0.5467],
  20068. [0.7022, 0.4410, 0.8480, 0.2651, 0.4787, 0.2276, 0.6324, 0.5189]],
  20069. device='cuda:0', grad_fn=<AddmmBackward>)
  20070. landmarks are: tensor([[[0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5788, 0.5117],
  20071. [0.0000, 0.0000, 0.8462, 0.2550, 0.5850, 0.2133, 0.7129, 0.6072],
  20072. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600],
  20073. [0.0000, 0.0000, 0.7335, 0.2569, 0.3787, 0.2667, 0.5066, 0.5578],
  20074. [0.6154, 0.4117, 0.8687, 0.5167, 0.4895, 0.5647, 0.5524, 0.5136],
  20075. [0.6245, 0.4100, 0.7763, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  20076. [0.6216, 0.4099, 0.7225, 0.2033, 0.4187, 0.2217, 0.5975, 0.5283],
  20077. [0.6268, 0.4060, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300]]],
  20078. device='cuda:0')
  20079. loss_train_step before backward: tensor(0.0062, device='cuda:0', grad_fn=<MseLossBackward>)
  20080. loss_train_step after backward: tensor(0.0062, device='cuda:0', grad_fn=<MseLossBackward>)
  20081. loss_train: 0.07122712582349777
  20082. step: 45
  20083. running loss: 0.0015828250182999505
  20084. Train Steps: 45/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20085. torch.Size([8, 8])
  20086. tensor([[0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5787, 0.5117],
  20087. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  20088. [0.6268, 0.4029, 0.8500, 0.2683, 0.3937, 0.3500, 0.6860, 0.5297],
  20089. [0.6346, 0.4092, 0.7712, 0.5917, 0.4037, 0.4767, 0.7343, 0.5725],
  20090. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  20091. [0.6135, 0.4115, 0.8838, 0.4667, 0.4288, 0.6050, 0.5778, 0.5097],
  20092. [0.6293, 0.4097, 0.8800, 0.2517, 0.5262, 0.2600, 0.7430, 0.5378],
  20093. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617]],
  20094. device='cuda:0', dtype=torch.float64)
  20095. predictions are: tensor([[0.7331, 0.4592, 0.7735, 0.2241, 0.4503, 0.1924, 0.5746, 0.5400],
  20096. [0.6419, 0.4064, 0.8766, 0.4838, 0.4706, 0.4968, 0.5743, 0.5954],
  20097. [0.7422, 0.4753, 0.8519, 0.2872, 0.4188, 0.3272, 0.6793, 0.5440],
  20098. [0.6730, 0.4350, 0.8086, 0.5327, 0.4291, 0.4629, 0.6900, 0.5999],
  20099. [0.6489, 0.4106, 0.8866, 0.4799, 0.4025, 0.5063, 0.6117, 0.5371],
  20100. [0.6957, 0.4588, 0.8931, 0.4718, 0.4490, 0.5881, 0.5687, 0.5560],
  20101. [0.6324, 0.4016, 0.9098, 0.2668, 0.5592, 0.2225, 0.7408, 0.5735],
  20102. [0.6425, 0.4179, 0.9084, 0.5664, 0.4185, 0.4261, 0.5618, 0.6009]],
  20103. device='cuda:0', grad_fn=<AddmmBackward>)
  20104. landmarks are: tensor([[[0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5788, 0.5117],
  20105. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  20106. [0.6268, 0.4029, 0.8500, 0.2683, 0.3938, 0.3500, 0.6860, 0.5297],
  20107. [0.6346, 0.4092, 0.7713, 0.5917, 0.4038, 0.4767, 0.7343, 0.5725],
  20108. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  20109. [0.6135, 0.4115, 0.8838, 0.4667, 0.4288, 0.6050, 0.5778, 0.5097],
  20110. [0.6293, 0.4097, 0.8800, 0.2517, 0.5263, 0.2600, 0.7430, 0.5378],
  20111. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617]]],
  20112. device='cuda:0')
  20113. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  20114. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  20115. loss_train: 0.07247095857746899
  20116. step: 46
  20117. running loss: 0.001575455621249326
  20118. Train Steps: 46/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20119. torch.Size([8, 8])
  20120. tensor([[0.6117, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103],
  20121. [0.6255, 0.4017, 0.8688, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  20122. [0.6252, 0.4158, 0.8988, 0.4083, 0.3788, 0.4783, 0.6225, 0.5633],
  20123. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  20124. [0.6114, 0.4018, 0.7213, 0.1967, 0.3763, 0.2700, 0.5875, 0.5533],
  20125. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  20126. [0.6307, 0.4029, 0.8650, 0.5200, 0.3763, 0.4017, 0.7311, 0.5366],
  20127. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167]],
  20128. device='cuda:0', dtype=torch.float64)
  20129. predictions are: tensor([[0.6574, 0.4007, 0.7161, 0.2415, 0.4372, 0.2182, 0.5540, 0.5274],
  20130. [0.7188, 0.4395, 0.9126, 0.3551, 0.4240, 0.3446, 0.6664, 0.5151],
  20131. [0.6449, 0.4115, 0.9352, 0.4280, 0.4575, 0.4946, 0.6595, 0.5656],
  20132. [0.6935, 0.4357, 0.7239, 0.3226, 0.4177, 0.2775, 0.5561, 0.5865],
  20133. [0.6732, 0.4144, 0.7784, 0.2303, 0.4354, 0.2535, 0.6162, 0.5620],
  20134. [0.6671, 0.4141, 0.8405, 0.4290, 0.4107, 0.4286, 0.5424, 0.5497],
  20135. [0.6732, 0.4279, 0.9269, 0.5229, 0.4582, 0.3990, 0.7307, 0.5606],
  20136. [0.6479, 0.4105, 0.8007, 0.3074, 0.4547, 0.2792, 0.6077, 0.6354]],
  20137. device='cuda:0', grad_fn=<AddmmBackward>)
  20138. landmarks are: tensor([[[0.6116, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103],
  20139. [0.6255, 0.4017, 0.8687, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  20140. [0.6252, 0.4158, 0.8988, 0.4083, 0.3787, 0.4783, 0.6225, 0.5633],
  20141. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  20142. [0.6114, 0.4018, 0.7212, 0.1967, 0.3762, 0.2700, 0.5875, 0.5533],
  20143. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  20144. [0.6307, 0.4029, 0.8650, 0.5200, 0.3762, 0.4017, 0.7311, 0.5366],
  20145. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167]]],
  20146. device='cuda:0')
  20147. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  20148. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  20149. loss_train: 0.07407605787739158
  20150. step: 47
  20151. running loss: 0.001576086337816842
  20152. Train Steps: 47/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20153. torch.Size([8, 8])
  20154. tensor([[0.6336, 0.4191, 0.8938, 0.5167, 0.3937, 0.3517, 0.7343, 0.5748],
  20155. [0.6082, 0.4024, 0.8738, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  20156. [0.6107, 0.4013, 0.8700, 0.4650, 0.5049, 0.5176, 0.5850, 0.5550],
  20157. [0.6277, 0.4118, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  20158. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  20159. [0.6163, 0.4114, 0.7650, 0.2017, 0.3763, 0.2867, 0.5631, 0.5071],
  20160. [0.6075, 0.4007, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280],
  20161. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683]],
  20162. device='cuda:0', dtype=torch.float64)
  20163. predictions are: tensor([[0.6368, 0.3935, 0.8968, 0.4891, 0.4338, 0.3506, 0.7530, 0.5938],
  20164. [0.6039, 0.3732, 0.8718, 0.3862, 0.3848, 0.3704, 0.5681, 0.5286],
  20165. [0.6556, 0.4214, 0.8814, 0.4510, 0.5159, 0.4949, 0.6190, 0.5738],
  20166. [0.6569, 0.4089, 0.9095, 0.3692, 0.4205, 0.2534, 0.6525, 0.5490],
  20167. [0.6146, 0.3732, 0.8748, 0.4358, 0.4161, 0.4537, 0.5730, 0.5365],
  20168. [0.7169, 0.4363, 0.7584, 0.1976, 0.4032, 0.2690, 0.6160, 0.5167],
  20169. [0.6273, 0.3904, 0.8306, 0.4688, 0.4328, 0.4919, 0.5570, 0.5485],
  20170. [0.6928, 0.4262, 0.8915, 0.5248, 0.4117, 0.3709, 0.5685, 0.5948]],
  20171. device='cuda:0', grad_fn=<AddmmBackward>)
  20172. landmarks are: tensor([[[0.6336, 0.4191, 0.8938, 0.5167, 0.3938, 0.3517, 0.7343, 0.5748],
  20173. [0.6082, 0.4024, 0.8737, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  20174. [0.6107, 0.4013, 0.8700, 0.4650, 0.5049, 0.5176, 0.5850, 0.5550],
  20175. [0.6277, 0.4117, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  20176. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  20177. [0.6163, 0.4114, 0.7650, 0.2017, 0.3762, 0.2867, 0.5631, 0.5071],
  20178. [0.6075, 0.4006, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280],
  20179. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683]]],
  20180. device='cuda:0')
  20181. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  20182. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  20183. loss_train: 0.07483209832571447
  20184. step: 48
  20185. running loss: 0.0015590020484523848
  20186.  
  20187. Train Steps: 48/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20188. torch.Size([8, 8])
  20189. tensor([[ nan, nan, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  20190. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  20191. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  20192. [0.6197, 0.3986, 0.8800, 0.4617, 0.4188, 0.4783, 0.5687, 0.5550],
  20193. [0.6079, 0.3964, 0.7420, 0.2958, 0.3563, 0.2917, 0.5351, 0.4980],
  20194. [ nan, nan, 0.7515, 0.2708, 0.3987, 0.2267, 0.5162, 0.5567],
  20195. [0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  20196. [0.6236, 0.3966, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183]],
  20197. device='cuda:0', dtype=torch.float64)
  20198. predictions are: tensor([[0.1510, 0.0802, 0.8048, 0.3031, 0.3931, 0.2843, 0.6019, 0.5653],
  20199. [0.7519, 0.4651, 0.8721, 0.4573, 0.4482, 0.5135, 0.6305, 0.5623],
  20200. [0.7314, 0.4637, 0.8811, 0.5660, 0.4295, 0.4587, 0.5821, 0.5647],
  20201. [0.7187, 0.4577, 0.8832, 0.4578, 0.4222, 0.4926, 0.6105, 0.5489],
  20202. [0.7343, 0.4611, 0.7655, 0.2794, 0.3611, 0.2828, 0.5849, 0.4956],
  20203. [0.0396, 0.0147, 0.7697, 0.2914, 0.4217, 0.2391, 0.5808, 0.5600],
  20204. [0.7838, 0.4756, 0.8688, 0.5498, 0.4112, 0.4997, 0.6354, 0.5302],
  20205. [0.7763, 0.4855, 0.8959, 0.4841, 0.3739, 0.4266, 0.6179, 0.5216]],
  20206. device='cuda:0', grad_fn=<AddmmBackward>)
  20207. landmarks are: tensor([[[0.0000, 0.0000, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  20208. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  20209. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  20210. [0.6197, 0.3986, 0.8800, 0.4617, 0.4187, 0.4783, 0.5688, 0.5550],
  20211. [0.6079, 0.3964, 0.7420, 0.2958, 0.3562, 0.2917, 0.5351, 0.4980],
  20212. [0.0000, 0.0000, 0.7515, 0.2708, 0.3988, 0.2267, 0.5163, 0.5567],
  20213. [0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  20214. [0.6236, 0.3965, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183]]],
  20215. device='cuda:0')
  20216. loss_train_step before backward: tensor(0.0030, device='cuda:0', grad_fn=<MseLossBackward>)
  20217. loss_train_step after backward: tensor(0.0030, device='cuda:0', grad_fn=<MseLossBackward>)
  20218. loss_train: 0.07787081459537148
  20219. step: 49
  20220. running loss: 0.0015892002978647242
  20221. Train Steps: 49/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20222. torch.Size([8, 8])
  20223. tensor([[ nan, nan, 0.7225, 0.2167, 0.3987, 0.2283, 0.5427, 0.5181],
  20224. [0.6250, 0.4146, 0.8838, 0.3933, 0.3588, 0.4283, 0.6162, 0.5367],
  20225. [0.6064, 0.4019, 0.8650, 0.4517, 0.4037, 0.5367, 0.5703, 0.5609],
  20226. [0.6245, 0.4115, 0.8700, 0.4883, 0.4625, 0.5517, 0.6100, 0.5217],
  20227. [0.6162, 0.4134, 0.6700, 0.2467, 0.3962, 0.2533, 0.5737, 0.5467],
  20228. [0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  20229. [0.6263, 0.4065, 0.9038, 0.4317, 0.3588, 0.4550, 0.6325, 0.5250],
  20230. [0.6286, 0.4086, 0.8408, 0.2801, 0.4163, 0.2800, 0.6725, 0.5393]],
  20231. device='cuda:0', dtype=torch.float64)
  20232. predictions are: tensor([[-0.0325, -0.0318, 0.7265, 0.2437, 0.4054, 0.2141, 0.5640, 0.5116],
  20233. [ 0.5875, 0.3782, 0.8540, 0.4057, 0.3558, 0.4153, 0.6077, 0.5356],
  20234. [ 0.6916, 0.4526, 0.8666, 0.4584, 0.4000, 0.5510, 0.5852, 0.5289],
  20235. [ 0.6534, 0.4066, 0.8812, 0.5019, 0.4452, 0.5662, 0.6000, 0.5215],
  20236. [ 0.6641, 0.4348, 0.6917, 0.2722, 0.3654, 0.2443, 0.5700, 0.5324],
  20237. [ 0.6990, 0.4371, 0.8753, 0.3732, 0.3904, 0.2664, 0.6206, 0.4665],
  20238. [ 0.6395, 0.3998, 0.9076, 0.4593, 0.3602, 0.4819, 0.6314, 0.5227],
  20239. [ 0.5846, 0.3655, 0.8533, 0.2823, 0.4021, 0.2868, 0.7068, 0.5403]],
  20240. device='cuda:0', grad_fn=<AddmmBackward>)
  20241. landmarks are: tensor([[[0.0000, 0.0000, 0.7225, 0.2167, 0.3988, 0.2283, 0.5427, 0.5181],
  20242. [0.6250, 0.4146, 0.8838, 0.3933, 0.3587, 0.4283, 0.6162, 0.5367],
  20243. [0.6064, 0.4019, 0.8650, 0.4517, 0.4038, 0.5367, 0.5703, 0.5609],
  20244. [0.6245, 0.4115, 0.8700, 0.4883, 0.4625, 0.5517, 0.6100, 0.5217],
  20245. [0.6162, 0.4134, 0.6700, 0.2467, 0.3963, 0.2533, 0.5738, 0.5467],
  20246. [0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  20247. [0.6263, 0.4065, 0.9038, 0.4317, 0.3587, 0.4550, 0.6325, 0.5250],
  20248. [0.6286, 0.4086, 0.8408, 0.2801, 0.4162, 0.2800, 0.6725, 0.5393]]],
  20249. device='cuda:0')
  20250. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  20251. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  20252. loss_train: 0.07848756696330383
  20253. step: 50
  20254. running loss: 0.0015697513392660767
  20255. Train Steps: 50/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20256. torch.Size([8, 8])
  20257. tensor([[0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  20258. [0.6222, 0.4169, 0.8638, 0.5650, 0.4313, 0.4783, 0.5637, 0.5633],
  20259. [ nan, nan, 0.7981, 0.3194, 0.3625, 0.3167, 0.5040, 0.5563],
  20260. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  20261. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284],
  20262. [ nan, nan, 0.8300, 0.3150, 0.3588, 0.3383, 0.5208, 0.5194],
  20263. [0.6336, 0.4191, 0.8938, 0.5167, 0.3937, 0.3517, 0.7343, 0.5748],
  20264. [0.6064, 0.3953, 0.8738, 0.4417, 0.3663, 0.4683, 0.5511, 0.5416]],
  20265. device='cuda:0', dtype=torch.float64)
  20266. predictions are: tensor([[0.6583, 0.4061, 0.7852, 0.2628, 0.3901, 0.2268, 0.5825, 0.5011],
  20267. [0.6119, 0.4043, 0.8460, 0.5443, 0.3911, 0.4834, 0.5273, 0.5403],
  20268. [0.0739, 0.0346, 0.7801, 0.3044, 0.3294, 0.3241, 0.5190, 0.5207],
  20269. [0.6127, 0.3999, 0.8732, 0.4730, 0.3982, 0.5596, 0.6111, 0.5051],
  20270. [0.6407, 0.4011, 0.7635, 0.2083, 0.4314, 0.1839, 0.5845, 0.4947],
  20271. [0.1436, 0.0885, 0.7909, 0.3158, 0.3326, 0.3305, 0.5361, 0.5008],
  20272. [0.6194, 0.4075, 0.8671, 0.4859, 0.3701, 0.3710, 0.7046, 0.5399],
  20273. [0.6698, 0.4239, 0.8582, 0.4529, 0.3303, 0.4754, 0.5364, 0.4863]],
  20274. device='cuda:0', grad_fn=<AddmmBackward>)
  20275. landmarks are: tensor([[[0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  20276. [0.6222, 0.4169, 0.8637, 0.5650, 0.4313, 0.4783, 0.5638, 0.5633],
  20277. [0.0000, 0.0000, 0.7981, 0.3194, 0.3625, 0.3167, 0.5040, 0.5563],
  20278. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  20279. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284],
  20280. [0.0000, 0.0000, 0.8300, 0.3150, 0.3587, 0.3383, 0.5208, 0.5194],
  20281. [0.6336, 0.4191, 0.8938, 0.5167, 0.3938, 0.3517, 0.7343, 0.5748],
  20282. [0.6064, 0.3952, 0.8737, 0.4417, 0.3663, 0.4683, 0.5511, 0.5416]]],
  20283. device='cuda:0')
  20284. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  20285. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  20286. loss_train: 0.07955434458563104
  20287. step: 51
  20288. running loss: 0.0015598891095221772
  20289. Train Steps: 51/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20290. torch.Size([8, 8])
  20291. tensor([[0.6197, 0.4050, 0.7527, 0.2000, 0.4042, 0.2249, 0.5895, 0.4995],
  20292. [0.6204, 0.4007, 0.7838, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  20293. [0.6213, 0.4131, 0.8438, 0.3550, 0.3513, 0.4400, 0.5716, 0.5123],
  20294. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  20295. [0.6189, 0.4029, 0.8375, 0.5767, 0.4745, 0.4829, 0.5551, 0.5598],
  20296. [0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510],
  20297. [ nan, nan, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  20298. [0.6183, 0.4076, 0.8838, 0.4517, 0.3813, 0.4483, 0.5775, 0.5633]],
  20299. device='cuda:0', dtype=torch.float64)
  20300. predictions are: tensor([[ 0.5704, 0.3827, 0.7387, 0.2036, 0.3578, 0.2297, 0.5657, 0.4697],
  20301. [ 0.5516, 0.3512, 0.7640, 0.2236, 0.4117, 0.1581, 0.5666, 0.4943],
  20302. [ 0.5503, 0.3709, 0.8578, 0.3682, 0.3118, 0.4334, 0.5395, 0.4950],
  20303. [ 0.5605, 0.3689, 0.8584, 0.5159, 0.3863, 0.4964, 0.5315, 0.5135],
  20304. [ 0.5507, 0.3585, 0.8238, 0.5576, 0.4316, 0.4736, 0.5286, 0.5450],
  20305. [ 0.5587, 0.3720, 0.6769, 0.2708, 0.3132, 0.3024, 0.4931, 0.5178],
  20306. [-0.1807, -0.1097, 0.8726, 0.3411, 0.4731, 0.2189, 0.6576, 0.5594],
  20307. [ 0.5047, 0.3208, 0.8690, 0.4605, 0.3415, 0.4591, 0.5479, 0.5489]],
  20308. device='cuda:0', grad_fn=<AddmmBackward>)
  20309. landmarks are: tensor([[[0.6197, 0.4050, 0.7527, 0.2000, 0.4042, 0.2249, 0.5895, 0.4995],
  20310. [0.6204, 0.4007, 0.7837, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  20311. [0.6213, 0.4131, 0.8438, 0.3550, 0.3512, 0.4400, 0.5716, 0.5123],
  20312. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  20313. [0.6189, 0.4029, 0.8375, 0.5767, 0.4745, 0.4829, 0.5551, 0.5598],
  20314. [0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510],
  20315. [0.0000, 0.0000, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  20316. [0.6183, 0.4076, 0.8838, 0.4517, 0.3812, 0.4483, 0.5775, 0.5633]]],
  20317. device='cuda:0')
  20318. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  20319. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  20320. loss_train: 0.08149451989447698
  20321. step: 52
  20322. running loss: 0.001567202305663019
  20323.  
  20324. Train Steps: 52/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20325. torch.Size([8, 8])
  20326. tensor([[0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  20327. [0.6205, 0.4081, 0.8950, 0.4017, 0.3788, 0.4700, 0.5963, 0.5667],
  20328. [ nan, nan, 0.6900, 0.1917, 0.3937, 0.2367, 0.5240, 0.5246],
  20329. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  20330. [0.6262, 0.4163, 0.8850, 0.5183, 0.3763, 0.4150, 0.6025, 0.5500],
  20331. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  20332. [0.6076, 0.3953, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954],
  20333. [ nan, nan, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650]],
  20334. device='cuda:0', dtype=torch.float64)
  20335. predictions are: tensor([[ 0.6065, 0.3959, 0.8548, 0.5664, 0.3848, 0.4831, 0.5509, 0.5286],
  20336. [ 0.6251, 0.4124, 0.8550, 0.3978, 0.3658, 0.4721, 0.5616, 0.5464],
  20337. [-0.0914, -0.0326, 0.6882, 0.2071, 0.3913, 0.2299, 0.5179, 0.5272],
  20338. [ 0.5802, 0.3960, 0.7924, 0.1981, 0.4634, 0.1556, 0.6002, 0.4862],
  20339. [ 0.5282, 0.3601, 0.8603, 0.5014, 0.3412, 0.3819, 0.5508, 0.5523],
  20340. [ 0.5425, 0.3861, 0.8650, 0.4160, 0.3282, 0.3782, 0.5710, 0.5371],
  20341. [ 0.6114, 0.4252, 0.8005, 0.3769, 0.3246, 0.4035, 0.5220, 0.4932],
  20342. [-0.0446, -0.0088, 0.6980, 0.1995, 0.4239, 0.2405, 0.5390, 0.5463]],
  20343. device='cuda:0', grad_fn=<AddmmBackward>)
  20344. landmarks are: tensor([[[0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  20345. [0.6205, 0.4081, 0.8950, 0.4017, 0.3787, 0.4700, 0.5962, 0.5667],
  20346. [0.0000, 0.0000, 0.6900, 0.1917, 0.3938, 0.2367, 0.5240, 0.5246],
  20347. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  20348. [0.6262, 0.4163, 0.8850, 0.5183, 0.3762, 0.4150, 0.6025, 0.5500],
  20349. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  20350. [0.6076, 0.3952, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954],
  20351. [0.0000, 0.0000, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650]]],
  20352. device='cuda:0')
  20353. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  20354. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  20355. loss_train: 0.08231787505792454
  20356. step: 53
  20357. running loss: 0.0015531674539231044
  20358. Train Steps: 53/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20359. torch.Size([8, 8])
  20360. tensor([[0.6204, 0.4013, 0.8075, 0.2400, 0.4313, 0.2050, 0.5800, 0.5150],
  20361. [0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431],
  20362. [0.6110, 0.4047, 0.8700, 0.4483, 0.3713, 0.3967, 0.5088, 0.5517],
  20363. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  20364. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  20365. [0.6212, 0.4033, 0.8938, 0.4167, 0.3813, 0.4267, 0.5613, 0.5583],
  20366. [0.6058, 0.3978, 0.8287, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461],
  20367. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483]],
  20368. device='cuda:0', dtype=torch.float64)
  20369. predictions are: tensor([[0.4633, 0.3053, 0.7813, 0.2637, 0.4218, 0.2074, 0.5650, 0.5414],
  20370. [0.4423, 0.3266, 0.8592, 0.4890, 0.4347, 0.4827, 0.4844, 0.5379],
  20371. [0.4234, 0.3148, 0.8460, 0.4503, 0.3582, 0.4114, 0.4690, 0.5458],
  20372. [0.4643, 0.3248, 0.8627, 0.5094, 0.4220, 0.4801, 0.5468, 0.5422],
  20373. [0.4713, 0.3208, 0.8873, 0.5206, 0.4017, 0.5500, 0.6978, 0.5498],
  20374. [0.4593, 0.3276, 0.8906, 0.4396, 0.3651, 0.4203, 0.5323, 0.5475],
  20375. [0.4902, 0.3599, 0.8177, 0.3682, 0.3393, 0.4011, 0.5317, 0.5457],
  20376. [0.5154, 0.3562, 0.7241, 0.1829, 0.3850, 0.2528, 0.5751, 0.5372]],
  20377. device='cuda:0', grad_fn=<AddmmBackward>)
  20378. landmarks are: tensor([[[0.6204, 0.4013, 0.8075, 0.2400, 0.4313, 0.2050, 0.5800, 0.5150],
  20379. [0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431],
  20380. [0.6110, 0.4047, 0.8700, 0.4483, 0.3713, 0.3967, 0.5088, 0.5517],
  20381. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  20382. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  20383. [0.6212, 0.4033, 0.8938, 0.4167, 0.3812, 0.4267, 0.5612, 0.5583],
  20384. [0.6058, 0.3978, 0.8288, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461],
  20385. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483]]],
  20386. device='cuda:0')
  20387. loss_train_step before backward: tensor(0.0039, device='cuda:0', grad_fn=<MseLossBackward>)
  20388. loss_train_step after backward: tensor(0.0039, device='cuda:0', grad_fn=<MseLossBackward>)
  20389. loss_train: 0.08622756163822487
  20390. step: 54
  20391. running loss: 0.0015968066970041643
  20392. Train Steps: 54/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20393. torch.Size([8, 8])
  20394. tensor([[0.6299, 0.4303, 0.7963, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  20395. [0.6284, 0.4127, 0.8538, 0.5867, 0.4363, 0.5083, 0.6038, 0.5433],
  20396. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323],
  20397. [0.6132, 0.4066, 0.7259, 0.2402, 0.3588, 0.3300, 0.6000, 0.5600],
  20398. [ nan, nan, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  20399. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  20400. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  20401. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283]],
  20402. device='cuda:0', dtype=torch.float64)
  20403. predictions are: tensor([[ 0.4549, 0.3279, 0.8062, 0.3474, 0.4819, 0.2351, 0.5100, 0.6041],
  20404. [ 0.4452, 0.3079, 0.8588, 0.5672, 0.4430, 0.4954, 0.5422, 0.5496],
  20405. [ 0.5699, 0.3944, 0.7603, 0.1899, 0.4359, 0.2076, 0.5866, 0.5161],
  20406. [ 0.5667, 0.3971, 0.7179, 0.2186, 0.3601, 0.3267, 0.5661, 0.5567],
  20407. [-0.0764, -0.0384, 0.7836, 0.2590, 0.3608, 0.2841, 0.4919, 0.5454],
  20408. [ 0.4562, 0.3183, 0.9138, 0.3884, 0.4177, 0.3604, 0.6740, 0.5921],
  20409. [ 0.5548, 0.4049, 0.7312, 0.2623, 0.4146, 0.2199, 0.5294, 0.5516],
  20410. [ 0.5124, 0.3712, 0.8831, 0.3997, 0.3388, 0.3850, 0.5597, 0.5397]],
  20411. device='cuda:0', grad_fn=<AddmmBackward>)
  20412. landmarks are: tensor([[[0.6299, 0.4303, 0.7962, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  20413. [0.6284, 0.4127, 0.8537, 0.5867, 0.4363, 0.5083, 0.6037, 0.5433],
  20414. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323],
  20415. [0.6132, 0.4066, 0.7259, 0.2402, 0.3587, 0.3300, 0.6000, 0.5600],
  20416. [0.0000, 0.0000, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  20417. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  20418. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  20419. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283]]],
  20420. device='cuda:0')
  20421. loss_train_step before backward: tensor(0.0029, device='cuda:0', grad_fn=<MseLossBackward>)
  20422. loss_train_step after backward: tensor(0.0029, device='cuda:0', grad_fn=<MseLossBackward>)
  20423. loss_train: 0.0891112532117404
  20424. step: 55
  20425. running loss: 0.0016202046038498255
  20426. Train Steps: 55/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20427. torch.Size([8, 8])
  20428. tensor([[0.6353, 0.4128, 0.9138, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  20429. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  20430. [0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050],
  20431. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  20432. [0.6102, 0.4005, 0.8688, 0.5100, 0.4813, 0.5400, 0.5404, 0.5064],
  20433. [0.6333, 0.4037, 0.8638, 0.5733, 0.4012, 0.4717, 0.6369, 0.4938],
  20434. [0.6200, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  20435. [0.6133, 0.4066, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358]],
  20436. device='cuda:0', dtype=torch.float64)
  20437. predictions are: tensor([[0.2841, 0.2073, 0.8971, 0.3288, 0.4746, 0.3006, 0.6896, 0.6023],
  20438. [0.4716, 0.3356, 0.9035, 0.3814, 0.4286, 0.3491, 0.6755, 0.6058],
  20439. [0.5666, 0.3886, 0.8688, 0.3893, 0.3700, 0.4776, 0.5625, 0.5985],
  20440. [0.5145, 0.3731, 0.8393, 0.5072, 0.3919, 0.4774, 0.6830, 0.5975],
  20441. [0.5068, 0.3861, 0.8645, 0.4706, 0.4755, 0.5139, 0.5026, 0.5105],
  20442. [0.5235, 0.3674, 0.8710, 0.5243, 0.4125, 0.4725, 0.5630, 0.5267],
  20443. [0.5493, 0.3891, 0.8318, 0.2677, 0.4237, 0.2318, 0.5512, 0.5450],
  20444. [0.5084, 0.3757, 0.6835, 0.2245, 0.3632, 0.2362, 0.4987, 0.5296]],
  20445. device='cuda:0', grad_fn=<AddmmBackward>)
  20446. landmarks are: tensor([[[0.6353, 0.4128, 0.9137, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  20447. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  20448. [0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050],
  20449. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  20450. [0.6102, 0.4005, 0.8687, 0.5100, 0.4812, 0.5400, 0.5404, 0.5064],
  20451. [0.6334, 0.4037, 0.8637, 0.5733, 0.4013, 0.4717, 0.6369, 0.4938],
  20452. [0.6199, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  20453. [0.6133, 0.4065, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358]]],
  20454. device='cuda:0')
  20455. loss_train_step before backward: tensor(0.0045, device='cuda:0', grad_fn=<MseLossBackward>)
  20456. loss_train_step after backward: tensor(0.0045, device='cuda:0', grad_fn=<MseLossBackward>)
  20457. loss_train: 0.09363721316913143
  20458. step: 56
  20459. running loss: 0.0016720930923059183
  20460.  
  20461. Train Steps: 56/90 Loss: 0.0017 torch.Size([8, 600, 800])
  20462. torch.Size([8, 8])
  20463. tensor([[0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  20464. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  20465. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  20466. [0.6229, 0.4066, 0.8513, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  20467. [0.6117, 0.4019, 0.8538, 0.4067, 0.3513, 0.3583, 0.5663, 0.5133],
  20468. [0.6321, 0.4048, 0.8738, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  20469. [0.6296, 0.4045, 0.9138, 0.4100, 0.4232, 0.4242, 0.7422, 0.5297],
  20470. [0.6200, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391]],
  20471. device='cuda:0', dtype=torch.float64)
  20472. predictions are: tensor([[0.5629, 0.4024, 0.8491, 0.5074, 0.3920, 0.4756, 0.7119, 0.6109],
  20473. [0.5554, 0.3866, 0.8785, 0.4322, 0.3812, 0.4299, 0.5717, 0.5626],
  20474. [0.5874, 0.3855, 0.8805, 0.4502, 0.4399, 0.5208, 0.6653, 0.5731],
  20475. [0.5419, 0.3820, 0.8656, 0.5302, 0.4593, 0.4908, 0.5758, 0.5832],
  20476. [0.6008, 0.4064, 0.8500, 0.3611, 0.3444, 0.3462, 0.5353, 0.5433],
  20477. [0.5880, 0.4068, 0.8797, 0.5273, 0.3888, 0.4454, 0.5944, 0.5271],
  20478. [0.5795, 0.4050, 0.9005, 0.3755, 0.4074, 0.4259, 0.6626, 0.5615],
  20479. [0.6131, 0.4276, 0.8416, 0.2677, 0.4248, 0.2244, 0.5816, 0.5584]],
  20480. device='cuda:0', grad_fn=<AddmmBackward>)
  20481. landmarks are: tensor([[[0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  20482. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  20483. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  20484. [0.6229, 0.4066, 0.8512, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  20485. [0.6116, 0.4019, 0.8537, 0.4067, 0.3512, 0.3583, 0.5663, 0.5133],
  20486. [0.6321, 0.4048, 0.8737, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  20487. [0.6296, 0.4045, 0.9137, 0.4100, 0.4232, 0.4242, 0.7422, 0.5297],
  20488. [0.6199, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391]]],
  20489. device='cuda:0')
  20490. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  20491. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  20492. loss_train: 0.09452643216354772
  20493. step: 57
  20494. running loss: 0.0016583584590096092
  20495. Train Steps: 57/90 Loss: 0.0017 torch.Size([8, 600, 800])
  20496. torch.Size([8, 8])
  20497. tensor([[0.6064, 0.4019, 0.8650, 0.4517, 0.4037, 0.5367, 0.5703, 0.5609],
  20498. [0.6190, 0.4135, 0.8000, 0.4883, 0.3566, 0.3647, 0.5613, 0.5900],
  20499. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  20500. [0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038],
  20501. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  20502. [0.6120, 0.4014, 0.6863, 0.2817, 0.3700, 0.2783, 0.5513, 0.5667],
  20503. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6188, 0.5283],
  20504. [ nan, nan, 0.7192, 0.2346, 0.4037, 0.2050, 0.5138, 0.5650]],
  20505. device='cuda:0', dtype=torch.float64)
  20506. predictions are: tensor([[0.6112, 0.4222, 0.8758, 0.4470, 0.4294, 0.5529, 0.6551, 0.5872],
  20507. [0.6808, 0.4543, 0.8444, 0.5078, 0.3648, 0.3772, 0.6150, 0.5839],
  20508. [0.6741, 0.4665, 0.8564, 0.2615, 0.5054, 0.1738, 0.6894, 0.5344],
  20509. [0.6551, 0.4307, 0.8836, 0.5045, 0.4541, 0.5239, 0.6172, 0.5367],
  20510. [0.6273, 0.4249, 0.8685, 0.2997, 0.4325, 0.2790, 0.6880, 0.5231],
  20511. [0.5907, 0.3875, 0.7082, 0.2678, 0.3756, 0.2957, 0.5998, 0.5691],
  20512. [0.7871, 0.5148, 0.9175, 0.3656, 0.4089, 0.2759, 0.7013, 0.5536],
  20513. [0.1344, 0.0876, 0.7245, 0.2278, 0.4096, 0.2233, 0.5401, 0.5740]],
  20514. device='cuda:0', grad_fn=<AddmmBackward>)
  20515. landmarks are: tensor([[[0.6064, 0.4019, 0.8650, 0.4517, 0.4038, 0.5367, 0.5703, 0.5609],
  20516. [0.6190, 0.4135, 0.8000, 0.4883, 0.3566, 0.3647, 0.5612, 0.5900],
  20517. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  20518. [0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038],
  20519. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  20520. [0.6120, 0.4013, 0.6862, 0.2817, 0.3700, 0.2783, 0.5512, 0.5667],
  20521. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6187, 0.5283],
  20522. [0.0000, 0.0000, 0.7192, 0.2346, 0.4038, 0.2050, 0.5138, 0.5650]]],
  20523. device='cuda:0')
  20524. loss_train_step before backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  20525. loss_train_step after backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  20526. loss_train: 0.09650458296528086
  20527. step: 58
  20528. running loss: 0.0016638721200910494
  20529. Train Steps: 58/90 Loss: 0.0017 torch.Size([8, 600, 800])
  20530. torch.Size([8, 8])
  20531. tensor([[0.6148, 0.4076, 0.8666, 0.4820, 0.4138, 0.5067, 0.5250, 0.5767],
  20532. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5837, 0.5583],
  20533. [0.6201, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  20534. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  20535. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6038, 0.4833],
  20536. [0.6293, 0.4024, 0.8750, 0.5000, 0.4012, 0.5733, 0.7121, 0.5633],
  20537. [ nan, nan, 0.6793, 0.2110, 0.4012, 0.2167, 0.5112, 0.5583],
  20538. [0.6075, 0.4000, 0.8513, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280]],
  20539. device='cuda:0', dtype=torch.float64)
  20540. predictions are: tensor([[0.7382, 0.4823, 0.8761, 0.5222, 0.4304, 0.4852, 0.6209, 0.5718],
  20541. [0.7312, 0.4900, 0.8813, 0.4541, 0.4890, 0.5090, 0.6586, 0.5728],
  20542. [0.6612, 0.4320, 0.7609, 0.2048, 0.4153, 0.2033, 0.6580, 0.5155],
  20543. [0.6932, 0.4531, 0.8719, 0.5102, 0.4333, 0.4629, 0.6154, 0.5461],
  20544. [0.7277, 0.4609, 0.8940, 0.5155, 0.4000, 0.4746, 0.6561, 0.5016],
  20545. [0.7316, 0.4616, 0.9022, 0.5313, 0.4126, 0.5640, 0.7744, 0.5645],
  20546. [0.1607, 0.0987, 0.6774, 0.2018, 0.4002, 0.1998, 0.5467, 0.5342],
  20547. [0.7042, 0.4623, 0.8553, 0.5422, 0.4580, 0.5011, 0.6179, 0.5370]],
  20548. device='cuda:0', grad_fn=<AddmmBackward>)
  20549. landmarks are: tensor([[[0.6148, 0.4076, 0.8666, 0.4820, 0.4137, 0.5067, 0.5250, 0.5767],
  20550. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5838, 0.5583],
  20551. [0.6202, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  20552. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  20553. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6037, 0.4833],
  20554. [0.6293, 0.4024, 0.8750, 0.5000, 0.4013, 0.5733, 0.7121, 0.5633],
  20555. [0.0000, 0.0000, 0.6793, 0.2110, 0.4013, 0.2167, 0.5113, 0.5583],
  20556. [0.6075, 0.4000, 0.8512, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280]]],
  20557. device='cuda:0')
  20558. loss_train_step before backward: tensor(0.0030, device='cuda:0', grad_fn=<MseLossBackward>)
  20559. loss_train_step after backward: tensor(0.0030, device='cuda:0', grad_fn=<MseLossBackward>)
  20560. loss_train: 0.09950864681741223
  20561. step: 59
  20562. running loss: 0.0016865872341934275
  20563. Train Steps: 59/90 Loss: 0.0017 torch.Size([8, 600, 800])
  20564. torch.Size([8, 8])
  20565. tensor([[0.6204, 0.4007, 0.7838, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  20566. [0.6202, 0.4066, 0.8398, 0.2648, 0.3925, 0.2627, 0.5845, 0.5124],
  20567. [0.6182, 0.3930, 0.8841, 0.3892, 0.3556, 0.4967, 0.6222, 0.5279],
  20568. [0.6273, 0.4143, 0.8750, 0.5700, 0.3987, 0.4717, 0.6013, 0.5467],
  20569. [0.6193, 0.4108, 0.7438, 0.2700, 0.3650, 0.3683, 0.6238, 0.5717],
  20570. [0.6259, 0.4133, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  20571. [0.6263, 0.4039, 0.9000, 0.4400, 0.3625, 0.4667, 0.6424, 0.4804],
  20572. [0.6251, 0.4163, 0.8662, 0.4467, 0.3625, 0.3567, 0.6038, 0.5533]],
  20573. device='cuda:0', dtype=torch.float64)
  20574. predictions are: tensor([[0.6482, 0.3921, 0.7578, 0.2343, 0.4396, 0.1740, 0.6194, 0.5175],
  20575. [0.7013, 0.4344, 0.8072, 0.2890, 0.4050, 0.2726, 0.6344, 0.5235],
  20576. [0.6638, 0.3961, 0.8857, 0.4162, 0.3702, 0.4941, 0.6593, 0.5240],
  20577. [0.6293, 0.3778, 0.8640, 0.6180, 0.4028, 0.4876, 0.6442, 0.5462],
  20578. [0.6033, 0.3709, 0.7571, 0.2885, 0.3631, 0.3846, 0.6689, 0.5528],
  20579. [0.7017, 0.4492, 0.8180, 0.2445, 0.4934, 0.1715, 0.6521, 0.4938],
  20580. [0.6669, 0.3932, 0.9262, 0.4911, 0.3702, 0.4925, 0.6464, 0.5024],
  20581. [0.7239, 0.4449, 0.8679, 0.4768, 0.3647, 0.3830, 0.6458, 0.5466]],
  20582. device='cuda:0', grad_fn=<AddmmBackward>)
  20583. landmarks are: tensor([[[0.6204, 0.4007, 0.7837, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  20584. [0.6202, 0.4066, 0.8398, 0.2648, 0.3925, 0.2627, 0.5845, 0.5124],
  20585. [0.6182, 0.3930, 0.8841, 0.3892, 0.3556, 0.4967, 0.6222, 0.5279],
  20586. [0.6273, 0.4143, 0.8750, 0.5700, 0.3988, 0.4717, 0.6012, 0.5467],
  20587. [0.6193, 0.4108, 0.7437, 0.2700, 0.3650, 0.3683, 0.6237, 0.5717],
  20588. [0.6259, 0.4132, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  20589. [0.6263, 0.4038, 0.9000, 0.4400, 0.3625, 0.4667, 0.6424, 0.4804],
  20590. [0.6252, 0.4162, 0.8662, 0.4467, 0.3625, 0.3567, 0.6037, 0.5533]]],
  20591. device='cuda:0')
  20592. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  20593. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  20594. loss_train: 0.10043607087573037
  20595. step: 60
  20596. running loss: 0.0016739345145955061
  20597.  
  20598. Train Steps: 60/90 Loss: 0.0017 torch.Size([8, 600, 800])
  20599. torch.Size([8, 8])
  20600. tensor([[0.6109, 0.4015, 0.7668, 0.3639, 0.3513, 0.3667, 0.5200, 0.5641],
  20601. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  20602. [0.6218, 0.4137, 0.7263, 0.2233, 0.4075, 0.2650, 0.6212, 0.5783],
  20603. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  20604. [0.6329, 0.4196, 0.9238, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748],
  20605. [0.6226, 0.4103, 0.8575, 0.3450, 0.4388, 0.2067, 0.5787, 0.5383],
  20606. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  20607. [0.6175, 0.4091, 0.7863, 0.2800, 0.3638, 0.3583, 0.6188, 0.5433]],
  20608. device='cuda:0', dtype=torch.float64)
  20609. predictions are: tensor([[0.6287, 0.3835, 0.7721, 0.3537, 0.3589, 0.3807, 0.5404, 0.5136],
  20610. [0.7034, 0.4113, 0.8991, 0.4467, 0.3619, 0.3990, 0.6077, 0.5131],
  20611. [0.6951, 0.4118, 0.7277, 0.2470, 0.4020, 0.2793, 0.6083, 0.5252],
  20612. [0.7054, 0.4191, 0.8589, 0.4918, 0.3990, 0.4911, 0.5657, 0.5131],
  20613. [0.7142, 0.4282, 0.9088, 0.4964, 0.4427, 0.3011, 0.7445, 0.5255],
  20614. [0.7862, 0.4827, 0.8517, 0.3759, 0.4657, 0.2416, 0.5966, 0.5169],
  20615. [0.7011, 0.4088, 0.8817, 0.5840, 0.3721, 0.4897, 0.6431, 0.4846],
  20616. [0.5986, 0.3631, 0.7848, 0.2676, 0.3712, 0.3852, 0.6174, 0.5040]],
  20617. device='cuda:0', grad_fn=<AddmmBackward>)
  20618. landmarks are: tensor([[[0.6109, 0.4015, 0.7668, 0.3639, 0.3512, 0.3667, 0.5200, 0.5641],
  20619. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  20620. [0.6218, 0.4137, 0.7262, 0.2233, 0.4075, 0.2650, 0.6212, 0.5783],
  20621. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  20622. [0.6329, 0.4196, 0.9237, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748],
  20623. [0.6226, 0.4103, 0.8575, 0.3450, 0.4387, 0.2067, 0.5788, 0.5383],
  20624. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  20625. [0.6175, 0.4091, 0.7862, 0.2800, 0.3638, 0.3583, 0.6187, 0.5433]]],
  20626. device='cuda:0')
  20627. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  20628. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  20629. loss_train: 0.10195993207162246
  20630. step: 61
  20631. running loss: 0.0016714742962561059
  20632. Train Steps: 61/90 Loss: 0.0017 torch.Size([8, 600, 800])
  20633. torch.Size([8, 8])
  20634. tensor([[0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  20635. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  20636. [0.6216, 0.4167, 0.8588, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667],
  20637. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  20638. [0.6193, 0.4050, 0.7313, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  20639. [0.6154, 0.4112, 0.7037, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  20640. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  20641. [0.6218, 0.4185, 0.7338, 0.2650, 0.4625, 0.1950, 0.5687, 0.5800]],
  20642. device='cuda:0', dtype=torch.float64)
  20643. predictions are: tensor([[0.6247, 0.3661, 0.7157, 0.2869, 0.4271, 0.2154, 0.5333, 0.5766],
  20644. [0.6447, 0.3784, 0.8774, 0.2905, 0.5059, 0.2369, 0.7035, 0.5209],
  20645. [0.7440, 0.4345, 0.8946, 0.6275, 0.4013, 0.5598, 0.6141, 0.5292],
  20646. [0.7049, 0.4133, 0.8685, 0.5509, 0.4432, 0.5565, 0.5157, 0.5046],
  20647. [0.6179, 0.3676, 0.7443, 0.2568, 0.4057, 0.2316, 0.5449, 0.5404],
  20648. [0.6160, 0.3691, 0.7046, 0.2429, 0.4079, 0.1970, 0.5292, 0.5258],
  20649. [0.6514, 0.3870, 0.9177, 0.3847, 0.3822, 0.4629, 0.6947, 0.5112],
  20650. [0.6073, 0.3739, 0.7232, 0.2653, 0.4281, 0.1929, 0.5387, 0.5596]],
  20651. device='cuda:0', grad_fn=<AddmmBackward>)
  20652. landmarks are: tensor([[[0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  20653. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  20654. [0.6216, 0.4167, 0.8587, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667],
  20655. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  20656. [0.6193, 0.4050, 0.7312, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  20657. [0.6154, 0.4112, 0.7038, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  20658. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  20659. [0.6218, 0.4185, 0.7337, 0.2650, 0.4625, 0.1950, 0.5688, 0.5800]]],
  20660. device='cuda:0')
  20661. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  20662. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  20663. loss_train: 0.10285078029846773
  20664. step: 62
  20665. running loss: 0.0016588835532010922
  20666. Train Steps: 62/90 Loss: 0.0017 torch.Size([8, 600, 800])
  20667. torch.Size([8, 8])
  20668. tensor([[0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  20669. [ nan, nan, 0.7412, 0.2200, 0.4450, 0.1517, 0.5312, 0.4983],
  20670. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  20671. [0.6145, 0.4007, 0.8775, 0.4533, 0.4562, 0.5533, 0.6088, 0.5533],
  20672. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  20673. [0.6126, 0.4039, 0.8237, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138],
  20674. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  20675. [0.6361, 0.4076, 0.8862, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297]],
  20676. device='cuda:0', dtype=torch.float64)
  20677. predictions are: tensor([[0.7079, 0.4305, 0.9018, 0.3640, 0.4346, 0.3790, 0.6754, 0.5370],
  20678. [0.1807, 0.0795, 0.7081, 0.2036, 0.4204, 0.1630, 0.4939, 0.5363],
  20679. [0.7387, 0.4639, 0.8380, 0.2717, 0.4993, 0.1799, 0.5849, 0.5192],
  20680. [0.7147, 0.4226, 0.8808, 0.4658, 0.4636, 0.5462, 0.5793, 0.5490],
  20681. [0.7203, 0.4444, 0.8488, 0.3098, 0.4027, 0.2778, 0.5885, 0.4940],
  20682. [0.6518, 0.3961, 0.8113, 0.4107, 0.3574, 0.3735, 0.5625, 0.5967],
  20683. [0.6976, 0.4263, 0.8730, 0.4764, 0.3820, 0.5138, 0.5499, 0.5146],
  20684. [0.7670, 0.4823, 0.8740, 0.5626, 0.3858, 0.4831, 0.6404, 0.5375]],
  20685. device='cuda:0', grad_fn=<AddmmBackward>)
  20686. landmarks are: tensor([[[0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  20687. [0.0000, 0.0000, 0.7412, 0.2200, 0.4450, 0.1517, 0.5312, 0.4983],
  20688. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  20689. [0.6145, 0.4007, 0.8775, 0.4533, 0.4563, 0.5533, 0.6087, 0.5533],
  20690. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810],
  20691. [0.6126, 0.4038, 0.8238, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138],
  20692. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  20693. [0.6361, 0.4076, 0.8863, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297]]],
  20694. device='cuda:0')
  20695. loss_train_step before backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  20696. loss_train_step after backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  20697. loss_train: 0.10497850127285346
  20698. step: 63
  20699. running loss: 0.00166632541702942
  20700. Train Steps: 63/90 Loss: 0.0017 torch.Size([8, 600, 800])
  20701. torch.Size([8, 8])
  20702. tensor([[0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5463, 0.5800],
  20703. [0.6273, 0.4100, 0.7137, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633],
  20704. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  20705. [0.6332, 0.4165, 0.9100, 0.3350, 0.4188, 0.3683, 0.7438, 0.5528],
  20706. [0.6075, 0.4000, 0.8513, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  20707. [0.6207, 0.4110, 0.8738, 0.5000, 0.4800, 0.5633, 0.6300, 0.5433],
  20708. [0.6126, 0.4073, 0.8750, 0.5133, 0.3800, 0.4333, 0.4986, 0.5378],
  20709. [0.6131, 0.4037, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680]],
  20710. device='cuda:0', dtype=torch.float64)
  20711. predictions are: tensor([[0.5489, 0.3741, 0.7715, 0.3019, 0.3528, 0.2394, 0.5202, 0.5679],
  20712. [0.5608, 0.3765, 0.7200, 0.2122, 0.3870, 0.2277, 0.5736, 0.5654],
  20713. [0.6949, 0.4254, 0.8806, 0.4887, 0.4158, 0.4699, 0.6067, 0.5323],
  20714. [0.6474, 0.4089, 0.9201, 0.3635, 0.4119, 0.3545, 0.7003, 0.5663],
  20715. [0.6503, 0.4275, 0.8552, 0.5129, 0.4399, 0.4933, 0.5123, 0.5289],
  20716. [0.7275, 0.4710, 0.8964, 0.4896, 0.4696, 0.5465, 0.5877, 0.5373],
  20717. [0.6598, 0.4386, 0.8856, 0.5192, 0.3787, 0.4187, 0.4934, 0.5376],
  20718. [0.5645, 0.3624, 0.7051, 0.2746, 0.3734, 0.2543, 0.5114, 0.5627]],
  20719. device='cuda:0', grad_fn=<AddmmBackward>)
  20720. landmarks are: tensor([[[0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5462, 0.5800],
  20721. [0.6273, 0.4099, 0.7138, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633],
  20722. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  20723. [0.6332, 0.4165, 0.9100, 0.3350, 0.4187, 0.3683, 0.7438, 0.5528],
  20724. [0.6075, 0.4000, 0.8512, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  20725. [0.6207, 0.4110, 0.8737, 0.5000, 0.4800, 0.5633, 0.6300, 0.5433],
  20726. [0.6126, 0.4073, 0.8750, 0.5133, 0.3800, 0.4333, 0.4986, 0.5378],
  20727. [0.6131, 0.4036, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680]]],
  20728. device='cuda:0')
  20729. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  20730. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  20731. loss_train: 0.10589164070552215
  20732. step: 64
  20733. running loss: 0.0016545568860237836
  20734.  
  20735. Train Steps: 64/90 Loss: 0.0017 torch.Size([8, 600, 800])
  20736. torch.Size([8, 8])
  20737. tensor([[0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683],
  20738. [0.6336, 0.4191, 0.8938, 0.5167, 0.3937, 0.3517, 0.7343, 0.5748],
  20739. [0.6058, 0.3978, 0.8287, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461],
  20740. [0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  20741. [0.6264, 0.4049, 0.8988, 0.4633, 0.3813, 0.4983, 0.6326, 0.4843],
  20742. [0.6113, 0.4006, 0.8700, 0.5350, 0.3638, 0.3767, 0.5097, 0.4882],
  20743. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  20744. [0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915]],
  20745. device='cuda:0', dtype=torch.float64)
  20746. predictions are: tensor([[0.6221, 0.4283, 0.8628, 0.4962, 0.3900, 0.3418, 0.5310, 0.5983],
  20747. [0.7085, 0.4884, 0.8724, 0.4659, 0.4139, 0.3262, 0.6839, 0.5850],
  20748. [0.5233, 0.3546, 0.8350, 0.3252, 0.3570, 0.3759, 0.5416, 0.5449],
  20749. [0.5835, 0.3883, 0.8672, 0.3821, 0.3898, 0.5025, 0.5429, 0.5312],
  20750. [0.6488, 0.4269, 0.8925, 0.4275, 0.3925, 0.4656, 0.6009, 0.5248],
  20751. [0.5921, 0.4094, 0.8436, 0.4683, 0.3680, 0.3727, 0.5096, 0.5457],
  20752. [0.6816, 0.4483, 0.8707, 0.4532, 0.4092, 0.5143, 0.6279, 0.5406],
  20753. [0.6375, 0.4225, 0.8894, 0.4271, 0.3852, 0.3125, 0.6002, 0.5315]],
  20754. device='cuda:0', grad_fn=<AddmmBackward>)
  20755. landmarks are: tensor([[[0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683],
  20756. [0.6336, 0.4191, 0.8938, 0.5167, 0.3938, 0.3517, 0.7343, 0.5748],
  20757. [0.6058, 0.3978, 0.8288, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461],
  20758. [0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  20759. [0.6264, 0.4049, 0.8988, 0.4633, 0.3812, 0.4983, 0.6326, 0.4843],
  20760. [0.6113, 0.4006, 0.8700, 0.5350, 0.3638, 0.3767, 0.5097, 0.4882],
  20761. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  20762. [0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915]]],
  20763. device='cuda:0')
  20764. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  20765. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  20766. loss_train: 0.10696358111454174
  20767. step: 65
  20768. running loss: 0.0016455935556083344
  20769. Train Steps: 65/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20770. torch.Size([8, 8])
  20771. tensor([[0.6201, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  20772. [0.6064, 0.4019, 0.8650, 0.4517, 0.4037, 0.5367, 0.5703, 0.5609],
  20773. [0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  20774. [0.6102, 0.4020, 0.8638, 0.3717, 0.3625, 0.5017, 0.6038, 0.5500],
  20775. [0.6286, 0.4078, 0.8063, 0.2267, 0.4788, 0.1533, 0.5953, 0.4913],
  20776. [ nan, nan, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  20777. [0.6336, 0.4191, 0.8938, 0.5167, 0.3937, 0.3517, 0.7343, 0.5748],
  20778. [0.6185, 0.4080, 0.8625, 0.3483, 0.3788, 0.2650, 0.5320, 0.5272]],
  20779. device='cuda:0', dtype=torch.float64)
  20780. predictions are: tensor([[0.5393, 0.3889, 0.7657, 0.1986, 0.4130, 0.1993, 0.5852, 0.5261],
  20781. [0.5829, 0.4071, 0.8496, 0.4321, 0.3864, 0.5271, 0.5797, 0.5473],
  20782. [0.6193, 0.4327, 0.8669, 0.4604, 0.4150, 0.4514, 0.5412, 0.5538],
  20783. [0.5629, 0.3962, 0.8623, 0.3543, 0.3623, 0.4686, 0.6162, 0.5318],
  20784. [0.4795, 0.3447, 0.7762, 0.2536, 0.4611, 0.1480, 0.5585, 0.5314],
  20785. [0.1735, 0.1281, 0.8874, 0.2998, 0.4933, 0.2157, 0.6412, 0.5723],
  20786. [0.6627, 0.4651, 0.8768, 0.4995, 0.3844, 0.3360, 0.7009, 0.5691],
  20787. [0.6295, 0.4397, 0.8468, 0.3628, 0.3846, 0.2779, 0.5188, 0.5498]],
  20788. device='cuda:0', grad_fn=<AddmmBackward>)
  20789. landmarks are: tensor([[[0.6202, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  20790. [0.6064, 0.4019, 0.8650, 0.4517, 0.4038, 0.5367, 0.5703, 0.5609],
  20791. [0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  20792. [0.6102, 0.4020, 0.8637, 0.3717, 0.3625, 0.5017, 0.6037, 0.5500],
  20793. [0.6286, 0.4078, 0.8062, 0.2267, 0.4787, 0.1533, 0.5953, 0.4913],
  20794. [0.0000, 0.0000, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  20795. [0.6336, 0.4191, 0.8938, 0.5167, 0.3938, 0.3517, 0.7343, 0.5748],
  20796. [0.6186, 0.4080, 0.8625, 0.3483, 0.3787, 0.2650, 0.5320, 0.5272]]],
  20797. device='cuda:0')
  20798. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  20799. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  20800. loss_train: 0.10856805794173852
  20801. step: 66
  20802. running loss: 0.001644970574874826
  20803. Train Steps: 66/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20804. torch.Size([8, 8])
  20805. tensor([[0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  20806. [0.6161, 0.4024, 0.8838, 0.4583, 0.3688, 0.3733, 0.5311, 0.5344],
  20807. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  20808. [0.6248, 0.4185, 0.8500, 0.5767, 0.4463, 0.4550, 0.5613, 0.5917],
  20809. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  20810. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  20811. [0.6203, 0.4021, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  20812. [0.6160, 0.4093, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808]],
  20813. device='cuda:0', dtype=torch.float64)
  20814. predictions are: tensor([[0.5448, 0.3894, 0.8145, 0.2338, 0.4142, 0.2702, 0.6812, 0.5341],
  20815. [0.5694, 0.3942, 0.8913, 0.4585, 0.3639, 0.3936, 0.5592, 0.5375],
  20816. [0.5310, 0.3859, 0.8618, 0.2606, 0.4940, 0.1833, 0.6343, 0.5224],
  20817. [0.5874, 0.4226, 0.8470, 0.5582, 0.4386, 0.4280, 0.5736, 0.5882],
  20818. [0.5255, 0.3685, 0.8450, 0.4086, 0.3649, 0.4420, 0.5704, 0.5609],
  20819. [0.5708, 0.3805, 0.8753, 0.4588, 0.4241, 0.4992, 0.6315, 0.5281],
  20820. [0.5485, 0.3756, 0.8864, 0.5135, 0.3552, 0.4017, 0.6095, 0.5225],
  20821. [0.5485, 0.3837, 0.8391, 0.4335, 0.3586, 0.4426, 0.5607, 0.5755]],
  20822. device='cuda:0', grad_fn=<AddmmBackward>)
  20823. landmarks are: tensor([[[0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  20824. [0.6161, 0.4024, 0.8838, 0.4583, 0.3688, 0.3733, 0.5311, 0.5344],
  20825. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  20826. [0.6248, 0.4185, 0.8500, 0.5767, 0.4462, 0.4550, 0.5612, 0.5917],
  20827. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  20828. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  20829. [0.6203, 0.4020, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  20830. [0.6160, 0.4092, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808]]],
  20831. device='cuda:0')
  20832. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  20833. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  20834. loss_train: 0.10938900749897584
  20835. step: 67
  20836. running loss: 0.0016326717537160573
  20837. Train Steps: 67/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20838. torch.Size([8, 8])
  20839. tensor([[0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5513, 0.5750],
  20840. [0.6100, 0.4071, 0.7601, 0.3444, 0.3400, 0.4117, 0.5625, 0.5617],
  20841. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  20842. [0.6223, 0.4028, 0.8988, 0.4200, 0.3763, 0.5733, 0.6375, 0.5167],
  20843. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  20844. [0.6268, 0.4029, 0.8500, 0.2683, 0.3937, 0.3500, 0.6860, 0.5297],
  20845. [0.6131, 0.4037, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  20846. [0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611]],
  20847. device='cuda:0', dtype=torch.float64)
  20848. predictions are: tensor([[0.4256, 0.3133, 0.7312, 0.2664, 0.4504, 0.1735, 0.5968, 0.5748],
  20849. [0.5070, 0.3664, 0.8311, 0.3491, 0.3637, 0.4180, 0.6030, 0.5581],
  20850. [0.5938, 0.4139, 0.9050, 0.5002, 0.4267, 0.5329, 0.6676, 0.5504],
  20851. [0.5597, 0.3864, 0.9127, 0.4380, 0.4046, 0.5862, 0.6737, 0.5258],
  20852. [0.4858, 0.3554, 0.8112, 0.3820, 0.3644, 0.4135, 0.5584, 0.5594],
  20853. [0.6025, 0.4247, 0.8645, 0.2894, 0.3939, 0.3295, 0.7073, 0.5438],
  20854. [0.5373, 0.3748, 0.7314, 0.3082, 0.4009, 0.2635, 0.5578, 0.5699],
  20855. [0.5954, 0.4201, 0.8981, 0.4821, 0.4482, 0.4694, 0.5724, 0.5615]],
  20856. device='cuda:0', grad_fn=<AddmmBackward>)
  20857. landmarks are: tensor([[[0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5512, 0.5750],
  20858. [0.6100, 0.4071, 0.7601, 0.3444, 0.3400, 0.4117, 0.5625, 0.5617],
  20859. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  20860. [0.6223, 0.4028, 0.8988, 0.4200, 0.3762, 0.5733, 0.6375, 0.5167],
  20861. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  20862. [0.6268, 0.4029, 0.8500, 0.2683, 0.3938, 0.3500, 0.6860, 0.5297],
  20863. [0.6131, 0.4036, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  20864. [0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611]]],
  20865. device='cuda:0')
  20866. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  20867. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  20868. loss_train: 0.11124868033220991
  20869. step: 68
  20870. running loss: 0.00163601000488544
  20871.  
  20872. Train Steps: 68/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20873. torch.Size([8, 8])
  20874. tensor([[0.6198, 0.4164, 0.8700, 0.5067, 0.4625, 0.5650, 0.5464, 0.5197],
  20875. [0.6069, 0.3975, 0.8625, 0.5083, 0.4388, 0.5483, 0.5650, 0.4967],
  20876. [0.6229, 0.4107, 0.8137, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  20877. [0.6192, 0.4128, 0.8513, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  20878. [0.6317, 0.4038, 0.8287, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  20879. [0.6218, 0.4137, 0.7263, 0.2233, 0.4075, 0.2650, 0.6212, 0.5783],
  20880. [0.6293, 0.4024, 0.8750, 0.5000, 0.4012, 0.5733, 0.7121, 0.5633],
  20881. [0.6250, 0.3993, 0.9138, 0.4333, 0.3763, 0.5217, 0.6995, 0.5320]],
  20882. device='cuda:0', dtype=torch.float64)
  20883. predictions are: tensor([[0.5563, 0.3870, 0.8563, 0.4827, 0.4403, 0.5231, 0.5433, 0.5044],
  20884. [0.5492, 0.3759, 0.8491, 0.4923, 0.4246, 0.5409, 0.5634, 0.5205],
  20885. [0.5895, 0.4069, 0.8284, 0.2998, 0.4732, 0.1896, 0.5892, 0.5576],
  20886. [0.5430, 0.3776, 0.8512, 0.5306, 0.4005, 0.5172, 0.5993, 0.5526],
  20887. [0.5404, 0.3731, 0.8525, 0.5451, 0.3619, 0.4570, 0.6332, 0.5041],
  20888. [0.5671, 0.3874, 0.7579, 0.2432, 0.3995, 0.2573, 0.6219, 0.5672],
  20889. [0.5508, 0.3591, 0.8674, 0.4709, 0.3768, 0.5651, 0.6735, 0.5446],
  20890. [0.5503, 0.3572, 0.9081, 0.4160, 0.3851, 0.5056, 0.7052, 0.5347]],
  20891. device='cuda:0', grad_fn=<AddmmBackward>)
  20892. landmarks are: tensor([[[0.6198, 0.4164, 0.8700, 0.5067, 0.4625, 0.5650, 0.5464, 0.5197],
  20893. [0.6069, 0.3975, 0.8625, 0.5083, 0.4387, 0.5483, 0.5650, 0.4967],
  20894. [0.6229, 0.4107, 0.8138, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  20895. [0.6192, 0.4128, 0.8512, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  20896. [0.6317, 0.4038, 0.8288, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  20897. [0.6218, 0.4137, 0.7262, 0.2233, 0.4075, 0.2650, 0.6212, 0.5783],
  20898. [0.6293, 0.4024, 0.8750, 0.5000, 0.4013, 0.5733, 0.7121, 0.5633],
  20899. [0.6250, 0.3993, 0.9137, 0.4333, 0.3762, 0.5217, 0.6995, 0.5320]]],
  20900. device='cuda:0')
  20901. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  20902. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  20903. loss_train: 0.11220953316660598
  20904. step: 69
  20905. running loss: 0.0016262251183566084
  20906. Train Steps: 69/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20907. torch.Size([8, 8])
  20908. tensor([[0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  20909. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  20910. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5787, 0.5550],
  20911. [0.6060, 0.3924, 0.8450, 0.5717, 0.4200, 0.5217, 0.5253, 0.4752],
  20912. [ nan, nan, 0.7097, 0.2346, 0.4250, 0.1850, 0.5175, 0.5583],
  20913. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5787, 0.5117],
  20914. [0.6255, 0.4017, 0.8688, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  20915. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309]],
  20916. device='cuda:0', dtype=torch.float64)
  20917. predictions are: tensor([[0.5894, 0.3943, 0.8446, 0.4361, 0.4505, 0.5642, 0.6108, 0.5490],
  20918. [0.5916, 0.3702, 0.8096, 0.5563, 0.4293, 0.5186, 0.5392, 0.4968],
  20919. [0.6256, 0.4082, 0.8094, 0.2928, 0.3440, 0.3792, 0.5929, 0.5366],
  20920. [0.5960, 0.3767, 0.8206, 0.5812, 0.4219, 0.5179, 0.5555, 0.4948],
  20921. [0.1566, 0.1218, 0.7175, 0.2339, 0.4316, 0.1995, 0.5217, 0.5481],
  20922. [0.6159, 0.4071, 0.7354, 0.2169, 0.4185, 0.2362, 0.6050, 0.5164],
  20923. [0.6355, 0.4033, 0.8359, 0.3240, 0.3578, 0.3834, 0.6605, 0.5133],
  20924. [0.6122, 0.3977, 0.8758, 0.4785, 0.3735, 0.3831, 0.6451, 0.5342]],
  20925. device='cuda:0', grad_fn=<AddmmBackward>)
  20926. landmarks are: tensor([[[0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  20927. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  20928. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5788, 0.5550],
  20929. [0.6060, 0.3924, 0.8450, 0.5717, 0.4200, 0.5217, 0.5253, 0.4752],
  20930. [0.0000, 0.0000, 0.7097, 0.2346, 0.4250, 0.1850, 0.5175, 0.5583],
  20931. [0.6161, 0.4040, 0.7525, 0.2133, 0.4125, 0.2067, 0.5788, 0.5117],
  20932. [0.6255, 0.4017, 0.8687, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  20933. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309]]],
  20934. device='cuda:0')
  20935. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  20936. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  20937. loss_train: 0.11311025434406474
  20938. step: 70
  20939. running loss: 0.001615860776343782
  20940. Train Steps: 70/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20941. torch.Size([8, 8])
  20942. tensor([[0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5513, 0.5750],
  20943. [0.6250, 0.4054, 0.8770, 0.4723, 0.4662, 0.5367, 0.6162, 0.5433],
  20944. [0.6257, 0.4167, 0.8775, 0.3433, 0.3563, 0.4133, 0.6200, 0.5667],
  20945. [0.6250, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6088, 0.5183],
  20946. [0.6275, 0.4048, 0.8488, 0.2883, 0.4463, 0.2033, 0.6321, 0.5155],
  20947. [0.6261, 0.3987, 0.9045, 0.4208, 0.3600, 0.4633, 0.6570, 0.5162],
  20948. [0.6097, 0.4000, 0.7325, 0.2667, 0.3450, 0.3517, 0.5284, 0.5045],
  20949. [0.6085, 0.4005, 0.8400, 0.4317, 0.3763, 0.4750, 0.5476, 0.5058]],
  20950. device='cuda:0', dtype=torch.float64)
  20951. predictions are: tensor([[0.5502, 0.3574, 0.6851, 0.2700, 0.4356, 0.2077, 0.5738, 0.5523],
  20952. [0.5865, 0.3874, 0.8746, 0.4779, 0.4575, 0.5594, 0.6084, 0.5240],
  20953. [0.6299, 0.3938, 0.8675, 0.3658, 0.3588, 0.4424, 0.6396, 0.5507],
  20954. [0.5341, 0.3422, 0.8595, 0.4866, 0.4463, 0.5918, 0.5941, 0.5159],
  20955. [0.6748, 0.4351, 0.8300, 0.2889, 0.4593, 0.2303, 0.6556, 0.5105],
  20956. [0.5803, 0.3658, 0.8792, 0.4309, 0.3655, 0.5083, 0.6469, 0.5170],
  20957. [0.5794, 0.3641, 0.7371, 0.3003, 0.3485, 0.3589, 0.5397, 0.5130],
  20958. [0.6032, 0.3811, 0.8296, 0.4438, 0.3641, 0.4955, 0.5187, 0.5050]],
  20959. device='cuda:0', grad_fn=<AddmmBackward>)
  20960. landmarks are: tensor([[[0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5512, 0.5750],
  20961. [0.6250, 0.4054, 0.8770, 0.4723, 0.4663, 0.5367, 0.6162, 0.5433],
  20962. [0.6257, 0.4167, 0.8775, 0.3433, 0.3562, 0.4133, 0.6200, 0.5667],
  20963. [0.6251, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6087, 0.5183],
  20964. [0.6275, 0.4048, 0.8487, 0.2883, 0.4462, 0.2033, 0.6321, 0.5155],
  20965. [0.6261, 0.3987, 0.9045, 0.4208, 0.3600, 0.4633, 0.6570, 0.5162],
  20966. [0.6097, 0.4000, 0.7325, 0.2667, 0.3450, 0.3517, 0.5284, 0.5045],
  20967. [0.6084, 0.4005, 0.8400, 0.4317, 0.3762, 0.4750, 0.5476, 0.5058]]],
  20968. device='cuda:0')
  20969. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  20970. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  20971. loss_train: 0.11381898814579472
  20972. step: 71
  20973. running loss: 0.0016030843400816157
  20974. Train Steps: 71/90 Loss: 0.0016 torch.Size([8, 600, 800])
  20975. torch.Size([8, 8])
  20976. tensor([[0.6307, 0.4029, 0.8988, 0.4817, 0.3937, 0.3500, 0.7311, 0.5378],
  20977. [0.6189, 0.4049, 0.8888, 0.4417, 0.4213, 0.5200, 0.5988, 0.5633],
  20978. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  20979. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  20980. [0.6125, 0.3999, 0.8750, 0.4883, 0.4750, 0.4700, 0.5533, 0.5617],
  20981. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  20982. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  20983. [0.6142, 0.3982, 0.8650, 0.4883, 0.3912, 0.4317, 0.5315, 0.5350]],
  20984. device='cuda:0', dtype=torch.float64)
  20985. predictions are: tensor([[0.6276, 0.3894, 0.8747, 0.4654, 0.3921, 0.3863, 0.6660, 0.5179],
  20986. [0.6014, 0.3731, 0.8574, 0.4411, 0.4104, 0.5811, 0.5857, 0.5430],
  20987. [0.6147, 0.3785, 0.7877, 0.2300, 0.4743, 0.1988, 0.6161, 0.4771],
  20988. [0.6622, 0.4007, 0.8515, 0.5078, 0.3759, 0.5819, 0.6995, 0.5133],
  20989. [0.5559, 0.3539, 0.8504, 0.4778, 0.4623, 0.5141, 0.5119, 0.5238],
  20990. [0.6134, 0.3951, 0.7303, 0.3397, 0.5047, 0.2150, 0.5570, 0.5806],
  20991. [0.5881, 0.3722, 0.7941, 0.3264, 0.3500, 0.3278, 0.4857, 0.5187],
  20992. [0.5848, 0.3471, 0.8362, 0.4906, 0.3879, 0.4857, 0.5017, 0.4967]],
  20993. device='cuda:0', grad_fn=<AddmmBackward>)
  20994. landmarks are: tensor([[[0.6307, 0.4029, 0.8988, 0.4817, 0.3938, 0.3500, 0.7311, 0.5378],
  20995. [0.6189, 0.4049, 0.8888, 0.4417, 0.4212, 0.5200, 0.5987, 0.5633],
  20996. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  20997. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  20998. [0.6125, 0.3999, 0.8750, 0.4883, 0.4750, 0.4700, 0.5533, 0.5617],
  20999. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  21000. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  21001. [0.6143, 0.3982, 0.8650, 0.4883, 0.3913, 0.4317, 0.5315, 0.5350]]],
  21002. device='cuda:0')
  21003. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  21004. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  21005. loss_train: 0.11468475608853623
  21006. step: 72
  21007. running loss: 0.0015928438345630032
  21008.  
  21009. Train Steps: 72/90 Loss: 0.0016 torch.Size([8, 600, 800])
  21010. torch.Size([8, 8])
  21011. tensor([[0.6098, 0.3991, 0.8638, 0.4717, 0.4263, 0.4967, 0.5212, 0.5650],
  21012. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  21013. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  21014. [0.6262, 0.4163, 0.8850, 0.5183, 0.3763, 0.4150, 0.6025, 0.5500],
  21015. [0.6228, 0.4119, 0.7938, 0.2233, 0.4674, 0.1773, 0.6188, 0.5433],
  21016. [0.6274, 0.4003, 0.8638, 0.5967, 0.3688, 0.4900, 0.6108, 0.4661],
  21017. [0.6135, 0.4115, 0.8838, 0.4667, 0.4288, 0.6050, 0.5778, 0.5097],
  21018. [0.6140, 0.4070, 0.8700, 0.5000, 0.4612, 0.4900, 0.5260, 0.5852]],
  21019. device='cuda:0', dtype=torch.float64)
  21020. predictions are: tensor([[0.6321, 0.3942, 0.8170, 0.4639, 0.4226, 0.4967, 0.5175, 0.5260],
  21021. [0.6072, 0.3800, 0.8428, 0.2650, 0.4421, 0.2778, 0.6690, 0.5455],
  21022. [0.6211, 0.3606, 0.8914, 0.4243, 0.4077, 0.3264, 0.6256, 0.5129],
  21023. [0.6232, 0.3805, 0.8581, 0.4738, 0.3662, 0.4002, 0.5599, 0.5180],
  21024. [0.6466, 0.4008, 0.7879, 0.2387, 0.4705, 0.1779, 0.5898, 0.5244],
  21025. [0.6400, 0.3735, 0.8262, 0.5660, 0.3768, 0.4563, 0.6083, 0.4759],
  21026. [0.5742, 0.3790, 0.8567, 0.4308, 0.4156, 0.5919, 0.5513, 0.5092],
  21027. [0.6213, 0.3897, 0.8506, 0.4837, 0.4374, 0.4786, 0.5193, 0.5460]],
  21028. device='cuda:0', grad_fn=<AddmmBackward>)
  21029. landmarks are: tensor([[[0.6098, 0.3991, 0.8637, 0.4717, 0.4263, 0.4967, 0.5213, 0.5650],
  21030. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  21031. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  21032. [0.6262, 0.4163, 0.8850, 0.5183, 0.3762, 0.4150, 0.6025, 0.5500],
  21033. [0.6228, 0.4119, 0.7937, 0.2233, 0.4674, 0.1773, 0.6187, 0.5433],
  21034. [0.6274, 0.4003, 0.8637, 0.5967, 0.3688, 0.4900, 0.6108, 0.4661],
  21035. [0.6135, 0.4115, 0.8838, 0.4667, 0.4288, 0.6050, 0.5778, 0.5097],
  21036. [0.6140, 0.4070, 0.8700, 0.5000, 0.4613, 0.4900, 0.5260, 0.5852]]],
  21037. device='cuda:0')
  21038. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  21039. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  21040. loss_train: 0.11528142628958449
  21041. step: 73
  21042. running loss: 0.001579197620405267
  21043. Train Steps: 73/90 Loss: 0.0016 torch.Size([8, 600, 800])
  21044. torch.Size([8, 8])
  21045. tensor([[0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317],
  21046. [0.6224, 0.4061, 0.8988, 0.4300, 0.3838, 0.4750, 0.6112, 0.5483],
  21047. [0.6270, 0.4267, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  21048. [ nan, nan, 0.7850, 0.2700, 0.4288, 0.1717, 0.5199, 0.4999],
  21049. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  21050. [0.6321, 0.4048, 0.8738, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  21051. [0.6273, 0.4143, 0.8750, 0.5700, 0.3987, 0.4717, 0.6013, 0.5467],
  21052. [0.6357, 0.4097, 0.9038, 0.3883, 0.4213, 0.2950, 0.6686, 0.5390]],
  21053. device='cuda:0', dtype=torch.float64)
  21054. predictions are: tensor([[0.7195, 0.4545, 0.8844, 0.5364, 0.3846, 0.4108, 0.5699, 0.5352],
  21055. [0.6740, 0.4419, 0.9011, 0.4222, 0.3800, 0.5256, 0.5829, 0.5424],
  21056. [0.6853, 0.4666, 0.7156, 0.2920, 0.4794, 0.2020, 0.5704, 0.5991],
  21057. [0.2321, 0.1452, 0.7913, 0.2523, 0.4388, 0.2067, 0.5140, 0.5255],
  21058. [0.6632, 0.4370, 0.8896, 0.4825, 0.3672, 0.4591, 0.6066, 0.5275],
  21059. [0.6851, 0.4356, 0.8823, 0.5456, 0.3809, 0.4654, 0.6006, 0.5075],
  21060. [0.6783, 0.4285, 0.8513, 0.5570, 0.3972, 0.4829, 0.5697, 0.5386],
  21061. [0.7579, 0.4764, 0.8849, 0.3669, 0.4426, 0.3150, 0.6563, 0.5561]],
  21062. device='cuda:0', grad_fn=<AddmmBackward>)
  21063. landmarks are: tensor([[[0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317],
  21064. [0.6224, 0.4061, 0.8988, 0.4300, 0.3837, 0.4750, 0.6112, 0.5483],
  21065. [0.6270, 0.4266, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  21066. [0.0000, 0.0000, 0.7850, 0.2700, 0.4288, 0.1717, 0.5199, 0.4999],
  21067. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  21068. [0.6321, 0.4048, 0.8737, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  21069. [0.6273, 0.4143, 0.8750, 0.5700, 0.3988, 0.4717, 0.6012, 0.5467],
  21070. [0.6357, 0.4097, 0.9038, 0.3883, 0.4212, 0.2950, 0.6686, 0.5390]]],
  21071. device='cuda:0')
  21072. loss_train_step before backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  21073. loss_train_step after backward: tensor(0.0022, device='cuda:0', grad_fn=<MseLossBackward>)
  21074. loss_train: 0.11752213741419837
  21075. step: 74
  21076. running loss: 0.0015881369920837617
  21077. Train Steps: 74/90 Loss: 0.0016 torch.Size([8, 600, 800])
  21078. torch.Size([8, 8])
  21079. tensor([[0.6102, 0.4020, 0.8638, 0.3717, 0.3625, 0.5017, 0.6038, 0.5500],
  21080. [0.6097, 0.3988, 0.8650, 0.5250, 0.4213, 0.5200, 0.5675, 0.5050],
  21081. [0.6212, 0.4159, 0.8675, 0.5783, 0.4088, 0.4317, 0.5613, 0.5917],
  21082. [0.6152, 0.4131, 0.6863, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  21083. [0.6267, 0.4065, 0.8313, 0.2467, 0.4788, 0.1733, 0.6312, 0.5133],
  21084. [0.6317, 0.4038, 0.8287, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  21085. [0.6273, 0.4100, 0.7137, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633],
  21086. [0.6115, 0.4081, 0.6725, 0.2433, 0.4088, 0.1933, 0.5167, 0.5544]],
  21087. device='cuda:0', dtype=torch.float64)
  21088. predictions are: tensor([[0.6591, 0.4333, 0.8948, 0.4122, 0.3788, 0.4922, 0.6208, 0.5571],
  21089. [0.6427, 0.4237, 0.9069, 0.5538, 0.4507, 0.5241, 0.5406, 0.5316],
  21090. [0.6362, 0.4339, 0.8928, 0.5974, 0.4273, 0.4166, 0.5866, 0.6091],
  21091. [0.6376, 0.4100, 0.7270, 0.2893, 0.3867, 0.3177, 0.5819, 0.5739],
  21092. [0.6525, 0.4261, 0.8762, 0.2520, 0.5085, 0.1691, 0.6501, 0.5437],
  21093. [0.6357, 0.4092, 0.8949, 0.6106, 0.3867, 0.4622, 0.6421, 0.5256],
  21094. [0.6220, 0.4065, 0.7542, 0.2496, 0.4231, 0.2386, 0.6143, 0.5739],
  21095. [0.6334, 0.4142, 0.7204, 0.2827, 0.4284, 0.2035, 0.5362, 0.5713]],
  21096. device='cuda:0', grad_fn=<AddmmBackward>)
  21097. landmarks are: tensor([[[0.6102, 0.4020, 0.8637, 0.3717, 0.3625, 0.5017, 0.6037, 0.5500],
  21098. [0.6097, 0.3988, 0.8650, 0.5250, 0.4212, 0.5200, 0.5675, 0.5050],
  21099. [0.6212, 0.4159, 0.8675, 0.5783, 0.4087, 0.4317, 0.5612, 0.5917],
  21100. [0.6152, 0.4131, 0.6862, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  21101. [0.6266, 0.4065, 0.8313, 0.2467, 0.4787, 0.1733, 0.6313, 0.5133],
  21102. [0.6317, 0.4038, 0.8288, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  21103. [0.6273, 0.4099, 0.7138, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633],
  21104. [0.6115, 0.4081, 0.6725, 0.2433, 0.4087, 0.1933, 0.5167, 0.5544]]],
  21105. device='cuda:0')
  21106. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  21107. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  21108. loss_train: 0.11819850071333349
  21109. step: 75
  21110. running loss: 0.001575980009511113
  21111. Train Steps: 75/90 Loss: 0.0016 torch.Size([8, 600, 800])
  21112. torch.Size([8, 8])
  21113. tensor([[0.6200, 0.3978, 0.8900, 0.4550, 0.3775, 0.5200, 0.6150, 0.5367],
  21114. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  21115. [0.6201, 0.3970, 0.8413, 0.4950, 0.4413, 0.5183, 0.6088, 0.5400],
  21116. [0.6175, 0.4091, 0.7863, 0.2800, 0.3638, 0.3583, 0.6188, 0.5433],
  21117. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  21118. [0.6249, 0.4142, 0.8350, 0.3283, 0.3613, 0.3700, 0.6188, 0.5400],
  21119. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  21120. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200]],
  21121. device='cuda:0', dtype=torch.float64)
  21122. predictions are: tensor([[0.6172, 0.3995, 0.8901, 0.5058, 0.3956, 0.4915, 0.6131, 0.5523],
  21123. [0.6882, 0.4489, 0.8780, 0.4083, 0.3668, 0.3666, 0.5890, 0.5351],
  21124. [0.5328, 0.3744, 0.8810, 0.5341, 0.4609, 0.4898, 0.5960, 0.5529],
  21125. [0.7211, 0.4788, 0.8036, 0.3136, 0.3870, 0.3325, 0.6272, 0.5799],
  21126. [0.6880, 0.4580, 0.8173, 0.2626, 0.4443, 0.2419, 0.6816, 0.5460],
  21127. [0.6753, 0.4548, 0.8753, 0.3632, 0.3737, 0.2981, 0.6539, 0.5604],
  21128. [0.6129, 0.3984, 0.9226, 0.4563, 0.3824, 0.3320, 0.6079, 0.5717],
  21129. [0.6587, 0.4387, 0.9108, 0.5261, 0.4108, 0.4428, 0.5994, 0.5486]],
  21130. device='cuda:0', grad_fn=<AddmmBackward>)
  21131. landmarks are: tensor([[[0.6199, 0.3978, 0.8900, 0.4550, 0.3775, 0.5200, 0.6150, 0.5367],
  21132. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  21133. [0.6201, 0.3970, 0.8413, 0.4950, 0.4412, 0.5183, 0.6087, 0.5400],
  21134. [0.6175, 0.4091, 0.7862, 0.2800, 0.3638, 0.3583, 0.6187, 0.5433],
  21135. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  21136. [0.6249, 0.4142, 0.8350, 0.3283, 0.3613, 0.3700, 0.6187, 0.5400],
  21137. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  21138. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200]]],
  21139. device='cuda:0')
  21140. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  21141. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  21142. loss_train: 0.11945557687431574
  21143. step: 76
  21144. running loss: 0.0015717839062409965
  21145.  
  21146. Train Steps: 76/90 Loss: 0.0016 torch.Size([8, 600, 800])
  21147. torch.Size([8, 8])
  21148. tensor([[0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  21149. [0.6197, 0.4118, 0.8688, 0.5517, 0.4037, 0.5233, 0.5875, 0.5600],
  21150. [0.6111, 0.4033, 0.8300, 0.3267, 0.3588, 0.3333, 0.5444, 0.5637],
  21151. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050],
  21152. [0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050],
  21153. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116],
  21154. [0.6265, 0.4088, 0.8025, 0.1850, 0.4163, 0.2500, 0.6290, 0.4947],
  21155. [0.6200, 0.4098, 0.8237, 0.2917, 0.4012, 0.2967, 0.6000, 0.5683]],
  21156. device='cuda:0', dtype=torch.float64)
  21157. predictions are: tensor([[0.6638, 0.4265, 0.8378, 0.3199, 0.3502, 0.3498, 0.6437, 0.5390],
  21158. [0.6458, 0.4382, 0.8623, 0.5751, 0.4099, 0.4719, 0.6228, 0.5623],
  21159. [0.6353, 0.4273, 0.8509, 0.3646, 0.3605, 0.2750, 0.5698, 0.5699],
  21160. [0.6264, 0.4253, 0.8758, 0.5270, 0.4639, 0.4688, 0.6128, 0.5361],
  21161. [0.6012, 0.4146, 0.8752, 0.4587, 0.3832, 0.4310, 0.6454, 0.6015],
  21162. [0.6165, 0.4263, 0.8658, 0.4376, 0.4099, 0.5349, 0.6132, 0.5368],
  21163. [0.6423, 0.4167, 0.8071, 0.2262, 0.4394, 0.1830, 0.6501, 0.5213],
  21164. [0.6808, 0.4402, 0.8347, 0.3076, 0.4046, 0.2507, 0.6324, 0.5565]],
  21165. device='cuda:0', grad_fn=<AddmmBackward>)
  21166. landmarks are: tensor([[[0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  21167. [0.6197, 0.4118, 0.8687, 0.5517, 0.4038, 0.5233, 0.5875, 0.5600],
  21168. [0.6111, 0.4033, 0.8300, 0.3267, 0.3587, 0.3333, 0.5444, 0.5637],
  21169. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050],
  21170. [0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050],
  21171. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116],
  21172. [0.6265, 0.4088, 0.8025, 0.1850, 0.4162, 0.2500, 0.6290, 0.4947],
  21173. [0.6200, 0.4098, 0.8238, 0.2917, 0.4013, 0.2967, 0.6000, 0.5683]]],
  21174. device='cuda:0')
  21175. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  21176. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  21177. loss_train: 0.12034609029069543
  21178. step: 77
  21179. running loss: 0.001562936237541499
  21180. Train Steps: 77/90 Loss: 0.0016 torch.Size([8, 600, 800])
  21181. torch.Size([8, 8])
  21182. tensor([[0.6259, 0.4156, 0.8812, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960],
  21183. [0.6185, 0.4067, 0.8838, 0.4450, 0.4037, 0.4733, 0.5213, 0.5142],
  21184. [0.6276, 0.4002, 0.8800, 0.5533, 0.3575, 0.4400, 0.6132, 0.4672],
  21185. [ nan, nan, 0.8300, 0.3150, 0.3588, 0.3383, 0.5208, 0.5194],
  21186. [0.6161, 0.4055, 0.8675, 0.3867, 0.3713, 0.4033, 0.5195, 0.5162],
  21187. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  21188. [0.6112, 0.4029, 0.8638, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  21189. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517]],
  21190. device='cuda:0', dtype=torch.float64)
  21191. predictions are: tensor([[0.6829, 0.4321, 0.8792, 0.2930, 0.4530, 0.1758, 0.6499, 0.5164],
  21192. [0.6334, 0.4151, 0.8681, 0.4462, 0.3797, 0.4307, 0.5416, 0.5134],
  21193. [0.7628, 0.4811, 0.8505, 0.5250, 0.3577, 0.4349, 0.6527, 0.5248],
  21194. [0.2527, 0.1632, 0.8017, 0.3036, 0.3434, 0.2813, 0.5369, 0.5235],
  21195. [0.5974, 0.3909, 0.8661, 0.3871, 0.3583, 0.3448, 0.5662, 0.5360],
  21196. [0.6522, 0.4458, 0.8638, 0.3950, 0.3395, 0.3446, 0.5653, 0.5265],
  21197. [0.6673, 0.4463, 0.8625, 0.4778, 0.4617, 0.4634, 0.6032, 0.5708],
  21198. [0.6972, 0.4597, 0.8791, 0.4854, 0.4432, 0.5452, 0.6703, 0.5676]],
  21199. device='cuda:0', grad_fn=<AddmmBackward>)
  21200. landmarks are: tensor([[[0.6259, 0.4156, 0.8813, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960],
  21201. [0.6185, 0.4067, 0.8838, 0.4450, 0.4038, 0.4733, 0.5213, 0.5142],
  21202. [0.6276, 0.4002, 0.8800, 0.5533, 0.3575, 0.4400, 0.6132, 0.4672],
  21203. [0.0000, 0.0000, 0.8300, 0.3150, 0.3587, 0.3383, 0.5208, 0.5194],
  21204. [0.6161, 0.4055, 0.8675, 0.3867, 0.3713, 0.4033, 0.5195, 0.5162],
  21205. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  21206. [0.6112, 0.4029, 0.8637, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  21207. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517]]],
  21208. device='cuda:0')
  21209. loss_train_step before backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  21210. loss_train_step after backward: tensor(0.0027, device='cuda:0', grad_fn=<MseLossBackward>)
  21211. loss_train: 0.12301964685320854
  21212. step: 78
  21213. running loss: 0.0015771749596565198
  21214. Train Steps: 78/90 Loss: 0.0016 torch.Size([8, 600, 800])
  21215. torch.Size([8, 8])
  21216. tensor([[0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  21217. [0.6128, 0.4022, 0.8738, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  21218. [0.6257, 0.4024, 0.8672, 0.5422, 0.4196, 0.5198, 0.6694, 0.5204],
  21219. [0.6124, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  21220. [0.6132, 0.4037, 0.6963, 0.2217, 0.4100, 0.1950, 0.5395, 0.5175],
  21221. [0.6250, 0.3993, 0.9138, 0.4333, 0.3763, 0.5217, 0.6995, 0.5320],
  21222. [ nan, nan, 0.6688, 0.2513, 0.4113, 0.2117, 0.5193, 0.5933],
  21223. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395]],
  21224. device='cuda:0', dtype=torch.float64)
  21225. predictions are: tensor([[0.6981, 0.4461, 0.7920, 0.1761, 0.4232, 0.2350, 0.6595, 0.5030],
  21226. [0.6089, 0.4098, 0.8837, 0.4997, 0.4948, 0.5058, 0.5205, 0.5163],
  21227. [0.6776, 0.4383, 0.8625, 0.5327, 0.4152, 0.5352, 0.6680, 0.5216],
  21228. [0.6112, 0.3980, 0.8476, 0.4931, 0.3843, 0.4628, 0.5391, 0.5513],
  21229. [0.5809, 0.3944, 0.7159, 0.2180, 0.3969, 0.1966, 0.5366, 0.4958],
  21230. [0.7036, 0.4519, 0.9253, 0.4453, 0.3807, 0.5351, 0.7349, 0.5451],
  21231. [0.0905, 0.0750, 0.6754, 0.2421, 0.3957, 0.2056, 0.5225, 0.5518],
  21232. [0.6353, 0.4183, 0.8821, 0.5299, 0.3436, 0.3670, 0.5775, 0.5352]],
  21233. device='cuda:0', grad_fn=<AddmmBackward>)
  21234. landmarks are: tensor([[[0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  21235. [0.6128, 0.4022, 0.8737, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  21236. [0.6257, 0.4024, 0.8672, 0.5422, 0.4196, 0.5198, 0.6694, 0.5204],
  21237. [0.6123, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  21238. [0.6132, 0.4037, 0.6963, 0.2217, 0.4100, 0.1950, 0.5395, 0.5175],
  21239. [0.6250, 0.3993, 0.9137, 0.4333, 0.3762, 0.5217, 0.6995, 0.5320],
  21240. [0.0000, 0.0000, 0.6688, 0.2513, 0.4112, 0.2117, 0.5193, 0.5933],
  21241. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395]]],
  21242. device='cuda:0')
  21243. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  21244. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  21245. loss_train: 0.12372523982776329
  21246. step: 79
  21247. running loss: 0.0015661422763008012
  21248. Train Steps: 79/90 Loss: 0.0016 torch.Size([8, 600, 800])
  21249. torch.Size([8, 8])
  21250. tensor([[0.6250, 0.4106, 0.8700, 0.3717, 0.3588, 0.4967, 0.6038, 0.5167],
  21251. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5963, 0.6217],
  21252. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  21253. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  21254. [0.6346, 0.4086, 0.7938, 0.5500, 0.3962, 0.4867, 0.7343, 0.5702],
  21255. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  21256. [0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500],
  21257. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459]],
  21258. device='cuda:0', dtype=torch.float64)
  21259. predictions are: tensor([[0.6288, 0.4135, 0.8620, 0.3507, 0.3612, 0.5392, 0.5979, 0.4968],
  21260. [0.5228, 0.3384, 0.7171, 0.2279, 0.4291, 0.2401, 0.5491, 0.5494],
  21261. [0.5521, 0.3427, 0.9261, 0.4161, 0.4091, 0.3444, 0.6386, 0.4978],
  21262. [0.5937, 0.3695, 0.8879, 0.5150, 0.3578, 0.4873, 0.5942, 0.4688],
  21263. [0.6119, 0.3992, 0.8160, 0.5159, 0.3865, 0.5166, 0.6645, 0.5292],
  21264. [0.5653, 0.3847, 0.8176, 0.5684, 0.3951, 0.4822, 0.5524, 0.5696],
  21265. [0.5591, 0.3564, 0.7779, 0.2521, 0.3498, 0.4319, 0.5708, 0.5020],
  21266. [0.5886, 0.3754, 0.8601, 0.2555, 0.4770, 0.2343, 0.6406, 0.5053]],
  21267. device='cuda:0', grad_fn=<AddmmBackward>)
  21268. landmarks are: tensor([[[0.6250, 0.4105, 0.8700, 0.3717, 0.3587, 0.4967, 0.6037, 0.5167],
  21269. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5962, 0.6217],
  21270. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  21271. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  21272. [0.6346, 0.4086, 0.7937, 0.5500, 0.3963, 0.4867, 0.7343, 0.5702],
  21273. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  21274. [0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500],
  21275. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459]]],
  21276. device='cuda:0')
  21277. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  21278. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  21279. loss_train: 0.12507811159593984
  21280. step: 80
  21281. running loss: 0.001563476394949248
  21282.  
  21283. Train Steps: 80/90 Loss: 0.0016 torch.Size([8, 600, 800])
  21284. torch.Size([8, 8])
  21285. tensor([[0.6277, 0.4103, 0.8087, 0.5717, 0.4188, 0.4750, 0.5663, 0.6083],
  21286. [0.6227, 0.4083, 0.8938, 0.4800, 0.3800, 0.2950, 0.5737, 0.5350],
  21287. [0.6218, 0.4185, 0.7338, 0.2650, 0.4625, 0.1950, 0.5687, 0.5800],
  21288. [0.6085, 0.4008, 0.8588, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  21289. [0.6198, 0.4101, 0.8838, 0.5283, 0.3763, 0.5267, 0.5913, 0.5567],
  21290. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  21291. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  21292. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117]],
  21293. device='cuda:0', dtype=torch.float64)
  21294. predictions are: tensor([[0.6061, 0.3938, 0.8173, 0.5448, 0.4088, 0.4903, 0.6078, 0.5864],
  21295. [0.6013, 0.3893, 0.8713, 0.4460, 0.3765, 0.3447, 0.5802, 0.5218],
  21296. [0.5587, 0.3764, 0.7229, 0.2238, 0.4488, 0.2153, 0.5614, 0.5439],
  21297. [0.5299, 0.3311, 0.8490, 0.4794, 0.4807, 0.5188, 0.5202, 0.4970],
  21298. [0.5694, 0.3743, 0.8593, 0.5075, 0.3752, 0.5695, 0.5847, 0.5236],
  21299. [0.5815, 0.3632, 0.8861, 0.4526, 0.3648, 0.5317, 0.6349, 0.4731],
  21300. [0.5626, 0.3566, 0.8745, 0.3104, 0.3800, 0.3518, 0.5987, 0.5117],
  21301. [0.5221, 0.3276, 0.8707, 0.3038, 0.3811, 0.3485, 0.5562, 0.4916]],
  21302. device='cuda:0', grad_fn=<AddmmBackward>)
  21303. landmarks are: tensor([[[0.6277, 0.4103, 0.8087, 0.5717, 0.4187, 0.4750, 0.5663, 0.6083],
  21304. [0.6227, 0.4083, 0.8938, 0.4800, 0.3800, 0.2950, 0.5738, 0.5350],
  21305. [0.6218, 0.4185, 0.7337, 0.2650, 0.4625, 0.1950, 0.5688, 0.5800],
  21306. [0.6084, 0.4008, 0.8587, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  21307. [0.6198, 0.4101, 0.8838, 0.5283, 0.3762, 0.5267, 0.5913, 0.5567],
  21308. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  21309. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  21310. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117]]],
  21311. device='cuda:0')
  21312. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  21313. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  21314. loss_train: 0.12628521776059642
  21315. step: 81
  21316. running loss: 0.001559076762476499
  21317. Train Steps: 81/90 Loss: 0.0016 torch.Size([8, 600, 800])
  21318. torch.Size([8, 8])
  21319. tensor([[0.6109, 0.4041, 0.6975, 0.3167, 0.3513, 0.3383, 0.5153, 0.5319],
  21320. [0.6083, 0.3957, 0.8638, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  21321. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467],
  21322. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  21323. [0.6132, 0.4118, 0.8200, 0.3633, 0.3563, 0.5400, 0.5787, 0.5136],
  21324. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650],
  21325. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  21326. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609]],
  21327. device='cuda:0', dtype=torch.float64)
  21328. predictions are: tensor([[0.5418, 0.3460, 0.7083, 0.3052, 0.3531, 0.3603, 0.5018, 0.5154],
  21329. [0.5441, 0.3577, 0.8451, 0.4836, 0.4316, 0.5414, 0.5323, 0.4998],
  21330. [0.5531, 0.3617, 0.8243, 0.3293, 0.3425, 0.4509, 0.5539, 0.5333],
  21331. [0.5925, 0.3804, 0.8683, 0.4660, 0.3918, 0.4476, 0.5878, 0.5417],
  21332. [0.6002, 0.4011, 0.8016, 0.3649, 0.3546, 0.5386, 0.5675, 0.5146],
  21333. [0.5849, 0.4086, 0.8510, 0.4163, 0.3650, 0.3590, 0.5596, 0.5523],
  21334. [0.5821, 0.3788, 0.8625, 0.5083, 0.4160, 0.5501, 0.5990, 0.5181],
  21335. [0.4306, 0.2809, 0.8359, 0.2389, 0.5374, 0.2232, 0.7439, 0.5276]],
  21336. device='cuda:0', grad_fn=<AddmmBackward>)
  21337. landmarks are: tensor([[[0.6109, 0.4041, 0.6975, 0.3167, 0.3512, 0.3383, 0.5153, 0.5319],
  21338. [0.6083, 0.3957, 0.8637, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  21339. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467],
  21340. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  21341. [0.6132, 0.4118, 0.8200, 0.3633, 0.3562, 0.5400, 0.5787, 0.5136],
  21342. [0.6225, 0.4191, 0.8500, 0.4167, 0.3688, 0.3233, 0.5650, 0.5650],
  21343. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  21344. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609]]],
  21345. device='cuda:0')
  21346. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  21347. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  21348. loss_train: 0.12774768978124484
  21349. step: 82
  21350. running loss: 0.0015578986558688395
  21351. Train Steps: 82/90 Loss: 0.0016 torch.Size([8, 600, 800])
  21352. torch.Size([8, 8])
  21353. tensor([[0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  21354. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  21355. [0.6197, 0.3986, 0.8800, 0.4617, 0.4188, 0.4783, 0.5687, 0.5550],
  21356. [0.6222, 0.3957, 0.8838, 0.5017, 0.3937, 0.4600, 0.5900, 0.5017],
  21357. [ nan, nan, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  21358. [0.6219, 0.3934, 0.8688, 0.5267, 0.4313, 0.4967, 0.5988, 0.4983],
  21359. [0.6277, 0.4118, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  21360. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650]],
  21361. device='cuda:0', dtype=torch.float64)
  21362. predictions are: tensor([[ 0.6675, 0.4397, 0.8381, 0.5552, 0.3738, 0.5097, 0.6472, 0.5416],
  21363. [ 0.5951, 0.3972, 0.8532, 0.4696, 0.4206, 0.5650, 0.6118, 0.5556],
  21364. [ 0.6322, 0.4160, 0.8588, 0.4564, 0.4137, 0.5005, 0.5463, 0.5500],
  21365. [ 0.6032, 0.3870, 0.8658, 0.4831, 0.3925, 0.4840, 0.5734, 0.5137],
  21366. [-0.0529, -0.0354, 0.7400, 0.2637, 0.3871, 0.2772, 0.4593, 0.5618],
  21367. [ 0.6284, 0.4051, 0.8570, 0.5077, 0.4325, 0.5167, 0.5824, 0.5051],
  21368. [ 0.6477, 0.4342, 0.8797, 0.3638, 0.3833, 0.2962, 0.6136, 0.5325],
  21369. [ 0.6533, 0.4331, 0.8344, 0.5550, 0.4074, 0.4584, 0.5385, 0.5828]],
  21370. device='cuda:0', grad_fn=<AddmmBackward>)
  21371. landmarks are: tensor([[[0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  21372. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  21373. [0.6197, 0.3986, 0.8800, 0.4617, 0.4187, 0.4783, 0.5688, 0.5550],
  21374. [0.6222, 0.3957, 0.8838, 0.5017, 0.3938, 0.4600, 0.5900, 0.5017],
  21375. [0.0000, 0.0000, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  21376. [0.6219, 0.3934, 0.8687, 0.5267, 0.4313, 0.4967, 0.5987, 0.4983],
  21377. [0.6277, 0.4117, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  21378. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650]]],
  21379. device='cuda:0')
  21380. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  21381. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  21382. loss_train: 0.1281739636324346
  21383. step: 83
  21384. running loss: 0.0015442646220775254
  21385. Train Steps: 83/90 Loss: 0.0015 torch.Size([8, 600, 800])
  21386. torch.Size([8, 8])
  21387. tensor([[0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  21388. [0.6164, 0.4119, 0.7913, 0.2650, 0.3538, 0.3500, 0.5614, 0.5038],
  21389. [0.6127, 0.4118, 0.8650, 0.5083, 0.4088, 0.5367, 0.5300, 0.5456],
  21390. [0.6231, 0.3973, 0.8650, 0.3950, 0.3625, 0.3183, 0.5837, 0.5167],
  21391. [0.6277, 0.4083, 0.8350, 0.2717, 0.4562, 0.1800, 0.5918, 0.4878],
  21392. [0.6199, 0.4065, 0.7598, 0.2385, 0.4317, 0.1981, 0.5933, 0.5221],
  21393. [0.6186, 0.4060, 0.8750, 0.5050, 0.3538, 0.4367, 0.5813, 0.6083],
  21394. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378]],
  21395. device='cuda:0', dtype=torch.float64)
  21396. predictions are: tensor([[0.6597, 0.4264, 0.8902, 0.4278, 0.3658, 0.4644, 0.6411, 0.5222],
  21397. [0.5860, 0.3856, 0.7773, 0.2711, 0.3504, 0.3750, 0.5729, 0.5375],
  21398. [0.5656, 0.3925, 0.8694, 0.5230, 0.4260, 0.5606, 0.5162, 0.5660],
  21399. [0.6103, 0.3844, 0.8472, 0.4252, 0.3782, 0.3238, 0.5537, 0.5528],
  21400. [0.5067, 0.3293, 0.8092, 0.2659, 0.4708, 0.2124, 0.5685, 0.5118],
  21401. [0.5839, 0.3781, 0.7251, 0.2463, 0.4241, 0.2149, 0.5706, 0.5286],
  21402. [0.5335, 0.3611, 0.8560, 0.5066, 0.3550, 0.4511, 0.5697, 0.6224],
  21403. [0.5556, 0.3549, 0.8647, 0.5021, 0.4329, 0.5528, 0.6810, 0.5749]],
  21404. device='cuda:0', grad_fn=<AddmmBackward>)
  21405. landmarks are: tensor([[[0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  21406. [0.6164, 0.4119, 0.7912, 0.2650, 0.3537, 0.3500, 0.5614, 0.5038],
  21407. [0.6127, 0.4118, 0.8650, 0.5083, 0.4087, 0.5367, 0.5300, 0.5456],
  21408. [0.6231, 0.3973, 0.8650, 0.3950, 0.3625, 0.3183, 0.5838, 0.5167],
  21409. [0.6277, 0.4083, 0.8350, 0.2717, 0.4563, 0.1800, 0.5918, 0.4878],
  21410. [0.6199, 0.4065, 0.7598, 0.2385, 0.4317, 0.1981, 0.5933, 0.5221],
  21411. [0.6186, 0.4060, 0.8750, 0.5050, 0.3537, 0.4367, 0.5813, 0.6083],
  21412. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378]]],
  21413. device='cuda:0')
  21414. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  21415. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  21416. loss_train: 0.12917014537379146
  21417. step: 84
  21418. running loss: 0.0015377398258784698
  21419.  
  21420. Train Steps: 84/90 Loss: 0.0015 torch.Size([8, 600, 800])
  21421. torch.Size([8, 8])
  21422. tensor([[0.6250, 0.4054, 0.8770, 0.4723, 0.4662, 0.5367, 0.6162, 0.5433],
  21423. [0.6059, 0.4002, 0.7562, 0.2767, 0.3538, 0.3033, 0.5529, 0.5455],
  21424. [0.6145, 0.4007, 0.8775, 0.4533, 0.4562, 0.5533, 0.6088, 0.5533],
  21425. [0.6165, 0.4106, 0.7575, 0.1733, 0.3838, 0.2650, 0.5680, 0.5116],
  21426. [0.6250, 0.4013, 0.8525, 0.5417, 0.4037, 0.5117, 0.6325, 0.5017],
  21427. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5787, 0.5550],
  21428. [0.6202, 0.4053, 0.8638, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  21429. [0.6200, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391]],
  21430. device='cuda:0', dtype=torch.float64)
  21431. predictions are: tensor([[0.6153, 0.4149, 0.8889, 0.4847, 0.4558, 0.5046, 0.6244, 0.5498],
  21432. [0.5586, 0.3738, 0.7486, 0.2882, 0.3541, 0.2953, 0.5606, 0.5527],
  21433. [0.6257, 0.4160, 0.8654, 0.4678, 0.4573, 0.5368, 0.5980, 0.5739],
  21434. [0.5703, 0.3918, 0.7460, 0.1940, 0.3798, 0.2339, 0.5797, 0.5188],
  21435. [0.5688, 0.3682, 0.8581, 0.5712, 0.4034, 0.4855, 0.6434, 0.5377],
  21436. [0.6166, 0.4107, 0.8232, 0.3031, 0.3427, 0.3449, 0.5720, 0.5591],
  21437. [0.6142, 0.3968, 0.8740, 0.5402, 0.4438, 0.4831, 0.6002, 0.5364],
  21438. [0.5238, 0.3544, 0.8481, 0.3092, 0.4216, 0.2158, 0.5670, 0.5483]],
  21439. device='cuda:0', grad_fn=<AddmmBackward>)
  21440. landmarks are: tensor([[[0.6250, 0.4054, 0.8770, 0.4723, 0.4663, 0.5367, 0.6162, 0.5433],
  21441. [0.6059, 0.4002, 0.7563, 0.2767, 0.3537, 0.3033, 0.5529, 0.5455],
  21442. [0.6145, 0.4007, 0.8775, 0.4533, 0.4563, 0.5533, 0.6087, 0.5533],
  21443. [0.6165, 0.4106, 0.7575, 0.1733, 0.3837, 0.2650, 0.5680, 0.5116],
  21444. [0.6250, 0.4013, 0.8525, 0.5417, 0.4038, 0.5117, 0.6325, 0.5017],
  21445. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5788, 0.5550],
  21446. [0.6202, 0.4053, 0.8637, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  21447. [0.6199, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391]]],
  21448. device='cuda:0')
  21449. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  21450. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  21451. loss_train: 0.12972077063750476
  21452. step: 85
  21453. running loss: 0.001526126713382409
  21454. Train Steps: 85/90 Loss: 0.0015 torch.Size([8, 600, 800])
  21455. torch.Size([8, 8])
  21456. tensor([[0.6204, 0.4013, 0.8075, 0.2400, 0.4313, 0.2050, 0.5800, 0.5150],
  21457. [ nan, nan, 0.8463, 0.2550, 0.5850, 0.2133, 0.7129, 0.6072],
  21458. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  21459. [0.6250, 0.4236, 0.8638, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  21460. [0.6038, 0.3946, 0.8413, 0.4883, 0.3563, 0.4550, 0.5266, 0.4693],
  21461. [0.6161, 0.4055, 0.8675, 0.3867, 0.3713, 0.4033, 0.5195, 0.5162],
  21462. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  21463. [0.6199, 0.4102, 0.8950, 0.4417, 0.4012, 0.5367, 0.6112, 0.5967]],
  21464. device='cuda:0', dtype=torch.float64)
  21465. predictions are: tensor([[0.6324, 0.4024, 0.7867, 0.2706, 0.4302, 0.1842, 0.5873, 0.5167],
  21466. [0.1120, 0.0905, 0.8614, 0.2757, 0.5323, 0.2532, 0.7227, 0.5899],
  21467. [0.6636, 0.4510, 0.8866, 0.5082, 0.3449, 0.4067, 0.6281, 0.5082],
  21468. [0.6678, 0.4606, 0.8678, 0.3985, 0.3912, 0.2996, 0.5728, 0.5700],
  21469. [0.6000, 0.4066, 0.8471, 0.5015, 0.3561, 0.4322, 0.5428, 0.5055],
  21470. [0.6050, 0.4145, 0.8732, 0.3977, 0.3691, 0.3673, 0.5318, 0.5119],
  21471. [0.5998, 0.4105, 0.6665, 0.2674, 0.3806, 0.2248, 0.5654, 0.5679],
  21472. [0.6570, 0.4326, 0.8951, 0.4556, 0.3951, 0.5170, 0.6160, 0.5746]],
  21473. device='cuda:0', grad_fn=<AddmmBackward>)
  21474. landmarks are: tensor([[[0.6204, 0.4013, 0.8075, 0.2400, 0.4313, 0.2050, 0.5800, 0.5150],
  21475. [0.0000, 0.0000, 0.8462, 0.2550, 0.5850, 0.2133, 0.7129, 0.6072],
  21476. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  21477. [0.6250, 0.4236, 0.8637, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  21478. [0.6038, 0.3946, 0.8413, 0.4883, 0.3562, 0.4550, 0.5266, 0.4693],
  21479. [0.6161, 0.4055, 0.8675, 0.3867, 0.3713, 0.4033, 0.5195, 0.5162],
  21480. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  21481. [0.6199, 0.4102, 0.8950, 0.4417, 0.4013, 0.5367, 0.6112, 0.5967]]],
  21482. device='cuda:0')
  21483. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  21484. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  21485. loss_train: 0.13042904192116112
  21486. step: 86
  21487. running loss: 0.0015166167665251293
  21488. Train Steps: 86/90 Loss: 0.0015 torch.Size([8, 600, 800])
  21489. torch.Size([8, 8])
  21490. tensor([[0.6199, 0.4093, 0.7913, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700],
  21491. [0.6332, 0.4165, 0.9100, 0.3350, 0.4188, 0.3683, 0.7438, 0.5528],
  21492. [0.6216, 0.4099, 0.7225, 0.2033, 0.4188, 0.2217, 0.5975, 0.5283],
  21493. [0.6307, 0.4029, 0.8988, 0.4817, 0.3937, 0.3500, 0.7311, 0.5378],
  21494. [0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431],
  21495. [0.6200, 0.3993, 0.8519, 0.4923, 0.3962, 0.4717, 0.6013, 0.5433],
  21496. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  21497. [0.6060, 0.3924, 0.8450, 0.5717, 0.4200, 0.5217, 0.5253, 0.4752]],
  21498. device='cuda:0', dtype=torch.float64)
  21499. predictions are: tensor([[0.5883, 0.3915, 0.8001, 0.2407, 0.4200, 0.2433, 0.5929, 0.5700],
  21500. [0.6419, 0.4123, 0.9253, 0.3650, 0.4035, 0.3305, 0.7279, 0.5396],
  21501. [0.6671, 0.4461, 0.7152, 0.1981, 0.4120, 0.1980, 0.5596, 0.5310],
  21502. [0.6116, 0.4015, 0.9169, 0.4768, 0.3858, 0.3125, 0.7088, 0.5173],
  21503. [0.6014, 0.3990, 0.8853, 0.4906, 0.4388, 0.4627, 0.5007, 0.5111],
  21504. [0.5797, 0.3709, 0.8777, 0.5019, 0.3913, 0.4391, 0.5890, 0.5408],
  21505. [0.6075, 0.3936, 0.8572, 0.5368, 0.3973, 0.5263, 0.7019, 0.5504],
  21506. [0.6125, 0.3837, 0.8629, 0.5984, 0.4356, 0.4845, 0.5144, 0.4990]],
  21507. device='cuda:0', grad_fn=<AddmmBackward>)
  21508. landmarks are: tensor([[[0.6198, 0.4093, 0.7912, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700],
  21509. [0.6332, 0.4165, 0.9100, 0.3350, 0.4187, 0.3683, 0.7438, 0.5528],
  21510. [0.6216, 0.4099, 0.7225, 0.2033, 0.4187, 0.2217, 0.5975, 0.5283],
  21511. [0.6307, 0.4029, 0.8988, 0.4817, 0.3938, 0.3500, 0.7311, 0.5378],
  21512. [0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431],
  21513. [0.6200, 0.3993, 0.8519, 0.4923, 0.3963, 0.4717, 0.6012, 0.5433],
  21514. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  21515. [0.6060, 0.3924, 0.8450, 0.5717, 0.4200, 0.5217, 0.5253, 0.4752]]],
  21516. device='cuda:0')
  21517. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  21518. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  21519. loss_train: 0.13084795890608802
  21520. step: 87
  21521. running loss: 0.0015039995276561842
  21522. Train Steps: 87/90 Loss: 0.0015 torch.Size([8, 600, 800])
  21523. torch.Size([8, 8])
  21524. tensor([[0.6218, 0.4098, 0.7238, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  21525. [0.6135, 0.3994, 0.7913, 0.3050, 0.3625, 0.3050, 0.5837, 0.5050],
  21526. [0.6284, 0.4093, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  21527. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  21528. [0.6064, 0.3953, 0.8738, 0.4417, 0.3663, 0.4683, 0.5511, 0.5416],
  21529. [0.6202, 0.4053, 0.8638, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  21530. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  21531. [0.6276, 0.4235, 0.8888, 0.5333, 0.3800, 0.3117, 0.5427, 0.6164]],
  21532. device='cuda:0', dtype=torch.float64)
  21533. predictions are: tensor([[0.6102, 0.4038, 0.7431, 0.1998, 0.4382, 0.2402, 0.6259, 0.5483],
  21534. [0.6186, 0.3994, 0.7977, 0.3072, 0.3726, 0.3181, 0.5868, 0.5125],
  21535. [0.5738, 0.3796, 0.9057, 0.4642, 0.3647, 0.3801, 0.6272, 0.5086],
  21536. [0.6229, 0.4183, 0.8687, 0.2433, 0.5416, 0.2083, 0.7472, 0.5412],
  21537. [0.5756, 0.3719, 0.8908, 0.4649, 0.3554, 0.4701, 0.5623, 0.5266],
  21538. [0.6130, 0.3902, 0.8879, 0.5309, 0.4536, 0.5069, 0.5984, 0.5160],
  21539. [0.6295, 0.4091, 0.8896, 0.2798, 0.4066, 0.2617, 0.6012, 0.5375],
  21540. [0.6980, 0.4612, 0.8706, 0.5554, 0.3980, 0.3224, 0.5773, 0.5992]],
  21541. device='cuda:0', grad_fn=<AddmmBackward>)
  21542. landmarks are: tensor([[[0.6218, 0.4098, 0.7237, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  21543. [0.6135, 0.3994, 0.7912, 0.3050, 0.3625, 0.3050, 0.5838, 0.5050],
  21544. [0.6284, 0.4092, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  21545. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  21546. [0.6064, 0.3952, 0.8737, 0.4417, 0.3663, 0.4683, 0.5511, 0.5416],
  21547. [0.6202, 0.4053, 0.8637, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  21548. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  21549. [0.6276, 0.4235, 0.8888, 0.5333, 0.3800, 0.3117, 0.5427, 0.6164]]],
  21550. device='cuda:0')
  21551. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  21552. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  21553. loss_train: 0.1311794207431376
  21554. step: 88
  21555. running loss: 0.0014906752357174728
  21556.  
  21557. Train Steps: 88/90 Loss: 0.0015 torch.Size([8, 600, 800])
  21558. torch.Size([8, 8])
  21559. tensor([[0.6274, 0.4270, 0.8938, 0.4967, 0.3550, 0.4283, 0.5700, 0.5733],
  21560. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  21561. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  21562. [0.6163, 0.4006, 0.8788, 0.4683, 0.3663, 0.4883, 0.5887, 0.5017],
  21563. [0.6228, 0.4119, 0.7938, 0.2233, 0.4674, 0.1773, 0.6188, 0.5433],
  21564. [0.6207, 0.4110, 0.8738, 0.5000, 0.4800, 0.5633, 0.6300, 0.5433],
  21565. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  21566. [0.6030, 0.3969, 0.7988, 0.3917, 0.3450, 0.3667, 0.5266, 0.4700]],
  21567. device='cuda:0', dtype=torch.float64)
  21568. predictions are: tensor([[0.6583, 0.4411, 0.8900, 0.4753, 0.3661, 0.4107, 0.6070, 0.5610],
  21569. [0.7032, 0.4607, 0.8342, 0.5524, 0.4124, 0.4293, 0.6039, 0.6208],
  21570. [0.6121, 0.3993, 0.7666, 0.2818, 0.3541, 0.3070, 0.5600, 0.5471],
  21571. [0.6439, 0.4244, 0.8729, 0.4387, 0.3880, 0.4663, 0.5833, 0.5265],
  21572. [0.7073, 0.4632, 0.8083, 0.1969, 0.4801, 0.1462, 0.6399, 0.5277],
  21573. [0.6096, 0.3859, 0.9005, 0.4553, 0.4819, 0.5500, 0.6491, 0.5414],
  21574. [0.6181, 0.4030, 0.9020, 0.4781, 0.4361, 0.4963, 0.6299, 0.5102],
  21575. [0.6141, 0.4031, 0.8295, 0.3727, 0.3551, 0.3516, 0.5576, 0.4941]],
  21576. device='cuda:0', grad_fn=<AddmmBackward>)
  21577. landmarks are: tensor([[[0.6274, 0.4270, 0.8938, 0.4967, 0.3550, 0.4283, 0.5700, 0.5733],
  21578. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  21579. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  21580. [0.6163, 0.4006, 0.8788, 0.4683, 0.3663, 0.4883, 0.5888, 0.5017],
  21581. [0.6228, 0.4119, 0.7937, 0.2233, 0.4674, 0.1773, 0.6187, 0.5433],
  21582. [0.6207, 0.4110, 0.8737, 0.5000, 0.4800, 0.5633, 0.6300, 0.5433],
  21583. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  21584. [0.6030, 0.3969, 0.7987, 0.3917, 0.3450, 0.3667, 0.5266, 0.4700]]],
  21585. device='cuda:0')
  21586. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  21587. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  21588. loss_train: 0.13189973769476637
  21589. step: 89
  21590. running loss: 0.0014820195246602963
  21591. Train Steps: 89/90 Loss: 0.0015 torch.Size([8, 600, 800])
  21592. torch.Size([8, 8])
  21593. tensor([[0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  21594. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  21595. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667],
  21596. [0.6147, 0.4026, 0.6600, 0.2467, 0.4088, 0.2150, 0.5489, 0.5773],
  21597. [0.6357, 0.4118, 0.8400, 0.2500, 0.5413, 0.1633, 0.6725, 0.5586],
  21598. [0.6115, 0.3998, 0.7063, 0.2383, 0.4037, 0.1950, 0.5320, 0.4993],
  21599. [0.6261, 0.4045, 0.8865, 0.5369, 0.3895, 0.4859, 0.6683, 0.5249],
  21600. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717]],
  21601. device='cuda:0', dtype=torch.float64)
  21602. predictions are: tensor([[0.6113, 0.4123, 0.9156, 0.4842, 0.3787, 0.4339, 0.6803, 0.5313],
  21603. [0.5968, 0.3982, 0.8830, 0.4667, 0.3824, 0.5372, 0.5866, 0.4975],
  21604. [0.6615, 0.4335, 0.8602, 0.3215, 0.3607, 0.3666, 0.5859, 0.5735],
  21605. [0.6616, 0.4508, 0.6859, 0.2386, 0.4228, 0.2136, 0.5463, 0.5839],
  21606. [0.6522, 0.4216, 0.8659, 0.2377, 0.5468, 0.1720, 0.6670, 0.5417],
  21607. [0.6198, 0.4215, 0.7188, 0.2078, 0.4245, 0.2021, 0.5389, 0.5014],
  21608. [0.5990, 0.3871, 0.9110, 0.5404, 0.3889, 0.4972, 0.6665, 0.5114],
  21609. [0.5953, 0.4015, 0.6792, 0.2568, 0.4020, 0.2465, 0.5699, 0.5747]],
  21610. device='cuda:0', grad_fn=<AddmmBackward>)
  21611. landmarks are: tensor([[[0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  21612. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  21613. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667],
  21614. [0.6147, 0.4026, 0.6600, 0.2467, 0.4087, 0.2150, 0.5489, 0.5773],
  21615. [0.6357, 0.4118, 0.8400, 0.2500, 0.5412, 0.1633, 0.6725, 0.5586],
  21616. [0.6115, 0.3998, 0.7063, 0.2383, 0.4038, 0.1950, 0.5320, 0.4993],
  21617. [0.6261, 0.4045, 0.8865, 0.5369, 0.3895, 0.4859, 0.6683, 0.5249],
  21618. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717]]],
  21619. device='cuda:0')
  21620. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  21621. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  21622. loss_train: 0.13218482348020189
  21623. step: 90
  21624. running loss: 0.001468720260891132
  21625. Valid Steps: 10/10 Loss: nan 15
  21626. --------------------------------------------------
  21627. Epoch: 7 Train Loss: 0.0015 Valid Loss: nan
  21628. --------------------------------------------------
  21629. size of train loader is: 90
  21630. torch.Size([8, 600, 800])
  21631. torch.Size([8, 8])
  21632. tensor([[0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  21633. [0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6262, 0.5167],
  21634. [0.6250, 0.4054, 0.8770, 0.4723, 0.4662, 0.5367, 0.6162, 0.5433],
  21635. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  21636. [0.6250, 0.4146, 0.8838, 0.3933, 0.3588, 0.4283, 0.6162, 0.5367],
  21637. [0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500],
  21638. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  21639. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283]],
  21640. device='cuda:0', dtype=torch.float64)
  21641. predictions are: tensor([[0.6952, 0.4387, 0.8725, 0.4735, 0.3666, 0.3972, 0.6239, 0.5173],
  21642. [0.6620, 0.4335, 0.8837, 0.4076, 0.3828, 0.3460, 0.6416, 0.5286],
  21643. [0.6551, 0.4304, 0.8760, 0.4881, 0.4687, 0.5299, 0.6150, 0.5283],
  21644. [0.6975, 0.4636, 0.8789, 0.4989, 0.3623, 0.4215, 0.6250, 0.5132],
  21645. [0.6783, 0.4559, 0.8491, 0.3904, 0.3644, 0.3979, 0.6092, 0.5457],
  21646. [0.6325, 0.4093, 0.7707, 0.2715, 0.3652, 0.4031, 0.6094, 0.5560],
  21647. [0.6950, 0.4435, 0.8780, 0.4944, 0.4598, 0.6014, 0.6045, 0.5163],
  21648. [0.6692, 0.4511, 0.8783, 0.4058, 0.3701, 0.3708, 0.5801, 0.5369]],
  21649. device='cuda:0', grad_fn=<AddmmBackward>)
  21650. landmarks are: tensor([[[0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  21651. [0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6263, 0.5167],
  21652. [0.6250, 0.4054, 0.8770, 0.4723, 0.4663, 0.5367, 0.6162, 0.5433],
  21653. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  21654. [0.6250, 0.4146, 0.8838, 0.3933, 0.3587, 0.4283, 0.6162, 0.5367],
  21655. [0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500],
  21656. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  21657. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283]]],
  21658. device='cuda:0')
  21659. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  21660. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  21661. loss_train: 0.000634737079963088
  21662. step: 1
  21663. running loss: 0.000634737079963088
  21664. Train Steps: 1/90 Loss: 0.0006 torch.Size([8, 600, 800])
  21665. torch.Size([8, 8])
  21666. tensor([[0.6286, 0.4086, 0.8408, 0.2801, 0.4163, 0.2800, 0.6725, 0.5393],
  21667. [0.6115, 0.3998, 0.7063, 0.2383, 0.4037, 0.1950, 0.5320, 0.4993],
  21668. [0.6261, 0.4066, 0.8325, 0.2150, 0.4763, 0.2667, 0.7002, 0.5633],
  21669. [0.6153, 0.4119, 0.8463, 0.3833, 0.3600, 0.3200, 0.5106, 0.5563],
  21670. [0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050],
  21671. [0.6135, 0.4115, 0.8838, 0.4667, 0.4288, 0.6050, 0.5778, 0.5097],
  21672. [0.6163, 0.4114, 0.7650, 0.2017, 0.3763, 0.2867, 0.5631, 0.5071],
  21673. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309]],
  21674. device='cuda:0', dtype=torch.float64)
  21675. predictions are: tensor([[0.6766, 0.4289, 0.8286, 0.2968, 0.4127, 0.2863, 0.6748, 0.5418],
  21676. [0.6179, 0.4140, 0.6912, 0.2371, 0.4145, 0.2128, 0.5311, 0.5148],
  21677. [0.6247, 0.4153, 0.8053, 0.2420, 0.4797, 0.2715, 0.7081, 0.5556],
  21678. [0.5787, 0.3854, 0.8454, 0.4129, 0.3846, 0.3143, 0.5029, 0.5668],
  21679. [0.6452, 0.4174, 0.8559, 0.4356, 0.3796, 0.5023, 0.6076, 0.5970],
  21680. [0.6454, 0.4280, 0.8659, 0.4704, 0.4320, 0.6108, 0.5659, 0.5241],
  21681. [0.5738, 0.3689, 0.7364, 0.2182, 0.3825, 0.2897, 0.5679, 0.5108],
  21682. [0.6694, 0.4415, 0.8888, 0.4933, 0.3867, 0.3707, 0.6565, 0.5217]],
  21683. device='cuda:0', grad_fn=<AddmmBackward>)
  21684. landmarks are: tensor([[[0.6286, 0.4086, 0.8408, 0.2801, 0.4162, 0.2800, 0.6725, 0.5393],
  21685. [0.6115, 0.3998, 0.7063, 0.2383, 0.4038, 0.1950, 0.5320, 0.4993],
  21686. [0.6261, 0.4066, 0.8325, 0.2150, 0.4762, 0.2667, 0.7002, 0.5633],
  21687. [0.6153, 0.4119, 0.8462, 0.3833, 0.3600, 0.3200, 0.5106, 0.5563],
  21688. [0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050],
  21689. [0.6135, 0.4115, 0.8838, 0.4667, 0.4288, 0.6050, 0.5778, 0.5097],
  21690. [0.6163, 0.4114, 0.7650, 0.2017, 0.3762, 0.2867, 0.5631, 0.5071],
  21691. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309]]],
  21692. device='cuda:0')
  21693. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  21694. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  21695. loss_train: 0.0009739936213009059
  21696. step: 2
  21697. running loss: 0.00048699681065045297
  21698.  
  21699. Train Steps: 2/90 Loss: 0.0005 torch.Size([8, 600, 800])
  21700. torch.Size([8, 8])
  21701. tensor([[0.6307, 0.4029, 0.8650, 0.5200, 0.3763, 0.4017, 0.7311, 0.5366],
  21702. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  21703. [0.6200, 0.4059, 0.8700, 0.4900, 0.4163, 0.5000, 0.6162, 0.5467],
  21704. [0.6131, 0.4037, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  21705. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6063, 0.6083],
  21706. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309],
  21707. [0.6118, 0.4052, 0.8463, 0.3917, 0.3538, 0.3450, 0.5053, 0.5593],
  21708. [0.6224, 0.4061, 0.8988, 0.4300, 0.3838, 0.4750, 0.6112, 0.5483]],
  21709. device='cuda:0', dtype=torch.float64)
  21710. predictions are: tensor([[0.6684, 0.4218, 0.8629, 0.5071, 0.3801, 0.4074, 0.7064, 0.5017],
  21711. [0.6440, 0.3897, 0.8907, 0.4888, 0.3580, 0.5210, 0.6386, 0.4876],
  21712. [0.6618, 0.4176, 0.8623, 0.4999, 0.4258, 0.5417, 0.5987, 0.5448],
  21713. [0.6504, 0.4116, 0.6684, 0.2818, 0.3705, 0.3098, 0.5226, 0.5757],
  21714. [0.6074, 0.4032, 0.8490, 0.3834, 0.3605, 0.4385, 0.6010, 0.6027],
  21715. [0.7068, 0.4541, 0.8974, 0.4816, 0.3890, 0.3849, 0.6512, 0.5143],
  21716. [0.6663, 0.4288, 0.8405, 0.3848, 0.3601, 0.3499, 0.4727, 0.5486],
  21717. [0.6665, 0.4204, 0.8850, 0.4391, 0.3886, 0.5139, 0.6090, 0.5330]],
  21718. device='cuda:0', grad_fn=<AddmmBackward>)
  21719. landmarks are: tensor([[[0.6307, 0.4029, 0.8650, 0.5200, 0.3762, 0.4017, 0.7311, 0.5366],
  21720. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  21721. [0.6199, 0.4059, 0.8700, 0.4900, 0.4162, 0.5000, 0.6162, 0.5467],
  21722. [0.6131, 0.4036, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  21723. [0.6199, 0.4112, 0.8475, 0.3717, 0.3550, 0.4350, 0.6062, 0.6083],
  21724. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309],
  21725. [0.6118, 0.4052, 0.8462, 0.3917, 0.3537, 0.3450, 0.5053, 0.5593],
  21726. [0.6224, 0.4061, 0.8988, 0.4300, 0.3837, 0.4750, 0.6112, 0.5483]]],
  21727. device='cuda:0')
  21728. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  21729. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  21730. loss_train: 0.0014395180041901767
  21731. step: 3
  21732. running loss: 0.0004798393347300589
  21733. Train Steps: 3/90 Loss: 0.0005 torch.Size([8, 600, 800])
  21734. torch.Size([8, 8])
  21735. tensor([[0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  21736. [0.6193, 0.4050, 0.7313, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  21737. [0.6202, 0.4079, 0.8025, 0.2500, 0.3763, 0.3217, 0.6125, 0.5533],
  21738. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683],
  21739. [0.6200, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  21740. [0.6100, 0.4071, 0.7601, 0.3444, 0.3400, 0.4117, 0.5625, 0.5617],
  21741. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  21742. [0.6182, 0.4058, 0.8738, 0.4350, 0.3563, 0.3400, 0.5290, 0.5822]],
  21743. device='cuda:0', dtype=torch.float64)
  21744. predictions are: tensor([[0.6081, 0.3873, 0.8497, 0.4549, 0.3521, 0.3878, 0.5540, 0.5651],
  21745. [0.6429, 0.4091, 0.7206, 0.2405, 0.4152, 0.2421, 0.5812, 0.5704],
  21746. [0.6230, 0.3810, 0.7868, 0.2496, 0.3900, 0.3198, 0.6284, 0.5419],
  21747. [0.6585, 0.4105, 0.8837, 0.5345, 0.3681, 0.3895, 0.5702, 0.5555],
  21748. [0.6274, 0.3987, 0.8372, 0.3034, 0.4211, 0.2373, 0.6061, 0.5390],
  21749. [0.6069, 0.3815, 0.7686, 0.3443, 0.3403, 0.4284, 0.5813, 0.5239],
  21750. [0.6252, 0.3910, 0.8705, 0.4495, 0.4389, 0.6050, 0.6370, 0.5267],
  21751. [0.6166, 0.3865, 0.8532, 0.4427, 0.3593, 0.3599, 0.5404, 0.5638]],
  21752. device='cuda:0', grad_fn=<AddmmBackward>)
  21753. landmarks are: tensor([[[0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  21754. [0.6193, 0.4050, 0.7312, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  21755. [0.6202, 0.4079, 0.8025, 0.2500, 0.3762, 0.3217, 0.6125, 0.5533],
  21756. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683],
  21757. [0.6199, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  21758. [0.6100, 0.4071, 0.7601, 0.3444, 0.3400, 0.4117, 0.5625, 0.5617],
  21759. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  21760. [0.6182, 0.4058, 0.8737, 0.4350, 0.3562, 0.3400, 0.5290, 0.5822]]],
  21761. device='cuda:0')
  21762. loss_train_step before backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  21763. loss_train_step after backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  21764. loss_train: 0.001678855056525208
  21765. step: 4
  21766. running loss: 0.000419713764131302
  21767. Train Steps: 4/90 Loss: 0.0004 torch.Size([8, 600, 800])
  21768. torch.Size([8, 8])
  21769. tensor([[0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6262, 0.5167],
  21770. [0.6138, 0.4054, 0.8750, 0.4750, 0.4363, 0.5017, 0.5086, 0.5822],
  21771. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  21772. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  21773. [ nan, nan, 0.6688, 0.2513, 0.4113, 0.2117, 0.5193, 0.5933],
  21774. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  21775. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  21776. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837]],
  21777. device='cuda:0', dtype=torch.float64)
  21778. predictions are: tensor([[0.6535, 0.4268, 0.9218, 0.4248, 0.3633, 0.3752, 0.6456, 0.5334],
  21779. [0.6299, 0.4239, 0.8814, 0.4930, 0.4228, 0.5131, 0.5455, 0.5724],
  21780. [0.6410, 0.4221, 0.8750, 0.2759, 0.4460, 0.2838, 0.7228, 0.5797],
  21781. [0.6569, 0.4039, 0.9135, 0.5521, 0.3506, 0.4886, 0.6536, 0.4932],
  21782. [0.0340, 0.0284, 0.6861, 0.2670, 0.4044, 0.2420, 0.5503, 0.6324],
  21783. [0.7165, 0.4682, 0.7702, 0.1931, 0.4608, 0.1776, 0.6070, 0.5090],
  21784. [0.6661, 0.4247, 0.8615, 0.4843, 0.3716, 0.5002, 0.5485, 0.5507],
  21785. [0.6522, 0.4341, 0.6852, 0.3236, 0.3517, 0.3098, 0.5355, 0.5977]],
  21786. device='cuda:0', grad_fn=<AddmmBackward>)
  21787. landmarks are: tensor([[[0.6268, 0.4094, 0.9038, 0.4300, 0.3700, 0.3483, 0.6263, 0.5167],
  21788. [0.6138, 0.4054, 0.8750, 0.4750, 0.4363, 0.5017, 0.5086, 0.5822],
  21789. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  21790. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  21791. [0.0000, 0.0000, 0.6688, 0.2513, 0.4112, 0.2117, 0.5193, 0.5933],
  21792. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  21793. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  21794. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837]]],
  21795. device='cuda:0')
  21796. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  21797. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  21798. loss_train: 0.002325661844224669
  21799. step: 5
  21800. running loss: 0.0004651323688449338
  21801. Train Steps: 5/90 Loss: 0.0005 torch.Size([8, 600, 800])
  21802. torch.Size([8, 8])
  21803. tensor([[0.6239, 0.4206, 0.8750, 0.5400, 0.3688, 0.4850, 0.5737, 0.5700],
  21804. [0.6261, 0.3987, 0.8688, 0.4917, 0.4300, 0.5333, 0.7010, 0.5309],
  21805. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  21806. [0.6276, 0.4002, 0.8800, 0.5533, 0.3575, 0.4400, 0.6132, 0.4672],
  21807. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  21808. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  21809. [0.6177, 0.4086, 0.8738, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700],
  21810. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167]],
  21811. device='cuda:0', dtype=torch.float64)
  21812. predictions are: tensor([[0.6606, 0.4245, 0.8836, 0.5583, 0.3539, 0.4847, 0.5659, 0.5619],
  21813. [0.6123, 0.3990, 0.8744, 0.5123, 0.4089, 0.5486, 0.6709, 0.5649],
  21814. [0.5972, 0.3926, 0.8888, 0.3162, 0.3896, 0.4101, 0.6946, 0.5617],
  21815. [0.5883, 0.3712, 0.8911, 0.5428, 0.3359, 0.4267, 0.6049, 0.5045],
  21816. [0.6754, 0.4337, 0.7266, 0.2542, 0.3625, 0.2963, 0.5969, 0.5932],
  21817. [0.5790, 0.3739, 0.8706, 0.4205, 0.3542, 0.4721, 0.5582, 0.5324],
  21818. [0.5863, 0.3972, 0.8765, 0.4084, 0.3567, 0.5534, 0.6111, 0.5487],
  21819. [0.6434, 0.4218, 0.7339, 0.3618, 0.4775, 0.1961, 0.5330, 0.6443]],
  21820. device='cuda:0', grad_fn=<AddmmBackward>)
  21821. landmarks are: tensor([[[0.6239, 0.4206, 0.8750, 0.5400, 0.3688, 0.4850, 0.5738, 0.5700],
  21822. [0.6261, 0.3987, 0.8687, 0.4917, 0.4300, 0.5333, 0.7010, 0.5309],
  21823. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  21824. [0.6276, 0.4002, 0.8800, 0.5533, 0.3575, 0.4400, 0.6132, 0.4672],
  21825. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  21826. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  21827. [0.6177, 0.4085, 0.8737, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700],
  21828. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167]]],
  21829. device='cuda:0')
  21830. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  21831. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  21832. loss_train: 0.0027636906161205843
  21833. step: 6
  21834. running loss: 0.00046061510268676403
  21835.  
  21836. Train Steps: 6/90 Loss: 0.0005 torch.Size([8, 600, 800])
  21837. torch.Size([8, 8])
  21838. tensor([[0.6208, 0.4082, 0.8538, 0.3067, 0.3588, 0.3717, 0.6112, 0.5517],
  21839. [0.6277, 0.4118, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  21840. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  21841. [0.6203, 0.4096, 0.8862, 0.4267, 0.3538, 0.4117, 0.6025, 0.5650],
  21842. [ nan, nan, 0.7725, 0.2611, 0.3675, 0.2733, 0.5413, 0.5167],
  21843. [0.6203, 0.4073, 0.8189, 0.2398, 0.4400, 0.2054, 0.5929, 0.5501],
  21844. [0.6336, 0.4154, 0.8900, 0.2767, 0.4988, 0.2867, 0.7422, 0.5540],
  21845. [0.6058, 0.3978, 0.8287, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461]],
  21846. device='cuda:0', dtype=torch.float64)
  21847. predictions are: tensor([[0.6093, 0.4016, 0.8455, 0.3136, 0.3467, 0.3746, 0.5997, 0.5630],
  21848. [0.6374, 0.4256, 0.8931, 0.3969, 0.3701, 0.2547, 0.6111, 0.5143],
  21849. [0.6286, 0.4087, 0.8342, 0.4794, 0.3684, 0.4703, 0.5179, 0.5471],
  21850. [0.6256, 0.4069, 0.8910, 0.4117, 0.3544, 0.4030, 0.5732, 0.5737],
  21851. [0.0518, 0.0447, 0.7377, 0.2758, 0.3662, 0.2630, 0.5332, 0.5880],
  21852. [0.6257, 0.4142, 0.7941, 0.2575, 0.4349, 0.1931, 0.5780, 0.5478],
  21853. [0.6132, 0.4034, 0.8644, 0.3036, 0.4870, 0.2848, 0.7201, 0.5546],
  21854. [0.6095, 0.4068, 0.8269, 0.3836, 0.3205, 0.4037, 0.5424, 0.5314]],
  21855. device='cuda:0', grad_fn=<AddmmBackward>)
  21856. landmarks are: tensor([[[0.6208, 0.4082, 0.8537, 0.3067, 0.3587, 0.3717, 0.6112, 0.5517],
  21857. [0.6277, 0.4117, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  21858. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  21859. [0.6203, 0.4096, 0.8863, 0.4267, 0.3537, 0.4117, 0.6025, 0.5650],
  21860. [0.0000, 0.0000, 0.7725, 0.2611, 0.3675, 0.2733, 0.5412, 0.5167],
  21861. [0.6203, 0.4073, 0.8189, 0.2398, 0.4400, 0.2054, 0.5929, 0.5501],
  21862. [0.6336, 0.4154, 0.8900, 0.2767, 0.4988, 0.2867, 0.7422, 0.5540],
  21863. [0.6058, 0.3978, 0.8288, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461]]],
  21864. device='cuda:0')
  21865. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  21866. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  21867. loss_train: 0.003107938144239597
  21868. step: 7
  21869. running loss: 0.0004439911634627996
  21870. Train Steps: 7/90 Loss: 0.0004 torch.Size([8, 600, 800])
  21871. torch.Size([8, 8])
  21872. tensor([[0.6095, 0.3970, 0.8688, 0.4767, 0.4860, 0.4879, 0.5191, 0.4940],
  21873. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  21874. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  21875. [0.6138, 0.4054, 0.8750, 0.4750, 0.4363, 0.5017, 0.5086, 0.5822],
  21876. [ nan, nan, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  21877. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795],
  21878. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  21879. [0.6163, 0.4114, 0.7650, 0.2017, 0.3763, 0.2867, 0.5631, 0.5071]],
  21880. device='cuda:0', dtype=torch.float64)
  21881. predictions are: tensor([[ 0.6563, 0.4177, 0.8839, 0.4649, 0.4715, 0.4626, 0.5183, 0.5332],
  21882. [ 0.6144, 0.4091, 0.6869, 0.3082, 0.3530, 0.2762, 0.5134, 0.5890],
  21883. [ 0.6505, 0.4218, 0.7742, 0.1857, 0.4556, 0.1476, 0.5805, 0.5013],
  21884. [ 0.6151, 0.4132, 0.8729, 0.4727, 0.4237, 0.4703, 0.5191, 0.5730],
  21885. [-0.0467, -0.0208, 0.8933, 0.2386, 0.5126, 0.2212, 0.7333, 0.5870],
  21886. [ 0.6474, 0.4061, 0.8690, 0.5207, 0.3831, 0.5279, 0.6993, 0.5735],
  21887. [ 0.6017, 0.3957, 0.8910, 0.5184, 0.4018, 0.5002, 0.5945, 0.5251],
  21888. [ 0.6256, 0.4011, 0.7802, 0.1927, 0.3695, 0.2676, 0.5585, 0.5046]],
  21889. device='cuda:0', grad_fn=<AddmmBackward>)
  21890. landmarks are: tensor([[[0.6095, 0.3970, 0.8687, 0.4767, 0.4860, 0.4879, 0.5191, 0.4940],
  21891. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  21892. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  21893. [0.6138, 0.4054, 0.8750, 0.4750, 0.4363, 0.5017, 0.5086, 0.5822],
  21894. [0.0000, 0.0000, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  21895. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795],
  21896. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  21897. [0.6163, 0.4114, 0.7650, 0.2017, 0.3762, 0.2867, 0.5631, 0.5071]]],
  21898. device='cuda:0')
  21899. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  21900. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  21901. loss_train: 0.003420340406592004
  21902. step: 8
  21903. running loss: 0.0004275425508240005
  21904. Train Steps: 8/90 Loss: 0.0004 torch.Size([8, 600, 800])
  21905. torch.Size([8, 8])
  21906. tensor([[0.6055, 0.4015, 0.7425, 0.2033, 0.4113, 0.1883, 0.5217, 0.4823],
  21907. [0.6081, 0.3950, 0.8538, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954],
  21908. [0.6250, 0.4013, 0.8525, 0.5417, 0.4037, 0.5117, 0.6325, 0.5017],
  21909. [0.6197, 0.4091, 0.8800, 0.4783, 0.3538, 0.4767, 0.5950, 0.5550],
  21910. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5637, 0.5633],
  21911. [ nan, nan, 0.8525, 0.2217, 0.5413, 0.2367, 0.7367, 0.5482],
  21912. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  21913. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973]],
  21914. device='cuda:0', dtype=torch.float64)
  21915. predictions are: tensor([[0.5593, 0.3640, 0.7191, 0.2089, 0.4127, 0.1832, 0.5501, 0.5045],
  21916. [0.5864, 0.3865, 0.8691, 0.4170, 0.3943, 0.4823, 0.5247, 0.5135],
  21917. [0.6240, 0.3902, 0.8571, 0.5376, 0.4267, 0.5107, 0.6087, 0.5094],
  21918. [0.6303, 0.4107, 0.8826, 0.4569, 0.3914, 0.4682, 0.5766, 0.5644],
  21919. [0.6528, 0.4362, 0.8672, 0.4647, 0.3807, 0.3725, 0.5555, 0.5838],
  21920. [0.0477, 0.0448, 0.8623, 0.2045, 0.5571, 0.2246, 0.7361, 0.5878],
  21921. [0.5963, 0.4004, 0.9031, 0.4261, 0.3855, 0.3173, 0.5781, 0.5401],
  21922. [0.6139, 0.3800, 0.8932, 0.5163, 0.3783, 0.4553, 0.6279, 0.4962]],
  21923. device='cuda:0', grad_fn=<AddmmBackward>)
  21924. landmarks are: tensor([[[0.6055, 0.4015, 0.7425, 0.2033, 0.4112, 0.1883, 0.5217, 0.4823],
  21925. [0.6081, 0.3950, 0.8537, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954],
  21926. [0.6250, 0.4013, 0.8525, 0.5417, 0.4038, 0.5117, 0.6325, 0.5017],
  21927. [0.6197, 0.4091, 0.8800, 0.4783, 0.3537, 0.4767, 0.5950, 0.5550],
  21928. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5638, 0.5633],
  21929. [0.0000, 0.0000, 0.8525, 0.2217, 0.5412, 0.2367, 0.7367, 0.5482],
  21930. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  21931. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973]]],
  21932. device='cuda:0')
  21933. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  21934. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  21935. loss_train: 0.0038243775634327903
  21936. step: 9
  21937. running loss: 0.00042493084038142115
  21938. Train Steps: 9/90 Loss: 0.0004 torch.Size([8, 600, 800])
  21939. torch.Size([8, 8])
  21940. tensor([[0.6204, 0.4110, 0.7913, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367],
  21941. [0.6200, 0.3993, 0.8519, 0.4923, 0.3962, 0.4717, 0.6013, 0.5433],
  21942. [0.6193, 0.4108, 0.7438, 0.2700, 0.3650, 0.3683, 0.6238, 0.5717],
  21943. [0.6168, 0.4029, 0.8523, 0.3417, 0.3588, 0.5000, 0.6125, 0.5400],
  21944. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  21945. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  21946. [0.6213, 0.4001, 0.7712, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183],
  21947. [0.6178, 0.4059, 0.7525, 0.2250, 0.4313, 0.1783, 0.5404, 0.5240]],
  21948. device='cuda:0', dtype=torch.float64)
  21949. predictions are: tensor([[0.5691, 0.3796, 0.8062, 0.2591, 0.4247, 0.2604, 0.6077, 0.5386],
  21950. [0.5695, 0.3792, 0.8748, 0.5051, 0.4170, 0.4710, 0.6073, 0.5523],
  21951. [0.5765, 0.3793, 0.7552, 0.2764, 0.3750, 0.3704, 0.6181, 0.5690],
  21952. [0.5474, 0.3674, 0.8702, 0.3460, 0.3741, 0.4950, 0.6105, 0.5452],
  21953. [0.5676, 0.3823, 0.8079, 0.3202, 0.3746, 0.3489, 0.5917, 0.5139],
  21954. [0.5589, 0.3677, 0.8124, 0.2105, 0.4556, 0.2762, 0.6631, 0.5270],
  21955. [0.5715, 0.3594, 0.7832, 0.2142, 0.4573, 0.1787, 0.5668, 0.5262],
  21956. [0.3931, 0.2683, 0.7582, 0.2373, 0.4466, 0.1760, 0.5537, 0.5485]],
  21957. device='cuda:0', grad_fn=<AddmmBackward>)
  21958. landmarks are: tensor([[[0.6204, 0.4110, 0.7912, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367],
  21959. [0.6200, 0.3993, 0.8519, 0.4923, 0.3963, 0.4717, 0.6012, 0.5433],
  21960. [0.6193, 0.4108, 0.7437, 0.2700, 0.3650, 0.3683, 0.6237, 0.5717],
  21961. [0.6168, 0.4029, 0.8523, 0.3417, 0.3587, 0.5000, 0.6125, 0.5400],
  21962. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  21963. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  21964. [0.6213, 0.4001, 0.7713, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183],
  21965. [0.6178, 0.4059, 0.7525, 0.2250, 0.4313, 0.1783, 0.5404, 0.5240]]],
  21966. device='cuda:0')
  21967. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  21968. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  21969. loss_train: 0.005475856611155905
  21970. step: 10
  21971. running loss: 0.0005475856611155905
  21972.  
  21973. Train Steps: 10/90 Loss: 0.0005 torch.Size([8, 600, 800])
  21974. torch.Size([8, 8])
  21975. tensor([[0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  21976. [0.6183, 0.4076, 0.8838, 0.4517, 0.3813, 0.4483, 0.5775, 0.5633],
  21977. [0.6186, 0.4060, 0.8750, 0.5050, 0.3538, 0.4367, 0.5813, 0.6083],
  21978. [0.6230, 0.4152, 0.7588, 0.2283, 0.4012, 0.2883, 0.6200, 0.5767],
  21979. [0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431],
  21980. [0.6138, 0.4101, 0.8800, 0.5083, 0.4637, 0.5950, 0.5587, 0.5077],
  21981. [0.6257, 0.4034, 0.8287, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  21982. [0.6187, 0.4104, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683]],
  21983. device='cuda:0', dtype=torch.float64)
  21984. predictions are: tensor([[0.5778, 0.3631, 0.8630, 0.4630, 0.4427, 0.4802, 0.6320, 0.5042],
  21985. [0.5571, 0.3702, 0.8847, 0.4294, 0.3884, 0.4266, 0.5899, 0.5506],
  21986. [0.5534, 0.3684, 0.8602, 0.4760, 0.3661, 0.4216, 0.5846, 0.5699],
  21987. [0.5960, 0.3879, 0.7611, 0.2382, 0.4183, 0.2803, 0.6296, 0.5309],
  21988. [0.5524, 0.3774, 0.8657, 0.4767, 0.4540, 0.4675, 0.5284, 0.4991],
  21989. [0.5111, 0.3414, 0.8815, 0.5135, 0.4777, 0.5534, 0.5801, 0.5005],
  21990. [0.6022, 0.3933, 0.8279, 0.2484, 0.3888, 0.2621, 0.6464, 0.4693],
  21991. [0.5504, 0.3655, 0.7145, 0.2024, 0.3990, 0.2294, 0.5790, 0.5200]],
  21992. device='cuda:0', grad_fn=<AddmmBackward>)
  21993. landmarks are: tensor([[[0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  21994. [0.6183, 0.4076, 0.8838, 0.4517, 0.3812, 0.4483, 0.5775, 0.5633],
  21995. [0.6186, 0.4060, 0.8750, 0.5050, 0.3537, 0.4367, 0.5813, 0.6083],
  21996. [0.6230, 0.4152, 0.7588, 0.2283, 0.4013, 0.2883, 0.6200, 0.5767],
  21997. [0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431],
  21998. [0.6138, 0.4101, 0.8800, 0.5083, 0.4638, 0.5950, 0.5587, 0.5077],
  21999. [0.6257, 0.4034, 0.8288, 0.2333, 0.3925, 0.2717, 0.6330, 0.4901],
  22000. [0.6187, 0.4103, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683]]],
  22001. device='cuda:0')
  22002. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  22003. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  22004. loss_train: 0.006415960393496789
  22005. step: 11
  22006. running loss: 0.0005832691266815262
  22007. Train Steps: 11/90 Loss: 0.0006 torch.Size([8, 600, 800])
  22008. torch.Size([8, 8])
  22009. tensor([[0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  22010. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901],
  22011. [0.6261, 0.3987, 0.8688, 0.4917, 0.4300, 0.5333, 0.7010, 0.5309],
  22012. [ nan, nan, 0.7525, 0.2291, 0.3838, 0.3017, 0.6050, 0.5667],
  22013. [0.6115, 0.4081, 0.6725, 0.2433, 0.4088, 0.1933, 0.5167, 0.5544],
  22014. [0.6095, 0.3970, 0.8688, 0.4767, 0.4860, 0.4879, 0.5191, 0.4940],
  22015. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  22016. [0.6084, 0.3981, 0.8588, 0.5233, 0.4600, 0.5367, 0.5680, 0.5006]],
  22017. device='cuda:0', dtype=torch.float64)
  22018. predictions are: tensor([[0.6073, 0.3913, 0.7963, 0.2558, 0.4119, 0.2415, 0.6013, 0.5137],
  22019. [0.5290, 0.3648, 0.7915, 0.2882, 0.3676, 0.2543, 0.5319, 0.4917],
  22020. [0.6104, 0.4107, 0.8586, 0.4751, 0.4234, 0.5301, 0.6855, 0.5266],
  22021. [0.0803, 0.0688, 0.7548, 0.2152, 0.3793, 0.2732, 0.5980, 0.5617],
  22022. [0.5532, 0.3812, 0.6788, 0.2455, 0.4015, 0.1991, 0.5203, 0.5415],
  22023. [0.5990, 0.3987, 0.8612, 0.4508, 0.4700, 0.4684, 0.5258, 0.5003],
  22024. [0.5285, 0.3691, 0.8373, 0.2425, 0.4806, 0.1649, 0.6189, 0.4771],
  22025. [0.6092, 0.4158, 0.8430, 0.5042, 0.4554, 0.5251, 0.5594, 0.4881]],
  22026. device='cuda:0', grad_fn=<AddmmBackward>)
  22027. landmarks are: tensor([[[0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  22028. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901],
  22029. [0.6261, 0.3987, 0.8687, 0.4917, 0.4300, 0.5333, 0.7010, 0.5309],
  22030. [0.0000, 0.0000, 0.7525, 0.2291, 0.3837, 0.3017, 0.6050, 0.5667],
  22031. [0.6115, 0.4081, 0.6725, 0.2433, 0.4087, 0.1933, 0.5167, 0.5544],
  22032. [0.6095, 0.3970, 0.8687, 0.4767, 0.4860, 0.4879, 0.5191, 0.4940],
  22033. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  22034. [0.6084, 0.3981, 0.8587, 0.5233, 0.4600, 0.5367, 0.5680, 0.5006]]],
  22035. device='cuda:0')
  22036. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  22037. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  22038. loss_train: 0.007117415210814215
  22039. step: 12
  22040. running loss: 0.000593117934234518
  22041. Train Steps: 12/90 Loss: 0.0006 torch.Size([8, 600, 800])
  22042. torch.Size([8, 8])
  22043. tensor([[0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915],
  22044. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  22045. [0.6241, 0.4143, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  22046. [0.6122, 0.4006, 0.8850, 0.4217, 0.4088, 0.5517, 0.6063, 0.5517],
  22047. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  22048. [0.6081, 0.3950, 0.8538, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954],
  22049. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  22050. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5988, 0.5667]],
  22051. device='cuda:0', dtype=torch.float64)
  22052. predictions are: tensor([[0.5904, 0.3898, 0.8998, 0.4425, 0.3572, 0.3174, 0.6593, 0.4781],
  22053. [0.5662, 0.3711, 0.6896, 0.1980, 0.4226, 0.1710, 0.5716, 0.4987],
  22054. [0.5754, 0.3873, 0.8808, 0.4244, 0.4162, 0.5290, 0.6554, 0.5356],
  22055. [0.5711, 0.3879, 0.8730, 0.4059, 0.4015, 0.5412, 0.6156, 0.5187],
  22056. [0.6003, 0.4078, 0.8135, 0.5540, 0.3982, 0.4318, 0.6006, 0.4987],
  22057. [0.5563, 0.3824, 0.8493, 0.4039, 0.3722, 0.4691, 0.5495, 0.4847],
  22058. [0.5480, 0.3759, 0.8244, 0.4778, 0.4327, 0.5286, 0.5453, 0.5120],
  22059. [0.5987, 0.4079, 0.8454, 0.3477, 0.3766, 0.3348, 0.6164, 0.5470]],
  22060. device='cuda:0', grad_fn=<AddmmBackward>)
  22061. landmarks are: tensor([[[0.6280, 0.4101, 0.9050, 0.4533, 0.3775, 0.3217, 0.6338, 0.4915],
  22062. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  22063. [0.6241, 0.4142, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  22064. [0.6122, 0.4006, 0.8850, 0.4217, 0.4087, 0.5517, 0.6062, 0.5517],
  22065. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  22066. [0.6081, 0.3950, 0.8537, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954],
  22067. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  22068. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5987, 0.5667]]],
  22069. device='cuda:0')
  22070. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  22071. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  22072. loss_train: 0.007744449147139676
  22073. step: 13
  22074. running loss: 0.0005957268574722827
  22075. Train Steps: 13/90 Loss: 0.0006 torch.Size([8, 600, 800])
  22076. torch.Size([8, 8])
  22077. tensor([[0.6204, 0.4049, 0.7975, 0.2700, 0.3937, 0.2567, 0.5700, 0.5183],
  22078. [0.6321, 0.4048, 0.8738, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  22079. [0.6075, 0.4000, 0.8513, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  22080. [0.6222, 0.3957, 0.8838, 0.5017, 0.3937, 0.4600, 0.5900, 0.5017],
  22081. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5463, 0.5800],
  22082. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  22083. [ nan, nan, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  22084. [0.6200, 0.4071, 0.7338, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517]],
  22085. device='cuda:0', dtype=torch.float64)
  22086. predictions are: tensor([[0.6580, 0.4510, 0.7884, 0.2578, 0.3902, 0.2608, 0.5653, 0.5235],
  22087. [0.6374, 0.4277, 0.8697, 0.5432, 0.3760, 0.4533, 0.6301, 0.4782],
  22088. [0.6094, 0.4081, 0.8409, 0.5083, 0.4566, 0.5553, 0.5343, 0.5236],
  22089. [0.6468, 0.4245, 0.8740, 0.4897, 0.3949, 0.4702, 0.5960, 0.4987],
  22090. [0.6265, 0.4372, 0.7443, 0.2968, 0.3676, 0.2812, 0.5386, 0.5554],
  22091. [0.6135, 0.4098, 0.8615, 0.3593, 0.3455, 0.3919, 0.5960, 0.4974],
  22092. [0.1953, 0.1307, 0.6816, 0.2437, 0.3851, 0.2490, 0.5604, 0.5534],
  22093. [0.5964, 0.4031, 0.7323, 0.1757, 0.4227, 0.2616, 0.6200, 0.5441]],
  22094. device='cuda:0', grad_fn=<AddmmBackward>)
  22095. landmarks are: tensor([[[0.6204, 0.4049, 0.7975, 0.2700, 0.3938, 0.2567, 0.5700, 0.5183],
  22096. [0.6321, 0.4048, 0.8737, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  22097. [0.6075, 0.4000, 0.8512, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  22098. [0.6222, 0.3957, 0.8838, 0.5017, 0.3938, 0.4600, 0.5900, 0.5017],
  22099. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5462, 0.5800],
  22100. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  22101. [0.0000, 0.0000, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  22102. [0.6200, 0.4071, 0.7337, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517]]],
  22103. device='cuda:0')
  22104. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  22105. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  22106. loss_train: 0.008856130370986648
  22107. step: 14
  22108. running loss: 0.0006325807407847606
  22109.  
  22110. Train Steps: 14/90 Loss: 0.0006 torch.Size([8, 600, 800])
  22111. torch.Size([8, 8])
  22112. tensor([[0.6143, 0.4040, 0.8237, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  22113. [0.6113, 0.4006, 0.8700, 0.5350, 0.3638, 0.3767, 0.5097, 0.4882],
  22114. [0.6189, 0.4049, 0.8888, 0.4417, 0.4213, 0.5200, 0.5988, 0.5633],
  22115. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  22116. [ nan, nan, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  22117. [0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6088, 0.5583],
  22118. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5837, 0.5500],
  22119. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235]],
  22120. device='cuda:0', dtype=torch.float64)
  22121. predictions are: tensor([[0.5724, 0.3819, 0.8246, 0.3438, 0.4118, 0.2549, 0.5484, 0.5102],
  22122. [0.6460, 0.4393, 0.8570, 0.5261, 0.3701, 0.4076, 0.5309, 0.5025],
  22123. [0.7027, 0.4749, 0.8744, 0.4550, 0.4139, 0.5604, 0.6192, 0.5783],
  22124. [0.6758, 0.4445, 0.8976, 0.4109, 0.3588, 0.3803, 0.5950, 0.5482],
  22125. [0.1091, 0.0646, 0.6863, 0.2242, 0.4313, 0.2017, 0.5191, 0.5573],
  22126. [0.6893, 0.4546, 0.7051, 0.2249, 0.3795, 0.2830, 0.6013, 0.5383],
  22127. [0.6455, 0.4368, 0.8594, 0.4563, 0.4097, 0.5432, 0.5876, 0.5352],
  22128. [0.6468, 0.4288, 0.8710, 0.4869, 0.4198, 0.5645, 0.6359, 0.5339]],
  22129. device='cuda:0', grad_fn=<AddmmBackward>)
  22130. landmarks are: tensor([[[0.6143, 0.4040, 0.8238, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  22131. [0.6113, 0.4006, 0.8700, 0.5350, 0.3638, 0.3767, 0.5097, 0.4882],
  22132. [0.6189, 0.4049, 0.8888, 0.4417, 0.4212, 0.5200, 0.5987, 0.5633],
  22133. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  22134. [0.0000, 0.0000, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  22135. [0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6087, 0.5583],
  22136. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5838, 0.5500],
  22137. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235]]],
  22138. device='cuda:0')
  22139. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  22140. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  22141. loss_train: 0.00981652743939776
  22142. step: 15
  22143. running loss: 0.0006544351626265173
  22144. Train Steps: 15/90 Loss: 0.0007 torch.Size([8, 600, 800])
  22145. torch.Size([8, 8])
  22146. tensor([[0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  22147. [0.6092, 0.4001, 0.8638, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  22148. [0.6213, 0.4131, 0.8438, 0.3550, 0.3513, 0.4400, 0.5716, 0.5123],
  22149. [0.6109, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  22150. [0.6202, 0.4053, 0.8638, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  22151. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  22152. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6138, 0.5333],
  22153. [0.6128, 0.4084, 0.8738, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397]],
  22154. device='cuda:0', dtype=torch.float64)
  22155. predictions are: tensor([[0.5996, 0.3768, 0.8086, 0.2620, 0.4501, 0.1922, 0.6225, 0.5224],
  22156. [0.6341, 0.4035, 0.8394, 0.4824, 0.4247, 0.5578, 0.5244, 0.5363],
  22157. [0.6212, 0.4082, 0.8375, 0.3579, 0.3429, 0.4226, 0.5650, 0.5385],
  22158. [0.6247, 0.4267, 0.8590, 0.4681, 0.3597, 0.4251, 0.5458, 0.5396],
  22159. [0.6561, 0.4138, 0.8484, 0.5444, 0.4495, 0.5152, 0.5933, 0.5391],
  22160. [0.6217, 0.4007, 0.8524, 0.4722, 0.3707, 0.5247, 0.6042, 0.5369],
  22161. [0.6051, 0.4039, 0.8702, 0.4627, 0.3706, 0.4843, 0.6178, 0.5528],
  22162. [0.6225, 0.3987, 0.8799, 0.4781, 0.3568, 0.3649, 0.5072, 0.5542]],
  22163. device='cuda:0', grad_fn=<AddmmBackward>)
  22164. landmarks are: tensor([[[0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  22165. [0.6092, 0.4001, 0.8637, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  22166. [0.6213, 0.4131, 0.8438, 0.3550, 0.3512, 0.4400, 0.5716, 0.5123],
  22167. [0.6108, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  22168. [0.6202, 0.4053, 0.8637, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  22169. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  22170. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6137, 0.5333],
  22171. [0.6127, 0.4084, 0.8737, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397]]],
  22172. device='cuda:0')
  22173. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  22174. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  22175. loss_train: 0.010077989150886424
  22176. step: 16
  22177. running loss: 0.0006298743219304015
  22178. Train Steps: 16/90 Loss: 0.0006 torch.Size([8, 600, 800])
  22179. torch.Size([8, 8])
  22180. tensor([[0.6267, 0.4065, 0.8313, 0.2467, 0.4788, 0.1733, 0.6312, 0.5133],
  22181. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  22182. [0.6250, 0.4103, 0.8950, 0.4400, 0.3912, 0.5650, 0.6050, 0.5133],
  22183. [0.6132, 0.4037, 0.6963, 0.2217, 0.4100, 0.1950, 0.5395, 0.5175],
  22184. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  22185. [0.6264, 0.4049, 0.8988, 0.4633, 0.3813, 0.4983, 0.6326, 0.4843],
  22186. [0.6083, 0.3957, 0.8638, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  22187. [0.6229, 0.4066, 0.7612, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350]],
  22188. device='cuda:0', dtype=torch.float64)
  22189. predictions are: tensor([[0.5917, 0.3692, 0.8392, 0.2501, 0.4780, 0.1923, 0.6008, 0.5347],
  22190. [0.6336, 0.4016, 0.8444, 0.5631, 0.3798, 0.4912, 0.6907, 0.6007],
  22191. [0.6221, 0.3998, 0.8935, 0.4706, 0.3912, 0.5854, 0.5955, 0.5473],
  22192. [0.6236, 0.4051, 0.7214, 0.2565, 0.3991, 0.2110, 0.5162, 0.5318],
  22193. [0.6419, 0.4099, 0.8480, 0.5898, 0.3739, 0.4971, 0.6184, 0.5603],
  22194. [0.6154, 0.3999, 0.9063, 0.4833, 0.3787, 0.5135, 0.6094, 0.5174],
  22195. [0.6243, 0.3887, 0.8631, 0.5044, 0.4348, 0.5294, 0.5217, 0.5199],
  22196. [0.6385, 0.3931, 0.7728, 0.2902, 0.4140, 0.2608, 0.5593, 0.5696]],
  22197. device='cuda:0', grad_fn=<AddmmBackward>)
  22198. landmarks are: tensor([[[0.6266, 0.4065, 0.8313, 0.2467, 0.4787, 0.1733, 0.6313, 0.5133],
  22199. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  22200. [0.6250, 0.4103, 0.8950, 0.4400, 0.3913, 0.5650, 0.6050, 0.5133],
  22201. [0.6132, 0.4037, 0.6963, 0.2217, 0.4100, 0.1950, 0.5395, 0.5175],
  22202. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  22203. [0.6264, 0.4049, 0.8988, 0.4633, 0.3812, 0.4983, 0.6326, 0.4843],
  22204. [0.6083, 0.3957, 0.8637, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  22205. [0.6229, 0.4066, 0.7613, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350]]],
  22206. device='cuda:0')
  22207. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  22208. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  22209. loss_train: 0.01048219391668681
  22210. step: 17
  22211. running loss: 0.0006165996421580477
  22212. Train Steps: 17/90 Loss: 0.0006 torch.Size([8, 600, 800])
  22213. torch.Size([8, 8])
  22214. tensor([[0.6197, 0.4091, 0.8800, 0.4783, 0.3538, 0.4767, 0.5950, 0.5550],
  22215. [0.6274, 0.4099, 0.8625, 0.3233, 0.4400, 0.1983, 0.5876, 0.4869],
  22216. [0.6200, 0.3961, 0.8461, 0.5497, 0.4142, 0.4577, 0.5892, 0.5402],
  22217. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  22218. [0.6339, 0.4123, 0.8638, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  22219. [0.6092, 0.4001, 0.8638, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  22220. [0.6246, 0.4028, 0.8738, 0.4867, 0.4088, 0.5667, 0.6362, 0.5200],
  22221. [ nan, nan, 0.6900, 0.1917, 0.3937, 0.2367, 0.5240, 0.5246]],
  22222. device='cuda:0', dtype=torch.float64)
  22223. predictions are: tensor([[6.8750e-01, 4.4056e-01, 9.0004e-01, 4.9645e-01, 3.8054e-01, 4.6319e-01,
  22224. 5.7499e-01, 5.7014e-01],
  22225. [7.5926e-01, 4.6508e-01, 8.8464e-01, 3.6296e-01, 4.5571e-01, 1.9916e-01,
  22226. 5.7849e-01, 5.0027e-01],
  22227. [6.7550e-01, 4.1488e-01, 8.6852e-01, 5.8610e-01, 4.1434e-01, 4.5401e-01,
  22228. 5.9690e-01, 5.4086e-01],
  22229. [6.8842e-01, 4.4305e-01, 8.9183e-01, 4.6133e-01, 3.9197e-01, 4.4541e-01,
  22230. 5.1547e-01, 5.5174e-01],
  22231. [6.6559e-01, 4.2687e-01, 8.7362e-01, 5.4431e-01, 4.1048e-01, 5.4079e-01,
  22232. 7.0992e-01, 5.4931e-01],
  22233. [6.8891e-01, 4.2708e-01, 8.7033e-01, 5.0001e-01, 4.3708e-01, 5.4782e-01,
  22234. 5.0905e-01, 5.1298e-01],
  22235. [6.8797e-01, 4.2947e-01, 8.8706e-01, 5.1611e-01, 4.3128e-01, 5.7506e-01,
  22236. 6.0441e-01, 5.3036e-01],
  22237. [5.4723e-02, 5.4600e-04, 7.2110e-01, 2.2487e-01, 4.1556e-01, 2.5140e-01,
  22238. 5.0913e-01, 5.2042e-01]], device='cuda:0', grad_fn=<AddmmBackward>)
  22239. landmarks are: tensor([[[0.6197, 0.4091, 0.8800, 0.4783, 0.3537, 0.4767, 0.5950, 0.5550],
  22240. [0.6274, 0.4099, 0.8625, 0.3233, 0.4400, 0.1983, 0.5876, 0.4869],
  22241. [0.6200, 0.3961, 0.8461, 0.5497, 0.4142, 0.4577, 0.5892, 0.5402],
  22242. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  22243. [0.6339, 0.4123, 0.8637, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  22244. [0.6092, 0.4001, 0.8637, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  22245. [0.6246, 0.4028, 0.8737, 0.4867, 0.4087, 0.5667, 0.6363, 0.5200],
  22246. [0.0000, 0.0000, 0.6900, 0.1917, 0.3938, 0.2367, 0.5240, 0.5246]]],
  22247. device='cuda:0')
  22248. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  22249. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  22250. loss_train: 0.011588190231123008
  22251. step: 18
  22252. running loss: 0.0006437883461735004
  22253.  
  22254. Train Steps: 18/90 Loss: 0.0006 torch.Size([8, 600, 800])
  22255. torch.Size([8, 8])
  22256. tensor([[0.6189, 0.4049, 0.8888, 0.4417, 0.4213, 0.5200, 0.5988, 0.5633],
  22257. [0.6350, 0.4043, 0.8738, 0.5650, 0.3850, 0.4750, 0.6401, 0.4950],
  22258. [0.6197, 0.3986, 0.8800, 0.4617, 0.4188, 0.4783, 0.5687, 0.5550],
  22259. [0.6204, 0.4007, 0.7838, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  22260. [0.6022, 0.3994, 0.8025, 0.3350, 0.3350, 0.4400, 0.5565, 0.5025],
  22261. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  22262. [0.6156, 0.4125, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  22263. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383]],
  22264. device='cuda:0', dtype=torch.float64)
  22265. predictions are: tensor([[0.6147, 0.3887, 0.8863, 0.4724, 0.4163, 0.5164, 0.5955, 0.5731],
  22266. [0.6220, 0.3836, 0.8847, 0.6055, 0.3916, 0.4640, 0.6227, 0.5119],
  22267. [0.6361, 0.4026, 0.8757, 0.4867, 0.4067, 0.4620, 0.5409, 0.5462],
  22268. [0.6163, 0.3732, 0.7845, 0.2437, 0.4469, 0.1474, 0.5833, 0.5044],
  22269. [0.6080, 0.3835, 0.8076, 0.3627, 0.3498, 0.4062, 0.5678, 0.5156],
  22270. [0.5851, 0.3613, 0.8399, 0.2569, 0.4622, 0.2293, 0.7007, 0.5473],
  22271. [0.6045, 0.3970, 0.8956, 0.5041, 0.4461, 0.5515, 0.5509, 0.5079],
  22272. [0.5926, 0.3736, 0.8545, 0.4993, 0.4274, 0.4609, 0.5187, 0.5249]],
  22273. device='cuda:0', grad_fn=<AddmmBackward>)
  22274. landmarks are: tensor([[[0.6189, 0.4049, 0.8888, 0.4417, 0.4212, 0.5200, 0.5987, 0.5633],
  22275. [0.6350, 0.4043, 0.8737, 0.5650, 0.3850, 0.4750, 0.6401, 0.4950],
  22276. [0.6197, 0.3986, 0.8800, 0.4617, 0.4187, 0.4783, 0.5688, 0.5550],
  22277. [0.6204, 0.4007, 0.7837, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  22278. [0.6022, 0.3994, 0.8025, 0.3350, 0.3350, 0.4400, 0.5565, 0.5025],
  22279. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  22280. [0.6155, 0.4124, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  22281. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383]]],
  22282. device='cuda:0')
  22283. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  22284. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  22285. loss_train: 0.011935614966205321
  22286. step: 19
  22287. running loss: 0.0006281902613792274
  22288. Train Steps: 19/90 Loss: 0.0006 torch.Size([8, 600, 800])
  22289. torch.Size([8, 8])
  22290. tensor([[0.6082, 0.4024, 0.8738, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  22291. [0.6183, 0.4076, 0.8838, 0.4517, 0.3813, 0.4483, 0.5775, 0.5633],
  22292. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  22293. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  22294. [0.6133, 0.4094, 0.8495, 0.4028, 0.3588, 0.3200, 0.5003, 0.5407],
  22295. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  22296. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  22297. [0.6200, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913]],
  22298. device='cuda:0', dtype=torch.float64)
  22299. predictions are: tensor([[0.6263, 0.3925, 0.8732, 0.3921, 0.3849, 0.3956, 0.5455, 0.4974],
  22300. [0.6533, 0.4087, 0.8859, 0.4419, 0.3958, 0.4445, 0.5874, 0.5733],
  22301. [0.6267, 0.3943, 0.8719, 0.5444, 0.4120, 0.4272, 0.5390, 0.4761],
  22302. [0.6193, 0.3927, 0.8507, 0.5603, 0.3943, 0.4842, 0.6691, 0.5243],
  22303. [0.5988, 0.3827, 0.8665, 0.3997, 0.3696, 0.3102, 0.5047, 0.5236],
  22304. [0.6087, 0.3742, 0.9097, 0.4067, 0.3816, 0.3709, 0.6709, 0.5170],
  22305. [0.6347, 0.3979, 0.8637, 0.5529, 0.3940, 0.4270, 0.6110, 0.5263],
  22306. [0.5986, 0.3606, 0.8703, 0.5145, 0.4212, 0.5126, 0.6077, 0.4776]],
  22307. device='cuda:0', grad_fn=<AddmmBackward>)
  22308. landmarks are: tensor([[[0.6082, 0.4024, 0.8737, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  22309. [0.6183, 0.4076, 0.8838, 0.4517, 0.3812, 0.4483, 0.5775, 0.5633],
  22310. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  22311. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  22312. [0.6133, 0.4094, 0.8495, 0.4028, 0.3587, 0.3200, 0.5003, 0.5407],
  22313. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  22314. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  22315. [0.6199, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913]]],
  22316. device='cuda:0')
  22317. loss_train_step before backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  22318. loss_train_step after backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  22319. loss_train: 0.012161795937572606
  22320. step: 20
  22321. running loss: 0.0006080897968786303
  22322. Train Steps: 20/90 Loss: 0.0006 torch.Size([8, 600, 800])
  22323. torch.Size([8, 8])
  22324. tensor([[0.6259, 0.4133, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  22325. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309],
  22326. [0.6205, 0.4081, 0.8950, 0.4017, 0.3788, 0.4700, 0.5963, 0.5667],
  22327. [0.6228, 0.4004, 0.8750, 0.5250, 0.3825, 0.5233, 0.6362, 0.5000],
  22328. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  22329. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116],
  22330. [0.6346, 0.4165, 0.9138, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471],
  22331. [0.6126, 0.4067, 0.8638, 0.5383, 0.4188, 0.4850, 0.5016, 0.5392]],
  22332. device='cuda:0', dtype=torch.float64)
  22333. predictions are: tensor([[0.5987, 0.3954, 0.8173, 0.2192, 0.4885, 0.1446, 0.6238, 0.4868],
  22334. [0.5582, 0.3632, 0.8971, 0.4730, 0.3848, 0.3590, 0.6402, 0.5108],
  22335. [0.6092, 0.3868, 0.8781, 0.3871, 0.3858, 0.4487, 0.6047, 0.5587],
  22336. [0.5910, 0.3785, 0.8639, 0.5020, 0.3823, 0.4878, 0.6334, 0.5027],
  22337. [0.5871, 0.3863, 0.8189, 0.5730, 0.4102, 0.4278, 0.5889, 0.5085],
  22338. [0.5511, 0.3794, 0.8619, 0.4024, 0.4057, 0.5856, 0.5989, 0.5047],
  22339. [0.6610, 0.4281, 0.9137, 0.3863, 0.4046, 0.4023, 0.7138, 0.5366],
  22340. [0.6162, 0.4093, 0.8579, 0.5548, 0.4164, 0.4813, 0.5180, 0.5221]],
  22341. device='cuda:0', grad_fn=<AddmmBackward>)
  22342. landmarks are: tensor([[[0.6259, 0.4132, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  22343. [0.6361, 0.4165, 0.9038, 0.5017, 0.3750, 0.3783, 0.6448, 0.5309],
  22344. [0.6205, 0.4081, 0.8950, 0.4017, 0.3787, 0.4700, 0.5962, 0.5667],
  22345. [0.6228, 0.4004, 0.8750, 0.5250, 0.3825, 0.5233, 0.6363, 0.5000],
  22346. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  22347. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116],
  22348. [0.6346, 0.4165, 0.9137, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471],
  22349. [0.6126, 0.4067, 0.8637, 0.5383, 0.4187, 0.4850, 0.5016, 0.5392]]],
  22350. device='cuda:0')
  22351. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  22352. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  22353. loss_train: 0.012637686988455243
  22354. step: 21
  22355. running loss: 0.0006017946184978687
  22356. Train Steps: 21/90 Loss: 0.0006 torch.Size([8, 600, 800])
  22357. torch.Size([8, 8])
  22358. tensor([[0.6263, 0.4065, 0.9038, 0.4317, 0.3588, 0.4550, 0.6325, 0.5250],
  22359. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864],
  22360. [0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  22361. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  22362. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  22363. [ nan, nan, 0.6412, 0.1900, 0.4238, 0.1883, 0.5487, 0.5700],
  22364. [0.6266, 0.4070, 0.8712, 0.5600, 0.3713, 0.4783, 0.5775, 0.6100],
  22365. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446]],
  22366. device='cuda:0', dtype=torch.float64)
  22367. predictions are: tensor([[0.6649, 0.4329, 0.9265, 0.4418, 0.3756, 0.4570, 0.6535, 0.5047],
  22368. [0.6515, 0.4245, 0.8946, 0.5157, 0.3987, 0.5116, 0.7217, 0.5745],
  22369. [0.6301, 0.4162, 0.8319, 0.2472, 0.4605, 0.1993, 0.6361, 0.4867],
  22370. [0.6229, 0.3967, 0.9096, 0.5021, 0.4142, 0.5424, 0.6717, 0.4770],
  22371. [0.6070, 0.4140, 0.7183, 0.2585, 0.4197, 0.2502, 0.5961, 0.5439],
  22372. [0.1898, 0.1257, 0.6954, 0.2147, 0.4328, 0.2136, 0.5403, 0.5564],
  22373. [0.6513, 0.4347, 0.8825, 0.5707, 0.3661, 0.5091, 0.6079, 0.5719],
  22374. [0.6652, 0.4476, 0.8582, 0.4722, 0.3887, 0.4969, 0.5615, 0.5470]],
  22375. device='cuda:0', grad_fn=<AddmmBackward>)
  22376. landmarks are: tensor([[[0.6263, 0.4065, 0.9038, 0.4317, 0.3587, 0.4550, 0.6325, 0.5250],
  22377. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864],
  22378. [0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  22379. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  22380. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  22381. [0.0000, 0.0000, 0.6413, 0.1900, 0.4238, 0.1883, 0.5487, 0.5700],
  22382. [0.6266, 0.4070, 0.8712, 0.5600, 0.3713, 0.4783, 0.5775, 0.6100],
  22383. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446]]],
  22384. device='cuda:0')
  22385. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  22386. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  22387. loss_train: 0.013873206989956088
  22388. step: 22
  22389. running loss: 0.0006306003177252768
  22390.  
  22391. Train Steps: 22/90 Loss: 0.0006 torch.Size([8, 600, 800])
  22392. torch.Size([8, 8])
  22393. tensor([[0.6124, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  22394. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5513, 0.5750],
  22395. [0.6277, 0.4057, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  22396. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  22397. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  22398. [ nan, nan, 0.7335, 0.2569, 0.3788, 0.2667, 0.5066, 0.5578],
  22399. [0.6250, 0.3993, 0.9138, 0.4333, 0.3763, 0.5217, 0.6995, 0.5320],
  22400. [0.6179, 0.4082, 0.6688, 0.2667, 0.3588, 0.3317, 0.5750, 0.5783]],
  22401. device='cuda:0', dtype=torch.float64)
  22402. predictions are: tensor([[ 0.6789, 0.4531, 0.8366, 0.5151, 0.3868, 0.4881, 0.5628, 0.5444],
  22403. [ 0.5932, 0.3999, 0.6746, 0.2478, 0.4322, 0.2141, 0.5409, 0.5745],
  22404. [ 0.6355, 0.4195, 0.8298, 0.2498, 0.4403, 0.2101, 0.6484, 0.4799],
  22405. [ 0.6366, 0.4174, 0.8163, 0.2554, 0.3985, 0.2939, 0.6574, 0.5077],
  22406. [ 0.6709, 0.4485, 0.8697, 0.2799, 0.4090, 0.2833, 0.6203, 0.5327],
  22407. [-0.0907, -0.0473, 0.7337, 0.2581, 0.3986, 0.2763, 0.5389, 0.5449],
  22408. [ 0.6507, 0.4282, 0.9220, 0.4580, 0.3847, 0.5605, 0.7420, 0.5411],
  22409. [ 0.6149, 0.4125, 0.6879, 0.2764, 0.3574, 0.3549, 0.5735, 0.5754]],
  22410. device='cuda:0', grad_fn=<AddmmBackward>)
  22411. landmarks are: tensor([[[0.6123, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  22412. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5512, 0.5750],
  22413. [0.6277, 0.4056, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  22414. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  22415. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  22416. [0.0000, 0.0000, 0.7335, 0.2569, 0.3787, 0.2667, 0.5066, 0.5578],
  22417. [0.6250, 0.3993, 0.9137, 0.4333, 0.3762, 0.5217, 0.6995, 0.5320],
  22418. [0.6179, 0.4082, 0.6687, 0.2667, 0.3587, 0.3317, 0.5750, 0.5783]]],
  22419. device='cuda:0')
  22420. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  22421. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  22422. loss_train: 0.014488388042082079
  22423. step: 23
  22424. running loss: 0.0006299299148731338
  22425. Train Steps: 23/90 Loss: 0.0006 torch.Size([8, 600, 800])
  22426. torch.Size([8, 8])
  22427. tensor([[0.6198, 0.4076, 0.7945, 0.2453, 0.4475, 0.1871, 0.5897, 0.5350],
  22428. [0.6165, 0.4106, 0.7575, 0.1733, 0.3838, 0.2650, 0.5680, 0.5116],
  22429. [0.6262, 0.4052, 0.8888, 0.4700, 0.3675, 0.5117, 0.6350, 0.5233],
  22430. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  22431. [0.6055, 0.4015, 0.7425, 0.2033, 0.4113, 0.1883, 0.5217, 0.4823],
  22432. [0.6263, 0.4233, 0.7924, 0.4626, 0.3788, 0.2883, 0.5573, 0.6047],
  22433. [0.6265, 0.4251, 0.7113, 0.3550, 0.4375, 0.2117, 0.5587, 0.6118],
  22434. [0.6289, 0.4032, 0.8419, 0.5446, 0.4075, 0.5017, 0.6312, 0.5117]],
  22435. device='cuda:0', dtype=torch.float64)
  22436. predictions are: tensor([[0.6425, 0.4334, 0.7857, 0.2444, 0.4534, 0.2241, 0.6116, 0.5370],
  22437. [0.5337, 0.3651, 0.7608, 0.1682, 0.3684, 0.2772, 0.5912, 0.5040],
  22438. [0.6076, 0.4188, 0.9171, 0.4704, 0.3646, 0.5334, 0.6596, 0.5171],
  22439. [0.5587, 0.3836, 0.8954, 0.4398, 0.3733, 0.4680, 0.5639, 0.5459],
  22440. [0.4811, 0.3287, 0.7080, 0.1976, 0.3847, 0.2021, 0.5528, 0.4906],
  22441. [0.5850, 0.4175, 0.8096, 0.4730, 0.3751, 0.3225, 0.5790, 0.6054],
  22442. [0.5410, 0.3818, 0.7291, 0.3556, 0.4151, 0.2339, 0.5605, 0.6092],
  22443. [0.6152, 0.3929, 0.8487, 0.5331, 0.3923, 0.5295, 0.6696, 0.5108]],
  22444. device='cuda:0', grad_fn=<AddmmBackward>)
  22445. landmarks are: tensor([[[0.6198, 0.4076, 0.7945, 0.2453, 0.4475, 0.1871, 0.5897, 0.5350],
  22446. [0.6165, 0.4106, 0.7575, 0.1733, 0.3837, 0.2650, 0.5680, 0.5116],
  22447. [0.6262, 0.4052, 0.8888, 0.4700, 0.3675, 0.5117, 0.6350, 0.5233],
  22448. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  22449. [0.6055, 0.4015, 0.7425, 0.2033, 0.4112, 0.1883, 0.5217, 0.4823],
  22450. [0.6263, 0.4232, 0.7924, 0.4626, 0.3787, 0.2883, 0.5573, 0.6047],
  22451. [0.6265, 0.4251, 0.7113, 0.3550, 0.4375, 0.2117, 0.5587, 0.6118],
  22452. [0.6289, 0.4031, 0.8419, 0.5446, 0.4075, 0.5017, 0.6313, 0.5117]]],
  22453. device='cuda:0')
  22454. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  22455. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  22456. loss_train: 0.015468385492567904
  22457. step: 24
  22458. running loss: 0.0006445160621903293
  22459. Train Steps: 24/90 Loss: 0.0006 torch.Size([8, 600, 800])
  22460. torch.Size([8, 8])
  22461. tensor([[0.6284, 0.4093, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  22462. [0.6113, 0.4006, 0.8700, 0.5350, 0.3638, 0.3767, 0.5097, 0.4882],
  22463. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  22464. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  22465. [0.6307, 0.4029, 0.8650, 0.5200, 0.3763, 0.4017, 0.7311, 0.5366],
  22466. [0.6197, 0.3986, 0.8800, 0.4617, 0.4188, 0.4783, 0.5687, 0.5550],
  22467. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  22468. [0.6162, 0.4014, 0.8800, 0.5333, 0.3750, 0.4817, 0.5988, 0.5283]],
  22469. device='cuda:0', dtype=torch.float64)
  22470. predictions are: tensor([[0.6276, 0.4176, 0.8725, 0.4386, 0.3636, 0.3857, 0.6137, 0.5281],
  22471. [0.5841, 0.3924, 0.8455, 0.4871, 0.3654, 0.3801, 0.5033, 0.5044],
  22472. [0.5597, 0.3609, 0.8477, 0.4645, 0.4077, 0.4724, 0.5645, 0.5286],
  22473. [0.5953, 0.3937, 0.8515, 0.4216, 0.3513, 0.3099, 0.5330, 0.5782],
  22474. [0.6220, 0.4098, 0.8538, 0.4849, 0.3751, 0.3998, 0.7113, 0.5320],
  22475. [0.6160, 0.4092, 0.8647, 0.4288, 0.4002, 0.4723, 0.5555, 0.5552],
  22476. [0.6098, 0.3939, 0.8777, 0.4970, 0.3560, 0.4419, 0.6439, 0.5046],
  22477. [0.5940, 0.3962, 0.8593, 0.5128, 0.3781, 0.4755, 0.5813, 0.5490]],
  22478. device='cuda:0', grad_fn=<AddmmBackward>)
  22479. landmarks are: tensor([[[0.6284, 0.4092, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  22480. [0.6113, 0.4006, 0.8700, 0.5350, 0.3638, 0.3767, 0.5097, 0.4882],
  22481. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  22482. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  22483. [0.6307, 0.4029, 0.8650, 0.5200, 0.3762, 0.4017, 0.7311, 0.5366],
  22484. [0.6197, 0.3986, 0.8800, 0.4617, 0.4187, 0.4783, 0.5688, 0.5550],
  22485. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  22486. [0.6162, 0.4014, 0.8800, 0.5333, 0.3750, 0.4817, 0.5987, 0.5283]]],
  22487. device='cuda:0')
  22488. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  22489. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  22490. loss_train: 0.015839603831409477
  22491. step: 25
  22492. running loss: 0.000633584153256379
  22493. Train Steps: 25/90 Loss: 0.0006 torch.Size([8, 600, 800])
  22494. torch.Size([8, 8])
  22495. tensor([[0.6300, 0.4133, 0.8538, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  22496. [0.6083, 0.3957, 0.8638, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  22497. [0.6075, 0.4000, 0.8513, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  22498. [0.6271, 0.4040, 0.9138, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  22499. [0.6260, 0.4153, 0.9000, 0.4533, 0.4025, 0.2633, 0.6223, 0.4967],
  22500. [0.6275, 0.4111, 0.8463, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  22501. [0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  22502. [0.6257, 0.4167, 0.8775, 0.3433, 0.3563, 0.4133, 0.6200, 0.5667]],
  22503. device='cuda:0', dtype=torch.float64)
  22504. predictions are: tensor([[0.5226, 0.3504, 0.8105, 0.2154, 0.5130, 0.2438, 0.6988, 0.5625],
  22505. [0.5402, 0.3682, 0.8372, 0.4816, 0.4074, 0.4962, 0.5079, 0.5018],
  22506. [0.5685, 0.3850, 0.8234, 0.5167, 0.4233, 0.5160, 0.4992, 0.5361],
  22507. [0.5714, 0.3812, 0.8884, 0.3620, 0.4291, 0.2457, 0.6756, 0.5418],
  22508. [0.6046, 0.4120, 0.8562, 0.4381, 0.3657, 0.2545, 0.5891, 0.5079],
  22509. [0.5323, 0.3624, 0.7989, 0.2456, 0.4330, 0.1772, 0.5702, 0.5084],
  22510. [0.5645, 0.3777, 0.8732, 0.3930, 0.3376, 0.4368, 0.5953, 0.5110],
  22511. [0.5504, 0.3856, 0.8446, 0.3207, 0.3229, 0.3875, 0.5939, 0.5799]],
  22512. device='cuda:0', grad_fn=<AddmmBackward>)
  22513. landmarks are: tensor([[[0.6300, 0.4133, 0.8537, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  22514. [0.6083, 0.3957, 0.8637, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  22515. [0.6075, 0.4000, 0.8512, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  22516. [0.6271, 0.4040, 0.9137, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  22517. [0.6260, 0.4153, 0.9000, 0.4533, 0.4025, 0.2633, 0.6223, 0.4967],
  22518. [0.6275, 0.4111, 0.8462, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  22519. [0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  22520. [0.6257, 0.4167, 0.8775, 0.3433, 0.3562, 0.4133, 0.6200, 0.5667]]],
  22521. device='cuda:0')
  22522. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  22523. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  22524. loss_train: 0.01718238960893359
  22525. step: 26
  22526. running loss: 0.0006608611388051381
  22527.  
  22528. Train Steps: 26/90 Loss: 0.0007 torch.Size([8, 600, 800])
  22529. torch.Size([8, 8])
  22530. tensor([[0.6076, 0.3953, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954],
  22531. [ nan, nan, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  22532. [0.6091, 0.3997, 0.8314, 0.4334, 0.3788, 0.4550, 0.5213, 0.5656],
  22533. [0.6193, 0.4079, 0.7288, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  22534. [0.6055, 0.4015, 0.7425, 0.2033, 0.4113, 0.1883, 0.5217, 0.4823],
  22535. [0.6254, 0.3993, 0.8988, 0.4767, 0.3987, 0.5517, 0.6955, 0.5285],
  22536. [0.6259, 0.4156, 0.8812, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960],
  22537. [0.6114, 0.4018, 0.7213, 0.1967, 0.3763, 0.2700, 0.5875, 0.5533]],
  22538. device='cuda:0', dtype=torch.float64)
  22539. predictions are: tensor([[ 0.6554, 0.4223, 0.8275, 0.4110, 0.3327, 0.3789, 0.5315, 0.5074],
  22540. [-0.0383, -0.0347, 0.6959, 0.2131, 0.4342, 0.2363, 0.5394, 0.5475],
  22541. [ 0.7011, 0.4454, 0.8374, 0.4684, 0.3629, 0.4386, 0.5320, 0.5673],
  22542. [ 0.6363, 0.4132, 0.7260, 0.2594, 0.4300, 0.2263, 0.5920, 0.6204],
  22543. [ 0.6163, 0.4085, 0.7112, 0.2127, 0.3985, 0.1573, 0.5243, 0.5062],
  22544. [ 0.5961, 0.3852, 0.9198, 0.5011, 0.4065, 0.5625, 0.6961, 0.5370],
  22545. [ 0.7267, 0.4630, 0.8801, 0.3093, 0.4681, 0.1737, 0.6029, 0.5127],
  22546. [ 0.5984, 0.3974, 0.7103, 0.1963, 0.3715, 0.2508, 0.5763, 0.5615]],
  22547. device='cuda:0', grad_fn=<AddmmBackward>)
  22548. landmarks are: tensor([[[0.6076, 0.3952, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954],
  22549. [0.0000, 0.0000, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  22550. [0.6091, 0.3997, 0.8314, 0.4334, 0.3787, 0.4550, 0.5213, 0.5656],
  22551. [0.6193, 0.4078, 0.7287, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  22552. [0.6055, 0.4015, 0.7425, 0.2033, 0.4112, 0.1883, 0.5217, 0.4823],
  22553. [0.6254, 0.3993, 0.8988, 0.4767, 0.3988, 0.5517, 0.6955, 0.5285],
  22554. [0.6259, 0.4156, 0.8813, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960],
  22555. [0.6114, 0.4018, 0.7212, 0.1967, 0.3762, 0.2700, 0.5875, 0.5533]]],
  22556. device='cuda:0')
  22557. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  22558. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  22559. loss_train: 0.01784341588790994
  22560. step: 27
  22561. running loss: 0.0006608672551077756
  22562. Train Steps: 27/90 Loss: 0.0007 torch.Size([8, 600, 800])
  22563. torch.Size([8, 8])
  22564. tensor([[0.6286, 0.4078, 0.8063, 0.2267, 0.4788, 0.1533, 0.5953, 0.4913],
  22565. [ nan, nan, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  22566. [0.6085, 0.4008, 0.8588, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  22567. [0.6276, 0.4235, 0.8888, 0.5333, 0.3800, 0.3117, 0.5427, 0.6164],
  22568. [0.6198, 0.4076, 0.7945, 0.2453, 0.4475, 0.1871, 0.5897, 0.5350],
  22569. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5963, 0.6217],
  22570. [ nan, nan, 0.6688, 0.2513, 0.4113, 0.2117, 0.5193, 0.5933],
  22571. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355]],
  22572. device='cuda:0', dtype=torch.float64)
  22573. predictions are: tensor([[ 0.7125, 0.4531, 0.7847, 0.2330, 0.4667, 0.1434, 0.5637, 0.4987],
  22574. [ 0.0361, 0.0144, 0.6946, 0.1977, 0.4312, 0.2426, 0.5346, 0.5346],
  22575. [ 0.7027, 0.4471, 0.8598, 0.5141, 0.4767, 0.4845, 0.5079, 0.5195],
  22576. [ 0.6661, 0.4406, 0.8856, 0.5504, 0.3870, 0.3181, 0.5491, 0.5983],
  22577. [ 0.7124, 0.4633, 0.7861, 0.2478, 0.4556, 0.1856, 0.5815, 0.5314],
  22578. [ 0.7486, 0.4859, 0.7194, 0.2394, 0.4215, 0.1986, 0.5841, 0.6084],
  22579. [-0.0305, -0.0297, 0.6597, 0.2384, 0.4110, 0.2072, 0.5175, 0.5639],
  22580. [ 0.6786, 0.4400, 0.8910, 0.3322, 0.4522, 0.3354, 0.7192, 0.5160]],
  22581. device='cuda:0', grad_fn=<AddmmBackward>)
  22582. landmarks are: tensor([[[0.6286, 0.4078, 0.8062, 0.2267, 0.4787, 0.1533, 0.5953, 0.4913],
  22583. [0.0000, 0.0000, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  22584. [0.6084, 0.4008, 0.8587, 0.5200, 0.4959, 0.4977, 0.5175, 0.5283],
  22585. [0.6276, 0.4235, 0.8888, 0.5333, 0.3800, 0.3117, 0.5427, 0.6164],
  22586. [0.6198, 0.4076, 0.7945, 0.2453, 0.4475, 0.1871, 0.5897, 0.5350],
  22587. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5962, 0.6217],
  22588. [0.0000, 0.0000, 0.6688, 0.2513, 0.4112, 0.2117, 0.5193, 0.5933],
  22589. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355]]],
  22590. device='cuda:0')
  22591. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  22592. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  22593. loss_train: 0.018996318729477935
  22594. step: 28
  22595. running loss: 0.0006784399546242119
  22596. Train Steps: 28/90 Loss: 0.0007 torch.Size([8, 600, 800])
  22597. torch.Size([8, 8])
  22598. tensor([[0.6072, 0.4029, 0.7037, 0.2150, 0.3912, 0.2267, 0.5516, 0.5507],
  22599. [0.6277, 0.4083, 0.8350, 0.2717, 0.4562, 0.1800, 0.5918, 0.4878],
  22600. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  22601. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  22602. [0.6192, 0.4128, 0.8513, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  22603. [0.6346, 0.4165, 0.9138, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471],
  22604. [0.6259, 0.4133, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  22605. [0.6224, 0.4061, 0.8988, 0.4300, 0.3838, 0.4750, 0.6112, 0.5483]],
  22606. device='cuda:0', dtype=torch.float64)
  22607. predictions are: tensor([[0.5747, 0.3813, 0.6918, 0.2215, 0.3995, 0.2139, 0.5478, 0.5785],
  22608. [0.6207, 0.4144, 0.8046, 0.2775, 0.4792, 0.1880, 0.5633, 0.5005],
  22609. [0.5985, 0.3829, 0.8660, 0.3888, 0.3766, 0.3070, 0.5694, 0.5266],
  22610. [0.5662, 0.3804, 0.7525, 0.2375, 0.3794, 0.3052, 0.6041, 0.5262],
  22611. [0.5694, 0.3678, 0.8466, 0.5783, 0.4286, 0.4979, 0.5485, 0.5595],
  22612. [0.6115, 0.4068, 0.9104, 0.4118, 0.4048, 0.4170, 0.7017, 0.5510],
  22613. [0.6342, 0.4301, 0.8145, 0.2441, 0.5075, 0.1358, 0.6096, 0.5103],
  22614. [0.5751, 0.3817, 0.8953, 0.4413, 0.3908, 0.4685, 0.5948, 0.5521]],
  22615. device='cuda:0', grad_fn=<AddmmBackward>)
  22616. landmarks are: tensor([[[0.6072, 0.4029, 0.7038, 0.2150, 0.3913, 0.2267, 0.5516, 0.5507],
  22617. [0.6277, 0.4083, 0.8350, 0.2717, 0.4563, 0.1800, 0.5918, 0.4878],
  22618. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  22619. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  22620. [0.6192, 0.4128, 0.8512, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  22621. [0.6346, 0.4165, 0.9137, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471],
  22622. [0.6259, 0.4132, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  22623. [0.6224, 0.4061, 0.8988, 0.4300, 0.3837, 0.4750, 0.6112, 0.5483]]],
  22624. device='cuda:0')
  22625. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  22626. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  22627. loss_train: 0.019415606089751236
  22628. step: 29
  22629. running loss: 0.000669503658267284
  22630. Train Steps: 29/90 Loss: 0.0007 torch.Size([8, 600, 800])
  22631. torch.Size([8, 8])
  22632. tensor([[0.6117, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103],
  22633. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  22634. [ nan, nan, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  22635. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5513, 0.5750],
  22636. [ nan, nan, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  22637. [0.6193, 0.4108, 0.7438, 0.2700, 0.3650, 0.3683, 0.6238, 0.5717],
  22638. [0.6228, 0.4004, 0.8750, 0.5250, 0.3825, 0.5233, 0.6362, 0.5000],
  22639. [0.6162, 0.4134, 0.6700, 0.2467, 0.3962, 0.2533, 0.5737, 0.5467]],
  22640. device='cuda:0', dtype=torch.float64)
  22641. predictions are: tensor([[0.6897, 0.4506, 0.6948, 0.2281, 0.4008, 0.2410, 0.5203, 0.5133],
  22642. [0.7451, 0.4733, 0.8854, 0.5929, 0.4019, 0.4944, 0.6341, 0.5262],
  22643. [0.0708, 0.0346, 0.8714, 0.2453, 0.5403, 0.2397, 0.7223, 0.5470],
  22644. [0.6919, 0.4506, 0.7064, 0.2530, 0.4442, 0.1918, 0.5331, 0.5808],
  22645. [0.0642, 0.0430, 0.8695, 0.2792, 0.5340, 0.2003, 0.6744, 0.5507],
  22646. [0.6989, 0.4623, 0.7750, 0.2946, 0.3784, 0.3741, 0.6267, 0.5688],
  22647. [0.7815, 0.4923, 0.9183, 0.5498, 0.4017, 0.5253, 0.6147, 0.5207],
  22648. [0.7112, 0.4725, 0.7207, 0.2791, 0.4066, 0.2466, 0.5596, 0.5634]],
  22649. device='cuda:0', grad_fn=<AddmmBackward>)
  22650. landmarks are: tensor([[[0.6116, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103],
  22651. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  22652. [0.0000, 0.0000, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  22653. [0.6175, 0.3997, 0.6737, 0.2500, 0.4313, 0.1933, 0.5512, 0.5750],
  22654. [0.0000, 0.0000, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  22655. [0.6193, 0.4108, 0.7437, 0.2700, 0.3650, 0.3683, 0.6237, 0.5717],
  22656. [0.6228, 0.4004, 0.8750, 0.5250, 0.3825, 0.5233, 0.6363, 0.5000],
  22657. [0.6162, 0.4134, 0.6700, 0.2467, 0.3963, 0.2533, 0.5738, 0.5467]]],
  22658. device='cuda:0')
  22659. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  22660. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  22661. loss_train: 0.021278868938679807
  22662. step: 30
  22663. running loss: 0.0007092956312893269
  22664.  
  22665. Train Steps: 30/90 Loss: 0.0007 torch.Size([8, 600, 800])
  22666. torch.Size([8, 8])
  22667. tensor([[0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297],
  22668. [0.6212, 0.4171, 0.7875, 0.3633, 0.3813, 0.2933, 0.5675, 0.5700],
  22669. [0.6081, 0.3950, 0.8538, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954],
  22670. [0.6136, 0.4117, 0.8700, 0.5167, 0.4188, 0.5083, 0.5147, 0.5495],
  22671. [0.6164, 0.4076, 0.8838, 0.4117, 0.3713, 0.5550, 0.6238, 0.5350],
  22672. [0.6284, 0.4093, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  22673. [0.6058, 0.3978, 0.8287, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461],
  22674. [0.6284, 0.4127, 0.8538, 0.5867, 0.4363, 0.5083, 0.6038, 0.5433]],
  22675. device='cuda:0', dtype=torch.float64)
  22676. predictions are: tensor([[0.5884, 0.3731, 0.8906, 0.5048, 0.3881, 0.4043, 0.6788, 0.5305],
  22677. [0.6297, 0.4199, 0.7994, 0.3490, 0.3894, 0.2737, 0.6291, 0.5945],
  22678. [0.6037, 0.3834, 0.8661, 0.4282, 0.3997, 0.4526, 0.5736, 0.4915],
  22679. [0.5989, 0.3808, 0.8840, 0.5084, 0.4319, 0.4836, 0.5536, 0.5220],
  22680. [0.6533, 0.4188, 0.8890, 0.3762, 0.3981, 0.5358, 0.6787, 0.5290],
  22681. [0.6001, 0.3861, 0.8929, 0.4558, 0.3773, 0.3699, 0.6534, 0.5160],
  22682. [0.6126, 0.4025, 0.8269, 0.3528, 0.3623, 0.3757, 0.5925, 0.5358],
  22683. [0.6032, 0.3817, 0.8537, 0.5689, 0.4503, 0.4667, 0.6179, 0.5194]],
  22684. device='cuda:0', grad_fn=<AddmmBackward>)
  22685. landmarks are: tensor([[[0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297],
  22686. [0.6212, 0.4171, 0.7875, 0.3633, 0.3812, 0.2933, 0.5675, 0.5700],
  22687. [0.6081, 0.3950, 0.8537, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954],
  22688. [0.6136, 0.4117, 0.8700, 0.5167, 0.4187, 0.5083, 0.5147, 0.5495],
  22689. [0.6164, 0.4076, 0.8838, 0.4117, 0.3713, 0.5550, 0.6237, 0.5350],
  22690. [0.6284, 0.4092, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167],
  22691. [0.6058, 0.3978, 0.8288, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461],
  22692. [0.6284, 0.4127, 0.8537, 0.5867, 0.4363, 0.5083, 0.6037, 0.5433]]],
  22693. device='cuda:0')
  22694. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  22695. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  22696. loss_train: 0.021944493215414695
  22697. step: 31
  22698. running loss: 0.000707886877916603
  22699. Train Steps: 31/90 Loss: 0.0007 torch.Size([8, 600, 800])
  22700. torch.Size([8, 8])
  22701. tensor([[0.6268, 0.4061, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  22702. [0.6161, 0.4024, 0.8838, 0.4583, 0.3688, 0.3733, 0.5311, 0.5344],
  22703. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795],
  22704. [0.6199, 0.4071, 0.7600, 0.2117, 0.4037, 0.2767, 0.6138, 0.5550],
  22705. [0.6153, 0.4117, 0.8688, 0.5167, 0.4895, 0.5647, 0.5524, 0.5136],
  22706. [0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  22707. [0.6161, 0.4055, 0.8675, 0.3867, 0.3713, 0.4033, 0.5195, 0.5162],
  22708. [0.6261, 0.4066, 0.8325, 0.2150, 0.4763, 0.2667, 0.7002, 0.5633]],
  22709. device='cuda:0', dtype=torch.float64)
  22710. predictions are: tensor([[0.6376, 0.4051, 0.8445, 0.2461, 0.4565, 0.1849, 0.6426, 0.5119],
  22711. [0.5506, 0.3638, 0.8830, 0.4665, 0.3697, 0.3586, 0.5518, 0.5292],
  22712. [0.5575, 0.3506, 0.8518, 0.5144, 0.3972, 0.5224, 0.7341, 0.5738],
  22713. [0.5898, 0.3928, 0.7805, 0.2062, 0.4126, 0.2712, 0.6223, 0.5467],
  22714. [0.5837, 0.3926, 0.8823, 0.5018, 0.4803, 0.5185, 0.5774, 0.5102],
  22715. [0.5228, 0.3175, 0.8610, 0.5440, 0.4067, 0.4438, 0.6141, 0.5258],
  22716. [0.5061, 0.3396, 0.8566, 0.3850, 0.3814, 0.3897, 0.5234, 0.5217],
  22717. [0.6307, 0.4236, 0.8495, 0.2266, 0.4794, 0.2508, 0.7256, 0.5409]],
  22718. device='cuda:0', grad_fn=<AddmmBackward>)
  22719. landmarks are: tensor([[[0.6268, 0.4060, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  22720. [0.6161, 0.4024, 0.8838, 0.4583, 0.3688, 0.3733, 0.5311, 0.5344],
  22721. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795],
  22722. [0.6199, 0.4071, 0.7600, 0.2117, 0.4038, 0.2767, 0.6137, 0.5550],
  22723. [0.6154, 0.4117, 0.8687, 0.5167, 0.4895, 0.5647, 0.5524, 0.5136],
  22724. [0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  22725. [0.6161, 0.4055, 0.8675, 0.3867, 0.3713, 0.4033, 0.5195, 0.5162],
  22726. [0.6261, 0.4066, 0.8325, 0.2150, 0.4762, 0.2667, 0.7002, 0.5633]]],
  22727. device='cuda:0')
  22728. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  22729. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  22730. loss_train: 0.022897559145349078
  22731. step: 32
  22732. running loss: 0.0007155487232921587
  22733. Train Steps: 32/90 Loss: 0.0007 torch.Size([8, 600, 800])
  22734. torch.Size([8, 8])
  22735. tensor([[0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  22736. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  22737. [0.6156, 0.4125, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  22738. [0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767],
  22739. [0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  22740. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  22741. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  22742. [0.6058, 0.3978, 0.8287, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461]],
  22743. device='cuda:0', dtype=torch.float64)
  22744. predictions are: tensor([[0.6576, 0.4233, 0.7940, 0.2923, 0.3964, 0.2534, 0.6247, 0.5410],
  22745. [0.5724, 0.3634, 0.8504, 0.4078, 0.3608, 0.4618, 0.6095, 0.5409],
  22746. [0.6195, 0.4325, 0.8775, 0.4804, 0.4494, 0.5989, 0.6058, 0.5275],
  22747. [0.5582, 0.3695, 0.8496, 0.5030, 0.4034, 0.4863, 0.5666, 0.5961],
  22748. [0.5600, 0.3584, 0.8529, 0.3570, 0.3881, 0.2920, 0.6447, 0.4917],
  22749. [0.6144, 0.3968, 0.8115, 0.2539, 0.4409, 0.2646, 0.6695, 0.5365],
  22750. [0.5736, 0.3816, 0.8761, 0.4166, 0.3680, 0.4029, 0.6566, 0.5264],
  22751. [0.5960, 0.3976, 0.8047, 0.3775, 0.3493, 0.4274, 0.5664, 0.5595]],
  22752. device='cuda:0', grad_fn=<AddmmBackward>)
  22753. landmarks are: tensor([[[0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  22754. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  22755. [0.6155, 0.4124, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  22756. [0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767],
  22757. [0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  22758. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  22759. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  22760. [0.6058, 0.3978, 0.8288, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461]]],
  22761. device='cuda:0')
  22762. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  22763. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  22764. loss_train: 0.023489262952352874
  22765. step: 33
  22766. running loss: 0.0007117958470409961
  22767. Train Steps: 33/90 Loss: 0.0007 torch.Size([8, 600, 800])
  22768. torch.Size([8, 8])
  22769. tensor([[0.6272, 0.4071, 0.8738, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  22770. [0.6293, 0.4024, 0.8750, 0.5000, 0.4012, 0.5733, 0.7121, 0.5633],
  22771. [0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  22772. [ nan, nan, 0.8488, 0.2300, 0.5563, 0.2100, 0.7390, 0.5679],
  22773. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  22774. [0.6186, 0.4060, 0.8750, 0.5050, 0.3538, 0.4367, 0.5813, 0.6083],
  22775. [0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  22776. [0.6265, 0.4088, 0.8025, 0.1850, 0.4163, 0.2500, 0.6290, 0.4947]],
  22777. device='cuda:0', dtype=torch.float64)
  22778. predictions are: tensor([[0.6593, 0.4331, 0.8618, 0.5456, 0.3639, 0.4010, 0.5979, 0.4803],
  22779. [0.6536, 0.4266, 0.8628, 0.4762, 0.3945, 0.5987, 0.6864, 0.5626],
  22780. [0.6165, 0.3938, 0.8499, 0.5151, 0.4208, 0.5274, 0.6455, 0.5208],
  22781. [0.1122, 0.0747, 0.8574, 0.2338, 0.5162, 0.2554, 0.7275, 0.5477],
  22782. [0.6545, 0.4259, 0.8812, 0.4674, 0.4134, 0.5498, 0.5881, 0.5204],
  22783. [0.6227, 0.4213, 0.8596, 0.4772, 0.3509, 0.4564, 0.5783, 0.6005],
  22784. [0.6466, 0.4088, 0.8482, 0.5273, 0.3951, 0.4719, 0.5904, 0.5376],
  22785. [0.6413, 0.4213, 0.8073, 0.1971, 0.4224, 0.2603, 0.6283, 0.5116]],
  22786. device='cuda:0', grad_fn=<AddmmBackward>)
  22787. landmarks are: tensor([[[0.6272, 0.4071, 0.8737, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  22788. [0.6293, 0.4024, 0.8750, 0.5000, 0.4013, 0.5733, 0.7121, 0.5633],
  22789. [0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  22790. [0.0000, 0.0000, 0.8487, 0.2300, 0.5562, 0.2100, 0.7390, 0.5679],
  22791. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  22792. [0.6186, 0.4060, 0.8750, 0.5050, 0.3537, 0.4367, 0.5813, 0.6083],
  22793. [0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  22794. [0.6265, 0.4088, 0.8025, 0.1850, 0.4162, 0.2500, 0.6290, 0.4947]]],
  22795. device='cuda:0')
  22796. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  22797. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  22798. loss_train: 0.02405052950780373
  22799. step: 34
  22800. running loss: 0.0007073685149354039
  22801.  
  22802. Train Steps: 34/90 Loss: 0.0007 torch.Size([8, 600, 800])
  22803. torch.Size([8, 8])
  22804. tensor([[0.6081, 0.3950, 0.8538, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954],
  22805. [ nan, nan, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600],
  22806. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  22807. [0.6115, 0.4005, 0.8838, 0.3867, 0.3763, 0.4700, 0.5800, 0.5550],
  22808. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  22809. [0.6262, 0.4052, 0.8888, 0.4700, 0.3675, 0.5117, 0.6350, 0.5233],
  22810. [0.6239, 0.4206, 0.8750, 0.5400, 0.3688, 0.4850, 0.5737, 0.5700],
  22811. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927]],
  22812. device='cuda:0', dtype=torch.float64)
  22813. predictions are: tensor([[0.6548, 0.4210, 0.8627, 0.4514, 0.3951, 0.4915, 0.5448, 0.4980],
  22814. [0.1056, 0.0795, 0.7924, 0.3010, 0.4140, 0.2797, 0.5885, 0.5713],
  22815. [0.6397, 0.4316, 0.8833, 0.4366, 0.3750, 0.3403, 0.6105, 0.5304],
  22816. [0.6451, 0.4234, 0.8708, 0.4025, 0.3790, 0.4824, 0.5783, 0.5597],
  22817. [0.6710, 0.4289, 0.8743, 0.4705, 0.4077, 0.5930, 0.6590, 0.5049],
  22818. [0.6303, 0.4196, 0.8991, 0.4671, 0.3868, 0.5141, 0.6538, 0.5146],
  22819. [0.6531, 0.4207, 0.8667, 0.5391, 0.3833, 0.5041, 0.6007, 0.5643],
  22820. [0.6968, 0.4398, 0.8925, 0.4197, 0.3636, 0.4451, 0.6342, 0.5185]],
  22821. device='cuda:0', grad_fn=<AddmmBackward>)
  22822. landmarks are: tensor([[[0.6081, 0.3950, 0.8537, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954],
  22823. [0.0000, 0.0000, 0.8037, 0.2483, 0.3975, 0.2517, 0.5575, 0.5600],
  22824. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  22825. [0.6115, 0.4005, 0.8838, 0.3867, 0.3762, 0.4700, 0.5800, 0.5550],
  22826. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  22827. [0.6262, 0.4052, 0.8888, 0.4700, 0.3675, 0.5117, 0.6350, 0.5233],
  22828. [0.6239, 0.4206, 0.8750, 0.5400, 0.3688, 0.4850, 0.5738, 0.5700],
  22829. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927]]],
  22830. device='cuda:0')
  22831. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  22832. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  22833. loss_train: 0.0247619836445665
  22834. step: 35
  22835. running loss: 0.0007074852469876143
  22836. Train Steps: 35/90 Loss: 0.0007 torch.Size([8, 600, 800])
  22837. torch.Size([8, 8])
  22838. tensor([[0.6199, 0.4071, 0.7600, 0.2117, 0.4037, 0.2767, 0.6138, 0.5550],
  22839. [0.6156, 0.4125, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  22840. [0.6275, 0.4111, 0.8463, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  22841. [0.6267, 0.4065, 0.8313, 0.2467, 0.4788, 0.1733, 0.6312, 0.5133],
  22842. [0.6124, 0.4075, 0.7696, 0.4153, 0.3475, 0.3767, 0.5157, 0.5427],
  22843. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  22844. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6138, 0.5333],
  22845. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092]],
  22846. device='cuda:0', dtype=torch.float64)
  22847. predictions are: tensor([[0.5583, 0.3769, 0.7909, 0.2271, 0.3903, 0.3063, 0.6082, 0.5704],
  22848. [0.5810, 0.4007, 0.9030, 0.4943, 0.4276, 0.6022, 0.5849, 0.5148],
  22849. [0.5632, 0.3702, 0.8655, 0.2920, 0.4494, 0.2302, 0.6230, 0.5132],
  22850. [0.6146, 0.4037, 0.8404, 0.2668, 0.4663, 0.2104, 0.6326, 0.5392],
  22851. [0.5761, 0.3819, 0.8072, 0.4148, 0.3366, 0.4056, 0.5089, 0.5536],
  22852. [0.5267, 0.3407, 0.7181, 0.2295, 0.4272, 0.1918, 0.5570, 0.5498],
  22853. [0.5548, 0.3801, 0.8961, 0.4657, 0.3552, 0.5163, 0.6202, 0.5391],
  22854. [0.5590, 0.3516, 0.8754, 0.5194, 0.3642, 0.5391, 0.5913, 0.5152]],
  22855. device='cuda:0', grad_fn=<AddmmBackward>)
  22856. landmarks are: tensor([[[0.6199, 0.4071, 0.7600, 0.2117, 0.4038, 0.2767, 0.6137, 0.5550],
  22857. [0.6155, 0.4124, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  22858. [0.6275, 0.4111, 0.8462, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  22859. [0.6266, 0.4065, 0.8313, 0.2467, 0.4787, 0.1733, 0.6313, 0.5133],
  22860. [0.6124, 0.4075, 0.7696, 0.4153, 0.3475, 0.3767, 0.5157, 0.5427],
  22861. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  22862. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6137, 0.5333],
  22863. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092]]],
  22864. device='cuda:0')
  22865. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  22866. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  22867. loss_train: 0.025636436403146945
  22868. step: 36
  22869. running loss: 0.0007121232334207485
  22870. Train Steps: 36/90 Loss: 0.0007 torch.Size([8, 600, 800])
  22871. torch.Size([8, 8])
  22872. tensor([[0.6260, 0.4214, 0.8538, 0.5500, 0.3663, 0.3767, 0.5587, 0.5983],
  22873. [0.6226, 0.4098, 0.8912, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  22874. [0.6111, 0.3995, 0.8788, 0.4567, 0.3813, 0.4833, 0.5450, 0.5700],
  22875. [0.6201, 0.4017, 0.8871, 0.4621, 0.3517, 0.4675, 0.5999, 0.5106],
  22876. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  22877. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6038, 0.4833],
  22878. [0.6339, 0.4102, 0.8588, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089],
  22879. [0.6179, 0.3998, 0.8396, 0.3505, 0.3552, 0.3768, 0.6064, 0.5329]],
  22880. device='cuda:0', dtype=torch.float64)
  22881. predictions are: tensor([[0.5802, 0.3709, 0.8561, 0.5348, 0.3591, 0.4219, 0.5316, 0.5887],
  22882. [0.5363, 0.3499, 0.8974, 0.4331, 0.4099, 0.2639, 0.5611, 0.5330],
  22883. [0.5620, 0.3608, 0.8742, 0.4556, 0.3872, 0.5100, 0.5308, 0.5577],
  22884. [0.5790, 0.3515, 0.8831, 0.4553, 0.3560, 0.4888, 0.5747, 0.5238],
  22885. [0.5446, 0.3546, 0.7188, 0.2502, 0.3910, 0.3107, 0.6053, 0.5817],
  22886. [0.6499, 0.4162, 0.8837, 0.5015, 0.3725, 0.5004, 0.5665, 0.4842],
  22887. [0.6564, 0.4173, 0.8841, 0.3146, 0.4628, 0.2244, 0.6416, 0.4986],
  22888. [0.6285, 0.4154, 0.8605, 0.3613, 0.3461, 0.3816, 0.5735, 0.5304]],
  22889. device='cuda:0', grad_fn=<AddmmBackward>)
  22890. landmarks are: tensor([[[0.6260, 0.4214, 0.8537, 0.5500, 0.3663, 0.3767, 0.5587, 0.5983],
  22891. [0.6226, 0.4098, 0.8913, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  22892. [0.6111, 0.3995, 0.8788, 0.4567, 0.3812, 0.4833, 0.5450, 0.5700],
  22893. [0.6201, 0.4017, 0.8871, 0.4621, 0.3517, 0.4675, 0.5999, 0.5106],
  22894. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  22895. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6037, 0.4833],
  22896. [0.6339, 0.4102, 0.8587, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089],
  22897. [0.6179, 0.3998, 0.8396, 0.3505, 0.3552, 0.3768, 0.6064, 0.5329]]],
  22898. device='cuda:0')
  22899. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  22900. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  22901. loss_train: 0.0263910611247411
  22902. step: 37
  22903. running loss: 0.0007132719222903
  22904. Train Steps: 37/90 Loss: 0.0007 torch.Size([8, 600, 800])
  22905. torch.Size([8, 8])
  22906. tensor([[0.6263, 0.4065, 0.9038, 0.4317, 0.3588, 0.4550, 0.6325, 0.5250],
  22907. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5413, 0.5717],
  22908. [0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767],
  22909. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  22910. [0.6250, 0.4236, 0.8638, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  22911. [0.6109, 0.4036, 0.7188, 0.1750, 0.3850, 0.2550, 0.5863, 0.5567],
  22912. [0.6079, 0.3964, 0.7420, 0.2958, 0.3563, 0.2917, 0.5351, 0.4980],
  22913. [0.6136, 0.4085, 0.6688, 0.2317, 0.3862, 0.2367, 0.5517, 0.5783]],
  22914. device='cuda:0', dtype=torch.float64)
  22915. predictions are: tensor([[0.5990, 0.3769, 0.9140, 0.4488, 0.3714, 0.4375, 0.6351, 0.5066],
  22916. [0.6271, 0.3925, 0.8840, 0.5031, 0.4382, 0.4743, 0.5681, 0.5371],
  22917. [0.6133, 0.3925, 0.8837, 0.5092, 0.4141, 0.4641, 0.5364, 0.5583],
  22918. [0.5439, 0.3584, 0.8692, 0.3327, 0.3847, 0.2954, 0.5554, 0.5394],
  22919. [0.5175, 0.3475, 0.8748, 0.4092, 0.3998, 0.3059, 0.5479, 0.5673],
  22920. [0.5929, 0.3816, 0.7363, 0.2116, 0.3944, 0.2486, 0.5592, 0.5466],
  22921. [0.5964, 0.3888, 0.7736, 0.3105, 0.3548, 0.2785, 0.5229, 0.4887],
  22922. [0.5833, 0.3863, 0.6853, 0.2263, 0.3996, 0.2373, 0.5364, 0.5512]],
  22923. device='cuda:0', grad_fn=<AddmmBackward>)
  22924. landmarks are: tensor([[[0.6263, 0.4065, 0.9038, 0.4317, 0.3587, 0.4550, 0.6325, 0.5250],
  22925. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5412, 0.5717],
  22926. [0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767],
  22927. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  22928. [0.6250, 0.4236, 0.8637, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  22929. [0.6108, 0.4036, 0.7188, 0.1750, 0.3850, 0.2550, 0.5863, 0.5567],
  22930. [0.6079, 0.3964, 0.7420, 0.2958, 0.3562, 0.2917, 0.5351, 0.4980],
  22931. [0.6136, 0.4085, 0.6687, 0.2317, 0.3862, 0.2367, 0.5517, 0.5783]]],
  22932. device='cuda:0')
  22933. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  22934. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  22935. loss_train: 0.027080467421910726
  22936. step: 38
  22937. running loss: 0.0007126438795239665
  22938.  
  22939. Train Steps: 38/90 Loss: 0.0007 torch.Size([8, 600, 800])
  22940. torch.Size([8, 8])
  22941. tensor([[0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  22942. [0.6289, 0.4081, 0.8720, 0.3487, 0.3900, 0.3183, 0.6703, 0.5376],
  22943. [0.6289, 0.4032, 0.8419, 0.5446, 0.4075, 0.5017, 0.6312, 0.5117],
  22944. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  22945. [0.6112, 0.4029, 0.8638, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  22946. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  22947. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  22948. [0.6115, 0.3998, 0.7063, 0.2383, 0.4037, 0.1950, 0.5320, 0.4993]],
  22949. device='cuda:0', dtype=torch.float64)
  22950. predictions are: tensor([[0.5684, 0.3738, 0.7840, 0.1812, 0.4079, 0.1931, 0.6037, 0.5029],
  22951. [0.5567, 0.3666, 0.8722, 0.3481, 0.3623, 0.2860, 0.6309, 0.5371],
  22952. [0.5954, 0.3750, 0.8283, 0.5332, 0.3828, 0.4742, 0.5910, 0.4946],
  22953. [0.5968, 0.3983, 0.8844, 0.4523, 0.4209, 0.5508, 0.5773, 0.5281],
  22954. [0.5925, 0.3815, 0.8645, 0.4634, 0.4638, 0.4607, 0.5284, 0.5370],
  22955. [0.6031, 0.3976, 0.8330, 0.2334, 0.4379, 0.1981, 0.6606, 0.5579],
  22956. [0.5907, 0.3900, 0.8811, 0.4777, 0.4345, 0.5356, 0.5625, 0.5505],
  22957. [0.5687, 0.3794, 0.7224, 0.2112, 0.3896, 0.1612, 0.4978, 0.4984]],
  22958. device='cuda:0', grad_fn=<AddmmBackward>)
  22959. landmarks are: tensor([[[0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  22960. [0.6289, 0.4081, 0.8720, 0.3487, 0.3900, 0.3183, 0.6703, 0.5376],
  22961. [0.6289, 0.4031, 0.8419, 0.5446, 0.4075, 0.5017, 0.6313, 0.5117],
  22962. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  22963. [0.6112, 0.4029, 0.8637, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  22964. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  22965. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517],
  22966. [0.6115, 0.3998, 0.7063, 0.2383, 0.4038, 0.1950, 0.5320, 0.4993]]],
  22967. device='cuda:0')
  22968. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  22969. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  22970. loss_train: 0.027903765891096555
  22971. step: 39
  22972. running loss: 0.0007154811766947835
  22973. Train Steps: 39/90 Loss: 0.0007 torch.Size([8, 600, 800])
  22974. torch.Size([8, 8])
  22975. tensor([[0.6207, 0.4110, 0.8738, 0.5000, 0.4800, 0.5633, 0.6300, 0.5433],
  22976. [0.6125, 0.3974, 0.7725, 0.2517, 0.3538, 0.3317, 0.5887, 0.5500],
  22977. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  22978. [0.6329, 0.4196, 0.9238, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748],
  22979. [0.6276, 0.4235, 0.8888, 0.5333, 0.3800, 0.3117, 0.5427, 0.6164],
  22980. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  22981. [0.6185, 0.4080, 0.8625, 0.3483, 0.3788, 0.2650, 0.5320, 0.5272],
  22982. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283]],
  22983. device='cuda:0', dtype=torch.float64)
  22984. predictions are: tensor([[0.6086, 0.3948, 0.8793, 0.4601, 0.4524, 0.5521, 0.5810, 0.5304],
  22985. [0.6241, 0.4068, 0.7623, 0.2233, 0.3301, 0.3224, 0.5753, 0.5376],
  22986. [0.5699, 0.3948, 0.7303, 0.2490, 0.4237, 0.1971, 0.5398, 0.5759],
  22987. [0.5910, 0.3813, 0.9273, 0.4244, 0.4138, 0.2584, 0.7003, 0.5499],
  22988. [0.6257, 0.4145, 0.8526, 0.4932, 0.3787, 0.3170, 0.5312, 0.5960],
  22989. [0.6313, 0.4051, 0.8683, 0.4292, 0.3831, 0.5610, 0.6172, 0.4978],
  22990. [0.6149, 0.4161, 0.8498, 0.3317, 0.3739, 0.2650, 0.5100, 0.5272],
  22991. [0.6453, 0.4217, 0.8295, 0.5290, 0.3750, 0.4164, 0.5556, 0.5208]],
  22992. device='cuda:0', grad_fn=<AddmmBackward>)
  22993. landmarks are: tensor([[[0.6207, 0.4110, 0.8737, 0.5000, 0.4800, 0.5633, 0.6300, 0.5433],
  22994. [0.6125, 0.3974, 0.7725, 0.2517, 0.3537, 0.3317, 0.5888, 0.5500],
  22995. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  22996. [0.6329, 0.4196, 0.9237, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748],
  22997. [0.6276, 0.4235, 0.8888, 0.5333, 0.3800, 0.3117, 0.5427, 0.6164],
  22998. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  22999. [0.6186, 0.4080, 0.8625, 0.3483, 0.3787, 0.2650, 0.5320, 0.5272],
  23000. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283]]],
  23001. device='cuda:0')
  23002. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  23003. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  23004. loss_train: 0.028414304062607698
  23005. step: 40
  23006. running loss: 0.0007103576015651924
  23007. Train Steps: 40/90 Loss: 0.0007 torch.Size([8, 600, 800])
  23008. torch.Size([8, 8])
  23009. tensor([[0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050],
  23010. [0.6277, 0.4029, 0.8250, 0.2433, 0.4325, 0.2100, 0.6366, 0.5207],
  23011. [0.6091, 0.3997, 0.8314, 0.4334, 0.3788, 0.4550, 0.5213, 0.5656],
  23012. [ nan, nan, 0.7612, 0.3250, 0.4037, 0.2533, 0.5438, 0.5767],
  23013. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  23014. [0.6085, 0.4005, 0.8400, 0.4317, 0.3763, 0.4750, 0.5476, 0.5058],
  23015. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  23016. [0.6132, 0.4118, 0.8200, 0.3633, 0.3563, 0.5400, 0.5787, 0.5136]],
  23017. device='cuda:0', dtype=torch.float64)
  23018. predictions are: tensor([[0.6467, 0.4431, 0.8648, 0.3886, 0.3671, 0.4597, 0.6357, 0.5992],
  23019. [0.6807, 0.4511, 0.8120, 0.2432, 0.4405, 0.1834, 0.6839, 0.5110],
  23020. [0.6613, 0.4373, 0.8320, 0.4391, 0.3748, 0.4331, 0.5437, 0.5581],
  23021. [0.2648, 0.1831, 0.7805, 0.2907, 0.4217, 0.2114, 0.5688, 0.5838],
  23022. [0.7139, 0.4862, 0.8317, 0.5578, 0.3945, 0.4067, 0.6070, 0.5315],
  23023. [0.6573, 0.4524, 0.8424, 0.4128, 0.3647, 0.4494, 0.5517, 0.5118],
  23024. [0.6732, 0.4647, 0.8023, 0.3019, 0.3642, 0.2834, 0.5257, 0.5534],
  23025. [0.6547, 0.4513, 0.8251, 0.3458, 0.3673, 0.4919, 0.6024, 0.5175]],
  23026. device='cuda:0', grad_fn=<AddmmBackward>)
  23027. landmarks are: tensor([[[0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050],
  23028. [0.6277, 0.4029, 0.8250, 0.2433, 0.4325, 0.2100, 0.6366, 0.5207],
  23029. [0.6091, 0.3997, 0.8314, 0.4334, 0.3787, 0.4550, 0.5213, 0.5656],
  23030. [0.0000, 0.0000, 0.7613, 0.3250, 0.4038, 0.2533, 0.5437, 0.5767],
  23031. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  23032. [0.6084, 0.4005, 0.8400, 0.4317, 0.3762, 0.4750, 0.5476, 0.5058],
  23033. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  23034. [0.6132, 0.4118, 0.8200, 0.3633, 0.3562, 0.5400, 0.5787, 0.5136]]],
  23035. device='cuda:0')
  23036. loss_train_step before backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  23037. loss_train_step after backward: tensor(0.0026, device='cuda:0', grad_fn=<MseLossBackward>)
  23038. loss_train: 0.03098702679562848
  23039. step: 41
  23040. running loss: 0.0007557811413567922
  23041. Train Steps: 41/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23042. torch.Size([8, 8])
  23043. tensor([[0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116],
  23044. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  23045. [0.6307, 0.4029, 0.8650, 0.5200, 0.3763, 0.4017, 0.7311, 0.5366],
  23046. [0.6204, 0.4007, 0.7838, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  23047. [ nan, nan, 0.8363, 0.3317, 0.3563, 0.3367, 0.5329, 0.5142],
  23048. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  23049. [0.6126, 0.3954, 0.8538, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  23050. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261]],
  23051. device='cuda:0', dtype=torch.float64)
  23052. predictions are: tensor([[0.7133, 0.4949, 0.8636, 0.4180, 0.4097, 0.5950, 0.5953, 0.5434],
  23053. [0.7292, 0.4689, 0.8951, 0.4901, 0.3736, 0.5000, 0.6631, 0.5112],
  23054. [0.6799, 0.4530, 0.8630, 0.5163, 0.3925, 0.4040, 0.7449, 0.5381],
  23055. [0.6545, 0.4295, 0.7728, 0.1951, 0.4598, 0.1320, 0.6202, 0.5347],
  23056. [0.1823, 0.1531, 0.8082, 0.3131, 0.3363, 0.3036, 0.5638, 0.5512],
  23057. [0.6235, 0.4221, 0.7693, 0.3914, 0.3424, 0.4040, 0.5297, 0.5574],
  23058. [0.7180, 0.4684, 0.8529, 0.5158, 0.4348, 0.4772, 0.5545, 0.5509],
  23059. [0.6849, 0.4614, 0.8755, 0.4868, 0.4155, 0.4720, 0.5656, 0.5568]],
  23060. device='cuda:0', grad_fn=<AddmmBackward>)
  23061. landmarks are: tensor([[[0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116],
  23062. [0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  23063. [0.6307, 0.4029, 0.8650, 0.5200, 0.3762, 0.4017, 0.7311, 0.5366],
  23064. [0.6204, 0.4007, 0.7837, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  23065. [0.0000, 0.0000, 0.8363, 0.3317, 0.3562, 0.3367, 0.5329, 0.5142],
  23066. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  23067. [0.6126, 0.3954, 0.8537, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  23068. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261]]],
  23069. device='cuda:0')
  23070. loss_train_step before backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  23071. loss_train_step after backward: tensor(0.0021, device='cuda:0', grad_fn=<MseLossBackward>)
  23072. loss_train: 0.033130172334495
  23073. step: 42
  23074. running loss: 0.0007888136270117858
  23075.  
  23076. Train Steps: 42/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23077. torch.Size([8, 8])
  23078. tensor([[0.6040, 0.4002, 0.7338, 0.2267, 0.3975, 0.2100, 0.5231, 0.4778],
  23079. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  23080. [0.6245, 0.4100, 0.7762, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  23081. [0.6254, 0.3993, 0.8988, 0.4767, 0.3987, 0.5517, 0.6955, 0.5285],
  23082. [0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950],
  23083. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  23084. [0.6275, 0.4157, 0.8337, 0.5800, 0.3763, 0.4200, 0.5547, 0.6125],
  23085. [0.6203, 0.4073, 0.8189, 0.2398, 0.4400, 0.2054, 0.5929, 0.5501]],
  23086. device='cuda:0', dtype=torch.float64)
  23087. predictions are: tensor([[0.6053, 0.3980, 0.7228, 0.2082, 0.3935, 0.2359, 0.5442, 0.5130],
  23088. [0.5986, 0.3876, 0.7927, 0.2864, 0.3970, 0.2688, 0.5976, 0.5458],
  23089. [0.5661, 0.3729, 0.7679, 0.2413, 0.4758, 0.1661, 0.5989, 0.5626],
  23090. [0.6387, 0.4324, 0.9024, 0.4804, 0.4105, 0.5948, 0.7053, 0.5499],
  23091. [0.6400, 0.4180, 0.8542, 0.5340, 0.4257, 0.5459, 0.5904, 0.5136],
  23092. [0.6048, 0.4021, 0.7705, 0.2558, 0.4410, 0.1949, 0.5803, 0.5480],
  23093. [0.6067, 0.4071, 0.8136, 0.5844, 0.3786, 0.4786, 0.5582, 0.6415],
  23094. [0.5370, 0.3652, 0.8080, 0.2506, 0.4418, 0.2339, 0.6032, 0.5683]],
  23095. device='cuda:0', grad_fn=<AddmmBackward>)
  23096. landmarks are: tensor([[[0.6040, 0.4002, 0.7337, 0.2267, 0.3975, 0.2100, 0.5231, 0.4778],
  23097. [0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  23098. [0.6245, 0.4100, 0.7763, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  23099. [0.6254, 0.3993, 0.8988, 0.4767, 0.3988, 0.5517, 0.6955, 0.5285],
  23100. [0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950],
  23101. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  23102. [0.6275, 0.4157, 0.8338, 0.5800, 0.3762, 0.4200, 0.5547, 0.6125],
  23103. [0.6203, 0.4073, 0.8189, 0.2398, 0.4400, 0.2054, 0.5929, 0.5501]]],
  23104. device='cuda:0')
  23105. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  23106. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  23107. loss_train: 0.033696947430144064
  23108. step: 43
  23109. running loss: 0.0007836499402359084
  23110. Train Steps: 43/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23111. torch.Size([8, 8])
  23112. tensor([[0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  23113. [0.6268, 0.4029, 0.8500, 0.2683, 0.3937, 0.3500, 0.6860, 0.5297],
  23114. [0.6336, 0.4086, 0.8900, 0.3950, 0.3900, 0.2950, 0.6504, 0.5066],
  23115. [0.6234, 0.4179, 0.7825, 0.3450, 0.3813, 0.2867, 0.5675, 0.5617],
  23116. [ nan, nan, 0.8850, 0.3000, 0.5363, 0.2250, 0.7343, 0.5771],
  23117. [0.6064, 0.4019, 0.8650, 0.4517, 0.4037, 0.5367, 0.5703, 0.5609],
  23118. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  23119. [0.6206, 0.4001, 0.8900, 0.3933, 0.3588, 0.3567, 0.5837, 0.5083]],
  23120. device='cuda:0', dtype=torch.float64)
  23121. predictions are: tensor([[ 0.6887, 0.4348, 0.8442, 0.4089, 0.3687, 0.3939, 0.6161, 0.4755],
  23122. [ 0.6412, 0.4167, 0.8052, 0.2971, 0.3953, 0.3711, 0.6636, 0.5303],
  23123. [ 0.7081, 0.4484, 0.8750, 0.4183, 0.4020, 0.3306, 0.6459, 0.5059],
  23124. [ 0.5995, 0.4066, 0.7609, 0.3685, 0.3983, 0.3026, 0.5671, 0.5638],
  23125. [-0.0731, -0.0399, 0.8660, 0.2882, 0.5323, 0.2049, 0.7043, 0.6052],
  23126. [ 0.6184, 0.4114, 0.8384, 0.4840, 0.4189, 0.5765, 0.5598, 0.5455],
  23127. [ 0.7184, 0.4611, 0.8971, 0.4411, 0.4320, 0.4057, 0.7127, 0.5755],
  23128. [ 0.6672, 0.4277, 0.8704, 0.4425, 0.3659, 0.3813, 0.5603, 0.5099]],
  23129. device='cuda:0', grad_fn=<AddmmBackward>)
  23130. landmarks are: tensor([[[0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  23131. [0.6268, 0.4029, 0.8500, 0.2683, 0.3938, 0.3500, 0.6860, 0.5297],
  23132. [0.6336, 0.4086, 0.8900, 0.3950, 0.3900, 0.2950, 0.6504, 0.5066],
  23133. [0.6234, 0.4179, 0.7825, 0.3450, 0.3812, 0.2867, 0.5675, 0.5617],
  23134. [0.0000, 0.0000, 0.8850, 0.3000, 0.5362, 0.2250, 0.7343, 0.5771],
  23135. [0.6064, 0.4019, 0.8650, 0.4517, 0.4038, 0.5367, 0.5703, 0.5609],
  23136. [0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  23137. [0.6206, 0.4001, 0.8900, 0.3933, 0.3587, 0.3567, 0.5838, 0.5083]]],
  23138. device='cuda:0')
  23139. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  23140. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  23141. loss_train: 0.0345825739641441
  23142. step: 44
  23143. running loss: 0.0007859675900941842
  23144. Train Steps: 44/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23145. torch.Size([8, 8])
  23146. tensor([[0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  23147. [0.6311, 0.3998, 0.7975, 0.5767, 0.3838, 0.4850, 0.7327, 0.5343],
  23148. [0.6277, 0.4103, 0.8087, 0.5717, 0.4188, 0.4750, 0.5663, 0.6083],
  23149. [ nan, nan, 0.7625, 0.2433, 0.3713, 0.2867, 0.5235, 0.5220],
  23150. [0.6203, 0.4096, 0.8862, 0.4267, 0.3538, 0.4117, 0.6025, 0.5650],
  23151. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5963, 0.6217],
  23152. [0.6299, 0.4303, 0.7963, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  23153. [0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717]],
  23154. device='cuda:0', dtype=torch.float64)
  23155. predictions are: tensor([[0.6366, 0.4079, 0.8650, 0.5747, 0.3955, 0.5086, 0.6662, 0.5088],
  23156. [0.6423, 0.4086, 0.8256, 0.5419, 0.3860, 0.4891, 0.7217, 0.5036],
  23157. [0.6544, 0.4121, 0.8499, 0.5703, 0.4315, 0.4819, 0.5847, 0.5896],
  23158. [0.1003, 0.0558, 0.7747, 0.2473, 0.3955, 0.2725, 0.5176, 0.5151],
  23159. [0.5841, 0.3731, 0.9133, 0.4263, 0.3761, 0.4109, 0.6111, 0.5613],
  23160. [0.5823, 0.3797, 0.7245, 0.2530, 0.4314, 0.2382, 0.6005, 0.5880],
  23161. [0.6369, 0.4191, 0.8270, 0.3852, 0.4887, 0.2643, 0.5677, 0.5828],
  23162. [0.6239, 0.3974, 0.7727, 0.2497, 0.3892, 0.2870, 0.6044, 0.5587]],
  23163. device='cuda:0', grad_fn=<AddmmBackward>)
  23164. landmarks are: tensor([[[0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  23165. [0.6311, 0.3998, 0.7975, 0.5767, 0.3837, 0.4850, 0.7327, 0.5343],
  23166. [0.6277, 0.4103, 0.8087, 0.5717, 0.4187, 0.4750, 0.5663, 0.6083],
  23167. [0.0000, 0.0000, 0.7625, 0.2433, 0.3713, 0.2867, 0.5235, 0.5220],
  23168. [0.6203, 0.4096, 0.8863, 0.4267, 0.3537, 0.4117, 0.6025, 0.5650],
  23169. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5962, 0.6217],
  23170. [0.6299, 0.4303, 0.7962, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  23171. [0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717]]],
  23172. device='cuda:0')
  23173. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  23174. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  23175. loss_train: 0.03513337757613044
  23176. step: 45
  23177. running loss: 0.0007807417239140098
  23178. Train Steps: 45/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23179. torch.Size([8, 8])
  23180. tensor([[0.6179, 0.4040, 0.7412, 0.1850, 0.3825, 0.2783, 0.5837, 0.5600],
  23181. [ nan, nan, 0.7335, 0.2569, 0.3788, 0.2667, 0.5066, 0.5578],
  23182. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  23183. [0.6277, 0.4083, 0.8350, 0.2717, 0.4562, 0.1800, 0.5918, 0.4878],
  23184. [0.6201, 0.4102, 0.7288, 0.2417, 0.4150, 0.2383, 0.6100, 0.5500],
  23185. [0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  23186. [0.6254, 0.4076, 0.8700, 0.3267, 0.4150, 0.3083, 0.7050, 0.5609],
  23187. [0.6193, 0.4108, 0.7438, 0.2700, 0.3650, 0.3683, 0.6238, 0.5717]],
  23188. device='cuda:0', dtype=torch.float64)
  23189. predictions are: tensor([[0.6316, 0.4101, 0.7578, 0.2223, 0.4062, 0.2772, 0.5956, 0.5518],
  23190. [0.0738, 0.0471, 0.7439, 0.2771, 0.4110, 0.2470, 0.5182, 0.5831],
  23191. [0.6456, 0.3919, 0.9053, 0.5674, 0.3980, 0.4672, 0.6566, 0.4862],
  23192. [0.5411, 0.3353, 0.8373, 0.2930, 0.4931, 0.2127, 0.6068, 0.4882],
  23193. [0.6293, 0.3939, 0.7605, 0.2568, 0.4241, 0.2575, 0.6009, 0.5605],
  23194. [0.6253, 0.3865, 0.9118, 0.4759, 0.4028, 0.3663, 0.5960, 0.5789],
  23195. [0.6018, 0.3726, 0.8972, 0.3387, 0.4325, 0.2984, 0.7147, 0.5485],
  23196. [0.5627, 0.3518, 0.7641, 0.2984, 0.3925, 0.3745, 0.6367, 0.5628]],
  23197. device='cuda:0', grad_fn=<AddmmBackward>)
  23198. landmarks are: tensor([[[0.6179, 0.4040, 0.7412, 0.1850, 0.3825, 0.2783, 0.5838, 0.5600],
  23199. [0.0000, 0.0000, 0.7335, 0.2569, 0.3787, 0.2667, 0.5066, 0.5578],
  23200. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  23201. [0.6277, 0.4083, 0.8350, 0.2717, 0.4563, 0.1800, 0.5918, 0.4878],
  23202. [0.6201, 0.4102, 0.7287, 0.2417, 0.4150, 0.2383, 0.6100, 0.5500],
  23203. [0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  23204. [0.6254, 0.4076, 0.8700, 0.3267, 0.4150, 0.3083, 0.7050, 0.5609],
  23205. [0.6193, 0.4108, 0.7437, 0.2700, 0.3650, 0.3683, 0.6237, 0.5717]]],
  23206. device='cuda:0')
  23207. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  23208. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  23209. loss_train: 0.03593642565829214
  23210. step: 46
  23211. running loss: 0.0007812266447454813
  23212.  
  23213. Train Steps: 46/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23214. torch.Size([8, 8])
  23215. tensor([[0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683],
  23216. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  23217. [0.6229, 0.4066, 0.7612, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  23218. [0.6151, 0.4125, 0.8738, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  23219. [0.6187, 0.4104, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  23220. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  23221. [0.6197, 0.4091, 0.8800, 0.4783, 0.3538, 0.4767, 0.5950, 0.5550],
  23222. [0.6043, 0.4022, 0.6887, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136]],
  23223. device='cuda:0', dtype=torch.float64)
  23224. predictions are: tensor([[0.5747, 0.3567, 0.8971, 0.5689, 0.3822, 0.3893, 0.5378, 0.5800],
  23225. [0.6002, 0.3645, 0.8841, 0.5117, 0.4197, 0.4937, 0.5724, 0.5320],
  23226. [0.5817, 0.3587, 0.7647, 0.2957, 0.4275, 0.2383, 0.5902, 0.5493],
  23227. [0.5815, 0.3735, 0.8794, 0.4650, 0.3519, 0.4093, 0.5114, 0.5625],
  23228. [0.5597, 0.3534, 0.7042, 0.2082, 0.3968, 0.2579, 0.5914, 0.5546],
  23229. [0.5851, 0.3438, 0.8778, 0.2913, 0.5018, 0.2339, 0.7276, 0.5327],
  23230. [0.5661, 0.3607, 0.9040, 0.5104, 0.3726, 0.4870, 0.6096, 0.5701],
  23231. [0.5549, 0.3462, 0.6927, 0.2035, 0.3760, 0.2776, 0.5572, 0.5165]],
  23232. device='cuda:0', grad_fn=<AddmmBackward>)
  23233. landmarks are: tensor([[[0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683],
  23234. [0.6086, 0.3940, 0.8712, 0.4783, 0.4025, 0.4900, 0.5498, 0.5390],
  23235. [0.6229, 0.4066, 0.7613, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  23236. [0.6151, 0.4125, 0.8737, 0.4417, 0.3575, 0.3783, 0.5138, 0.5483],
  23237. [0.6187, 0.4103, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  23238. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  23239. [0.6197, 0.4091, 0.8800, 0.4783, 0.3537, 0.4767, 0.5950, 0.5550],
  23240. [0.6043, 0.4022, 0.6888, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136]]],
  23241. device='cuda:0')
  23242. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  23243. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  23244. loss_train: 0.03665601489774417
  23245. step: 47
  23246. running loss: 0.0007799152105903015
  23247. Train Steps: 47/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23248. torch.Size([8, 8])
  23249. tensor([[0.6286, 0.4086, 0.8408, 0.2801, 0.4163, 0.2800, 0.6725, 0.5393],
  23250. [0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297],
  23251. [ nan, nan, 0.7515, 0.2708, 0.3987, 0.2267, 0.5162, 0.5567],
  23252. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  23253. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  23254. [0.6132, 0.4118, 0.8200, 0.3633, 0.3563, 0.5400, 0.5787, 0.5136],
  23255. [0.6364, 0.4165, 0.9088, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  23256. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650]],
  23257. device='cuda:0', dtype=torch.float64)
  23258. predictions are: tensor([[ 0.6047, 0.3709, 0.8453, 0.2752, 0.4076, 0.2719, 0.6861, 0.5352],
  23259. [ 0.6716, 0.4370, 0.8774, 0.5461, 0.3600, 0.4168, 0.6446, 0.5402],
  23260. [-0.0229, -0.0220, 0.7496, 0.2566, 0.3989, 0.2077, 0.5066, 0.5667],
  23261. [ 0.5969, 0.3846, 0.8858, 0.5054, 0.4045, 0.5154, 0.6771, 0.5271],
  23262. [ 0.6160, 0.4008, 0.8757, 0.4793, 0.4367, 0.4673, 0.5788, 0.5784],
  23263. [ 0.6028, 0.3947, 0.8308, 0.3613, 0.3697, 0.4870, 0.5969, 0.5300],
  23264. [ 0.6021, 0.3922, 0.9019, 0.4375, 0.4062, 0.2941, 0.6574, 0.5230],
  23265. [ 0.5390, 0.3624, 0.7708, 0.2638, 0.3830, 0.2645, 0.5742, 0.5576]],
  23266. device='cuda:0', grad_fn=<AddmmBackward>)
  23267. landmarks are: tensor([[[0.6286, 0.4086, 0.8408, 0.2801, 0.4162, 0.2800, 0.6725, 0.5393],
  23268. [0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297],
  23269. [0.0000, 0.0000, 0.7515, 0.2708, 0.3988, 0.2267, 0.5163, 0.5567],
  23270. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  23271. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  23272. [0.6132, 0.4118, 0.8200, 0.3633, 0.3562, 0.5400, 0.5787, 0.5136],
  23273. [0.6364, 0.4165, 0.9087, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  23274. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650]]],
  23275. device='cuda:0')
  23276. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  23277. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  23278. loss_train: 0.03708239090337884
  23279. step: 48
  23280. running loss: 0.0007725498104870591
  23281. Train Steps: 48/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23282. torch.Size([8, 8])
  23283. tensor([[0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  23284. [0.6109, 0.4041, 0.6975, 0.3167, 0.3513, 0.3383, 0.5153, 0.5319],
  23285. [0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5988, 0.5700],
  23286. [0.6307, 0.4045, 0.8025, 0.5833, 0.3775, 0.4867, 0.6892, 0.5459],
  23287. [0.6286, 0.3977, 0.9038, 0.4733, 0.3900, 0.4150, 0.7074, 0.5320],
  23288. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  23289. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  23290. [0.6204, 0.4049, 0.7975, 0.2700, 0.3937, 0.2567, 0.5700, 0.5183]],
  23291. device='cuda:0', dtype=torch.float64)
  23292. predictions are: tensor([[0.5392, 0.3393, 0.8966, 0.4762, 0.4050, 0.5081, 0.6086, 0.5354],
  23293. [0.5919, 0.4023, 0.7150, 0.2929, 0.3263, 0.3258, 0.4798, 0.5289],
  23294. [0.4935, 0.3126, 0.7921, 0.2691, 0.3399, 0.3424, 0.5831, 0.5745],
  23295. [0.6140, 0.4021, 0.8272, 0.5483, 0.3646, 0.4634, 0.6519, 0.5380],
  23296. [0.5755, 0.3592, 0.9023, 0.4503, 0.3514, 0.4043, 0.6516, 0.5154],
  23297. [0.5746, 0.3584, 0.8680, 0.2639, 0.4762, 0.2027, 0.6777, 0.5342],
  23298. [0.5433, 0.3501, 0.8058, 0.2698, 0.3872, 0.2175, 0.5622, 0.5379],
  23299. [0.6003, 0.4070, 0.7896, 0.2489, 0.3604, 0.2387, 0.5386, 0.5305]],
  23300. device='cuda:0', grad_fn=<AddmmBackward>)
  23301. landmarks are: tensor([[[0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  23302. [0.6109, 0.4041, 0.6975, 0.3167, 0.3512, 0.3383, 0.5153, 0.5319],
  23303. [0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5987, 0.5700],
  23304. [0.6307, 0.4045, 0.8025, 0.5833, 0.3775, 0.4867, 0.6892, 0.5459],
  23305. [0.6286, 0.3977, 0.9038, 0.4733, 0.3900, 0.4150, 0.7074, 0.5320],
  23306. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  23307. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391],
  23308. [0.6204, 0.4049, 0.7975, 0.2700, 0.3938, 0.2567, 0.5700, 0.5183]]],
  23309. device='cuda:0')
  23310. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  23311. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  23312. loss_train: 0.03830938627652358
  23313. step: 49
  23314. running loss: 0.0007818242097249711
  23315. Train Steps: 49/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23316. torch.Size([8, 8])
  23317. tensor([[0.6199, 0.4102, 0.8950, 0.4417, 0.4012, 0.5367, 0.6112, 0.5967],
  23318. [0.6210, 0.4164, 0.7202, 0.2930, 0.4025, 0.2483, 0.5687, 0.5567],
  23319. [0.6346, 0.4144, 0.9088, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  23320. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  23321. [ nan, nan, 0.7850, 0.2700, 0.4288, 0.1717, 0.5199, 0.4999],
  23322. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  23323. [0.6178, 0.4012, 0.8900, 0.4495, 0.3891, 0.5617, 0.5972, 0.4879],
  23324. [0.6228, 0.4119, 0.7938, 0.2233, 0.4674, 0.1773, 0.6188, 0.5433]],
  23325. device='cuda:0', dtype=torch.float64)
  23326. predictions are: tensor([[0.6319, 0.4052, 0.9060, 0.4367, 0.3820, 0.5091, 0.6136, 0.5904],
  23327. [0.5923, 0.3935, 0.7427, 0.2612, 0.3888, 0.2576, 0.5829, 0.5510],
  23328. [0.6742, 0.4459, 0.9098, 0.4459, 0.3704, 0.4277, 0.7321, 0.5875],
  23329. [0.6198, 0.4075, 0.8555, 0.2908, 0.3433, 0.2992, 0.5808, 0.5473],
  23330. [0.0068, 0.0069, 0.7769, 0.2453, 0.4023, 0.1820, 0.5361, 0.5231],
  23331. [0.5984, 0.4031, 0.7380, 0.3369, 0.4850, 0.1821, 0.5506, 0.6082],
  23332. [0.6052, 0.4033, 0.8811, 0.4277, 0.3689, 0.5446, 0.6375, 0.4999],
  23333. [0.5963, 0.3933, 0.7923, 0.1988, 0.4512, 0.1713, 0.6343, 0.5319]],
  23334. device='cuda:0', grad_fn=<AddmmBackward>)
  23335. landmarks are: tensor([[[0.6199, 0.4102, 0.8950, 0.4417, 0.4013, 0.5367, 0.6112, 0.5967],
  23336. [0.6210, 0.4164, 0.7202, 0.2930, 0.4025, 0.2483, 0.5688, 0.5567],
  23337. [0.6346, 0.4144, 0.9087, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  23338. [0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  23339. [0.0000, 0.0000, 0.7850, 0.2700, 0.4288, 0.1717, 0.5199, 0.4999],
  23340. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  23341. [0.6178, 0.4012, 0.8900, 0.4495, 0.3891, 0.5617, 0.5972, 0.4879],
  23342. [0.6228, 0.4119, 0.7937, 0.2233, 0.4674, 0.1773, 0.6187, 0.5433]]],
  23343. device='cuda:0')
  23344. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  23345. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  23346. loss_train: 0.03863833627838176
  23347. step: 50
  23348. running loss: 0.0007727667255676352
  23349.  
  23350. Train Steps: 50/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23351. torch.Size([8, 8])
  23352. tensor([[0.6173, 0.4114, 0.7325, 0.2500, 0.4213, 0.1917, 0.5338, 0.5700],
  23353. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  23354. [ nan, nan, 0.8850, 0.2817, 0.5112, 0.2183, 0.7184, 0.5436],
  23355. [0.6257, 0.4167, 0.8775, 0.3433, 0.3563, 0.4133, 0.6200, 0.5667],
  23356. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  23357. [0.6284, 0.4127, 0.8538, 0.5867, 0.4363, 0.5083, 0.6038, 0.5433],
  23358. [0.6196, 0.4094, 0.7562, 0.2817, 0.3937, 0.3183, 0.6013, 0.6183],
  23359. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5988, 0.5667]],
  23360. device='cuda:0', dtype=torch.float64)
  23361. predictions are: tensor([[ 0.6040, 0.4282, 0.7308, 0.2059, 0.4077, 0.2071, 0.5314, 0.5556],
  23362. [ 0.6273, 0.4338, 0.6940, 0.2815, 0.3589, 0.2807, 0.5497, 0.5789],
  23363. [-0.0056, -0.0037, 0.8923, 0.2670, 0.4995, 0.2198, 0.7089, 0.5389],
  23364. [ 0.6117, 0.4112, 0.8859, 0.3281, 0.3341, 0.4142, 0.6316, 0.5609],
  23365. [ 0.6467, 0.4352, 0.7644, 0.1785, 0.4511, 0.1633, 0.5915, 0.4808],
  23366. [ 0.6760, 0.4614, 0.8418, 0.5814, 0.4076, 0.4921, 0.5650, 0.5409],
  23367. [ 0.6194, 0.4256, 0.7702, 0.2751, 0.3781, 0.3243, 0.6166, 0.6052],
  23368. [ 0.5795, 0.4023, 0.8600, 0.3549, 0.3603, 0.3462, 0.6070, 0.5654]],
  23369. device='cuda:0', grad_fn=<AddmmBackward>)
  23370. landmarks are: tensor([[[0.6173, 0.4114, 0.7325, 0.2500, 0.4212, 0.1917, 0.5337, 0.5700],
  23371. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  23372. [0.0000, 0.0000, 0.8850, 0.2817, 0.5113, 0.2183, 0.7184, 0.5436],
  23373. [0.6257, 0.4167, 0.8775, 0.3433, 0.3562, 0.4133, 0.6200, 0.5667],
  23374. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  23375. [0.6284, 0.4127, 0.8537, 0.5867, 0.4363, 0.5083, 0.6037, 0.5433],
  23376. [0.6196, 0.4094, 0.7563, 0.2817, 0.3938, 0.3183, 0.6012, 0.6183],
  23377. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5987, 0.5667]]],
  23378. device='cuda:0')
  23379. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  23380. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  23381. loss_train: 0.03896989063650835
  23382. step: 51
  23383. running loss: 0.0007641155026766344
  23384. Train Steps: 51/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23385. torch.Size([8, 8])
  23386. tensor([[0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  23387. [0.6160, 0.4093, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  23388. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  23389. [0.6069, 0.3975, 0.8625, 0.5083, 0.4388, 0.5483, 0.5650, 0.4967],
  23390. [0.6256, 0.4199, 0.8638, 0.5800, 0.3987, 0.4383, 0.5600, 0.5950],
  23391. [0.6230, 0.4113, 0.7213, 0.1983, 0.4325, 0.2367, 0.6262, 0.5400],
  23392. [0.6270, 0.4267, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  23393. [0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332]],
  23394. device='cuda:0', dtype=torch.float64)
  23395. predictions are: tensor([[0.6073, 0.4136, 0.8816, 0.4450, 0.3556, 0.4199, 0.5715, 0.5052],
  23396. [0.6147, 0.4262, 0.8284, 0.4442, 0.3397, 0.4541, 0.5100, 0.5697],
  23397. [0.6346, 0.4331, 0.7867, 0.2099, 0.4025, 0.2692, 0.6605, 0.5185],
  23398. [0.5712, 0.4087, 0.8530, 0.5211, 0.4311, 0.5627, 0.5652, 0.5148],
  23399. [0.5833, 0.4068, 0.8546, 0.5685, 0.3776, 0.4327, 0.5370, 0.6009],
  23400. [0.5960, 0.4159, 0.7190, 0.1794, 0.4325, 0.2432, 0.6347, 0.5452],
  23401. [0.5953, 0.4253, 0.6915, 0.2885, 0.4620, 0.1969, 0.5367, 0.5910],
  23402. [0.5961, 0.4056, 0.8904, 0.4810, 0.3547, 0.4024, 0.6478, 0.5261]],
  23403. device='cuda:0', grad_fn=<AddmmBackward>)
  23404. landmarks are: tensor([[[0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  23405. [0.6160, 0.4092, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  23406. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  23407. [0.6069, 0.3975, 0.8625, 0.5083, 0.4387, 0.5483, 0.5650, 0.4967],
  23408. [0.6256, 0.4199, 0.8637, 0.5800, 0.3988, 0.4383, 0.5600, 0.5950],
  23409. [0.6230, 0.4113, 0.7212, 0.1983, 0.4325, 0.2367, 0.6263, 0.5400],
  23410. [0.6270, 0.4266, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  23411. [0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332]]],
  23412. device='cuda:0')
  23413. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  23414. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  23415. loss_train: 0.03928698522213381
  23416. step: 52
  23417. running loss: 0.0007555189465794963
  23418. Train Steps: 52/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23419. torch.Size([8, 8])
  23420. tensor([[0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  23421. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395],
  23422. [0.6075, 0.4000, 0.8513, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  23423. [0.6149, 0.4054, 0.6713, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  23424. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  23425. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  23426. [0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  23427. [ nan, nan, 0.7240, 0.2722, 0.3900, 0.2567, 0.5168, 0.5933]],
  23428. device='cuda:0', dtype=torch.float64)
  23429. predictions are: tensor([[0.6855, 0.4642, 0.8893, 0.5334, 0.3714, 0.3921, 0.6544, 0.5189],
  23430. [0.7078, 0.4839, 0.7207, 0.2155, 0.4106, 0.2417, 0.6140, 0.5604],
  23431. [0.5593, 0.3895, 0.8413, 0.5402, 0.4644, 0.5556, 0.5571, 0.5589],
  23432. [0.6221, 0.4285, 0.6530, 0.2349, 0.4115, 0.2204, 0.5106, 0.5732],
  23433. [0.6814, 0.4617, 0.7495, 0.1913, 0.4757, 0.1683, 0.6202, 0.5073],
  23434. [0.6239, 0.4233, 0.7985, 0.2182, 0.5043, 0.1667, 0.6382, 0.5138],
  23435. [0.6505, 0.4486, 0.8450, 0.4286, 0.3794, 0.5564, 0.5866, 0.5351],
  23436. [0.0618, 0.0508, 0.6994, 0.2725, 0.3897, 0.2711, 0.5051, 0.5950]],
  23437. device='cuda:0', grad_fn=<AddmmBackward>)
  23438. landmarks are: tensor([[[0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  23439. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395],
  23440. [0.6075, 0.4000, 0.8512, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  23441. [0.6149, 0.4054, 0.6712, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  23442. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  23443. [0.6275, 0.4071, 0.8075, 0.2100, 0.4700, 0.1533, 0.6148, 0.4834],
  23444. [0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  23445. [0.0000, 0.0000, 0.7240, 0.2722, 0.3900, 0.2567, 0.5168, 0.5933]]],
  23446. device='cuda:0')
  23447. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  23448. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  23449. loss_train: 0.040181065662181936
  23450. step: 53
  23451. running loss: 0.0007581333143807913
  23452. Train Steps: 53/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23453. torch.Size([8, 8])
  23454. tensor([[0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  23455. [0.6265, 0.4091, 0.8950, 0.3533, 0.3600, 0.3967, 0.6295, 0.4901],
  23456. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5988, 0.5667],
  23457. [0.6216, 0.4167, 0.8588, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667],
  23458. [0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  23459. [0.6203, 0.4021, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  23460. [ nan, nan, 0.7612, 0.3250, 0.4037, 0.2533, 0.5438, 0.5767],
  23461. [0.6152, 0.4131, 0.6863, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305]],
  23462. device='cuda:0', dtype=torch.float64)
  23463. predictions are: tensor([[0.7365, 0.4962, 0.7534, 0.2470, 0.4813, 0.1644, 0.5918, 0.5210],
  23464. [0.6629, 0.4288, 0.8927, 0.3460, 0.3785, 0.3965, 0.6518, 0.5165],
  23465. [0.6685, 0.4532, 0.8460, 0.3589, 0.3941, 0.3674, 0.6022, 0.5691],
  23466. [0.6363, 0.4391, 0.8478, 0.5446, 0.4133, 0.5225, 0.5803, 0.5728],
  23467. [0.6585, 0.4363, 0.8502, 0.3563, 0.4291, 0.2718, 0.6160, 0.4798],
  23468. [0.6618, 0.4396, 0.8653, 0.5108, 0.3825, 0.4054, 0.5755, 0.5395],
  23469. [0.0434, 0.0198, 0.7778, 0.3281, 0.4289, 0.2592, 0.5454, 0.5704],
  23470. [0.6542, 0.4441, 0.6695, 0.2451, 0.3776, 0.3397, 0.5739, 0.5523]],
  23471. device='cuda:0', grad_fn=<AddmmBackward>)
  23472. landmarks are: tensor([[[0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  23473. [0.6265, 0.4091, 0.8950, 0.3533, 0.3600, 0.3967, 0.6295, 0.4901],
  23474. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5987, 0.5667],
  23475. [0.6216, 0.4167, 0.8587, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667],
  23476. [0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  23477. [0.6203, 0.4020, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405],
  23478. [0.0000, 0.0000, 0.7613, 0.3250, 0.4038, 0.2533, 0.5437, 0.5767],
  23479. [0.6152, 0.4131, 0.6862, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305]]],
  23480. device='cuda:0')
  23481. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  23482. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  23483. loss_train: 0.04089540946006309
  23484. step: 54
  23485. running loss: 0.0007573223974085758
  23486.  
  23487. Train Steps: 54/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23488. torch.Size([8, 8])
  23489. tensor([[0.6265, 0.4091, 0.8950, 0.3533, 0.3600, 0.3967, 0.6295, 0.4901],
  23490. [0.6246, 0.4028, 0.8738, 0.4867, 0.4088, 0.5667, 0.6362, 0.5200],
  23491. [ nan, nan, 0.7192, 0.2346, 0.4037, 0.2050, 0.5138, 0.5650],
  23492. [0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500],
  23493. [0.6250, 0.4110, 0.7238, 0.2067, 0.4263, 0.1883, 0.5625, 0.5633],
  23494. [0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  23495. [0.6133, 0.4094, 0.8495, 0.4028, 0.3588, 0.3200, 0.5003, 0.5407],
  23496. [0.6165, 0.4106, 0.7575, 0.1733, 0.3838, 0.2650, 0.5680, 0.5116]],
  23497. device='cuda:0', dtype=torch.float64)
  23498. predictions are: tensor([[0.6796, 0.4268, 0.9074, 0.3819, 0.3924, 0.3890, 0.6588, 0.5231],
  23499. [0.6585, 0.4226, 0.8758, 0.5104, 0.4502, 0.5876, 0.6470, 0.5290],
  23500. [0.0220, 0.0207, 0.7197, 0.2960, 0.4424, 0.2298, 0.4976, 0.5429],
  23501. [0.6749, 0.4446, 0.7745, 0.2926, 0.3850, 0.4026, 0.6171, 0.5571],
  23502. [0.6448, 0.4231, 0.7123, 0.2585, 0.4646, 0.1773, 0.5542, 0.5607],
  23503. [0.6679, 0.4259, 0.8739, 0.3943, 0.3975, 0.3687, 0.6295, 0.4941],
  23504. [0.6708, 0.4429, 0.8537, 0.4344, 0.3786, 0.3161, 0.5023, 0.5432],
  23505. [0.7364, 0.4848, 0.7535, 0.2071, 0.4139, 0.2637, 0.5635, 0.5158]],
  23506. device='cuda:0', grad_fn=<AddmmBackward>)
  23507. landmarks are: tensor([[[0.6265, 0.4091, 0.8950, 0.3533, 0.3600, 0.3967, 0.6295, 0.4901],
  23508. [0.6246, 0.4028, 0.8737, 0.4867, 0.4087, 0.5667, 0.6363, 0.5200],
  23509. [0.0000, 0.0000, 0.7192, 0.2346, 0.4038, 0.2050, 0.5138, 0.5650],
  23510. [0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500],
  23511. [0.6250, 0.4110, 0.7237, 0.2067, 0.4263, 0.1883, 0.5625, 0.5633],
  23512. [0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  23513. [0.6133, 0.4094, 0.8495, 0.4028, 0.3587, 0.3200, 0.5003, 0.5407],
  23514. [0.6165, 0.4106, 0.7575, 0.1733, 0.3837, 0.2650, 0.5680, 0.5116]]],
  23515. device='cuda:0')
  23516. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  23517. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  23518. loss_train: 0.04186867403041106
  23519. step: 55
  23520. running loss: 0.0007612486187347464
  23521. Train Steps: 55/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23522. torch.Size([8, 8])
  23523. tensor([[0.6277, 0.4083, 0.8350, 0.2717, 0.4562, 0.1800, 0.5918, 0.4878],
  23524. [0.6210, 0.4164, 0.7202, 0.2930, 0.4025, 0.2483, 0.5687, 0.5567],
  23525. [0.6193, 0.4034, 0.7757, 0.2347, 0.3733, 0.2919, 0.5930, 0.4926],
  23526. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  23527. [0.6231, 0.3973, 0.8650, 0.3950, 0.3625, 0.3183, 0.5837, 0.5167],
  23528. [0.6223, 0.4028, 0.8988, 0.4200, 0.3763, 0.5733, 0.6375, 0.5167],
  23529. [0.6185, 0.4098, 0.8838, 0.4900, 0.4537, 0.5800, 0.6288, 0.5400],
  23530. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586]],
  23531. device='cuda:0', dtype=torch.float64)
  23532. predictions are: tensor([[0.6891, 0.4279, 0.8262, 0.2814, 0.4698, 0.1748, 0.5825, 0.4751],
  23533. [0.6379, 0.4060, 0.7332, 0.2904, 0.4166, 0.2478, 0.5579, 0.5494],
  23534. [0.5869, 0.3768, 0.7651, 0.2341, 0.3812, 0.2749, 0.5775, 0.4873],
  23535. [0.5375, 0.3461, 0.8667, 0.5664, 0.4084, 0.4206, 0.5235, 0.5708],
  23536. [0.6034, 0.3690, 0.8506, 0.3974, 0.3891, 0.2989, 0.5728, 0.5234],
  23537. [0.5709, 0.3539, 0.8911, 0.4125, 0.3997, 0.5595, 0.6092, 0.5050],
  23538. [0.5394, 0.3326, 0.8761, 0.5040, 0.4639, 0.5729, 0.6081, 0.5213],
  23539. [0.6028, 0.3882, 0.8502, 0.2723, 0.4557, 0.2620, 0.7067, 0.5437]],
  23540. device='cuda:0', grad_fn=<AddmmBackward>)
  23541. landmarks are: tensor([[[0.6277, 0.4083, 0.8350, 0.2717, 0.4563, 0.1800, 0.5918, 0.4878],
  23542. [0.6210, 0.4164, 0.7202, 0.2930, 0.4025, 0.2483, 0.5688, 0.5567],
  23543. [0.6193, 0.4034, 0.7757, 0.2347, 0.3733, 0.2919, 0.5930, 0.4926],
  23544. [0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  23545. [0.6231, 0.3973, 0.8650, 0.3950, 0.3625, 0.3183, 0.5838, 0.5167],
  23546. [0.6223, 0.4028, 0.8988, 0.4200, 0.3762, 0.5733, 0.6375, 0.5167],
  23547. [0.6185, 0.4098, 0.8838, 0.4900, 0.4538, 0.5800, 0.6288, 0.5400],
  23548. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586]]],
  23549. device='cuda:0')
  23550. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  23551. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  23552. loss_train: 0.04255643083888572
  23553. step: 56
  23554. running loss: 0.0007599362649801021
  23555. Train Steps: 56/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23556. torch.Size([8, 8])
  23557. tensor([[0.6224, 0.4061, 0.8988, 0.4300, 0.3838, 0.4750, 0.6112, 0.5483],
  23558. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783],
  23559. [0.6257, 0.4167, 0.8775, 0.3433, 0.3563, 0.4133, 0.6200, 0.5667],
  23560. [ nan, nan, 0.7612, 0.3250, 0.4037, 0.2533, 0.5438, 0.5767],
  23561. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  23562. [0.6234, 0.4179, 0.7825, 0.3450, 0.3813, 0.2867, 0.5675, 0.5617],
  23563. [0.6128, 0.4022, 0.8738, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  23564. [0.6275, 0.4024, 0.8600, 0.2283, 0.5350, 0.1800, 0.7074, 0.5413]],
  23565. device='cuda:0', dtype=torch.float64)
  23566. predictions are: tensor([[ 0.6451, 0.3948, 0.8907, 0.4315, 0.3767, 0.4683, 0.6095, 0.5213],
  23567. [ 0.6609, 0.4179, 0.8393, 0.3767, 0.3521, 0.2983, 0.5139, 0.5342],
  23568. [ 0.6633, 0.4080, 0.8775, 0.3444, 0.3452, 0.4230, 0.6186, 0.5333],
  23569. [ 0.0151, -0.0232, 0.7885, 0.3428, 0.4042, 0.2410, 0.5492, 0.5479],
  23570. [ 0.6191, 0.3628, 0.8780, 0.4186, 0.3605, 0.4454, 0.6142, 0.4847],
  23571. [ 0.7265, 0.4772, 0.7809, 0.3525, 0.3794, 0.2705, 0.5712, 0.5360],
  23572. [ 0.6252, 0.4042, 0.8688, 0.5054, 0.4890, 0.5129, 0.5313, 0.4892],
  23573. [ 0.5614, 0.3362, 0.8642, 0.2500, 0.5491, 0.1969, 0.7115, 0.5232]],
  23574. device='cuda:0', grad_fn=<AddmmBackward>)
  23575. landmarks are: tensor([[[0.6224, 0.4061, 0.8988, 0.4300, 0.3837, 0.4750, 0.6112, 0.5483],
  23576. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783],
  23577. [0.6257, 0.4167, 0.8775, 0.3433, 0.3562, 0.4133, 0.6200, 0.5667],
  23578. [0.0000, 0.0000, 0.7613, 0.3250, 0.4038, 0.2533, 0.5437, 0.5767],
  23579. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  23580. [0.6234, 0.4179, 0.7825, 0.3450, 0.3812, 0.2867, 0.5675, 0.5617],
  23581. [0.6128, 0.4022, 0.8737, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  23582. [0.6275, 0.4024, 0.8600, 0.2283, 0.5350, 0.1800, 0.7074, 0.5413]]],
  23583. device='cuda:0')
  23584. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  23585. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  23586. loss_train: 0.04318221162247937
  23587. step: 57
  23588. running loss: 0.0007575826600434977
  23589. Train Steps: 57/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23590. torch.Size([8, 8])
  23591. tensor([[ nan, nan, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  23592. [0.6204, 0.4013, 0.8075, 0.2400, 0.4313, 0.2050, 0.5800, 0.5150],
  23593. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  23594. [0.6201, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  23595. [0.6135, 0.3994, 0.7913, 0.3050, 0.3625, 0.3050, 0.5837, 0.5050],
  23596. [0.6339, 0.4102, 0.9088, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  23597. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  23598. [0.6213, 0.4001, 0.7712, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183]],
  23599. device='cuda:0', dtype=torch.float64)
  23600. predictions are: tensor([[ 0.0021, -0.0080, 0.7031, 0.2469, 0.4050, 0.1991, 0.5280, 0.5715],
  23601. [ 0.6362, 0.3893, 0.8048, 0.2570, 0.4196, 0.2289, 0.5890, 0.5236],
  23602. [ 0.6615, 0.4203, 0.7796, 0.2559, 0.4539, 0.1697, 0.5960, 0.5247],
  23603. [ 0.7031, 0.4374, 0.7803, 0.1934, 0.4075, 0.2273, 0.5867, 0.4858],
  23604. [ 0.6264, 0.3888, 0.7986, 0.3021, 0.3568, 0.3290, 0.5865, 0.5070],
  23605. [ 0.6252, 0.3895, 0.9177, 0.4737, 0.3849, 0.5662, 0.7386, 0.5326],
  23606. [ 0.6240, 0.3859, 0.8420, 0.5800, 0.3921, 0.5099, 0.5730, 0.6145],
  23607. [ 0.6473, 0.3830, 0.7823, 0.2062, 0.4194, 0.1886, 0.5721, 0.5217]],
  23608. device='cuda:0', grad_fn=<AddmmBackward>)
  23609. landmarks are: tensor([[[0.0000, 0.0000, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  23610. [0.6204, 0.4013, 0.8075, 0.2400, 0.4313, 0.2050, 0.5800, 0.5150],
  23611. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  23612. [0.6202, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  23613. [0.6135, 0.3994, 0.7912, 0.3050, 0.3625, 0.3050, 0.5838, 0.5050],
  23614. [0.6339, 0.4102, 0.9087, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  23615. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  23616. [0.6213, 0.4001, 0.7713, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183]]],
  23617. device='cuda:0')
  23618. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  23619. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  23620. loss_train: 0.04352180591376964
  23621. step: 58
  23622. running loss: 0.0007503759640305111
  23623.  
  23624. Train Steps: 58/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23625. torch.Size([8, 8])
  23626. tensor([[0.6133, 0.4094, 0.8495, 0.4028, 0.3588, 0.3200, 0.5003, 0.5407],
  23627. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  23628. [0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  23629. [0.6136, 0.4117, 0.8700, 0.5167, 0.4188, 0.5083, 0.5147, 0.5495],
  23630. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050],
  23631. [0.6111, 0.3995, 0.8788, 0.4567, 0.3813, 0.4833, 0.5450, 0.5700],
  23632. [0.6245, 0.4100, 0.7762, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  23633. [0.6153, 0.4119, 0.8463, 0.3833, 0.3600, 0.3200, 0.5106, 0.5563]],
  23634. device='cuda:0', dtype=torch.float64)
  23635. predictions are: tensor([[0.5926, 0.3791, 0.8699, 0.3925, 0.3340, 0.3080, 0.5218, 0.5129],
  23636. [0.5920, 0.3798, 0.8317, 0.5436, 0.3708, 0.4394, 0.5764, 0.5986],
  23637. [0.5888, 0.3719, 0.8841, 0.4800, 0.3814, 0.5299, 0.6021, 0.5108],
  23638. [0.6247, 0.3948, 0.8895, 0.5144, 0.4037, 0.5002, 0.5442, 0.5280],
  23639. [0.5756, 0.3660, 0.8945, 0.4966, 0.4299, 0.5158, 0.6156, 0.5064],
  23640. [0.5044, 0.3224, 0.8838, 0.4386, 0.3716, 0.4736, 0.5628, 0.5452],
  23641. [0.6710, 0.4162, 0.7985, 0.2600, 0.4690, 0.1384, 0.5956, 0.5194],
  23642. [0.6026, 0.3820, 0.8637, 0.3745, 0.3446, 0.3149, 0.5279, 0.5321]],
  23643. device='cuda:0', grad_fn=<AddmmBackward>)
  23644. landmarks are: tensor([[[0.6133, 0.4094, 0.8495, 0.4028, 0.3587, 0.3200, 0.5003, 0.5407],
  23645. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  23646. [0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  23647. [0.6136, 0.4117, 0.8700, 0.5167, 0.4187, 0.5083, 0.5147, 0.5495],
  23648. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050],
  23649. [0.6111, 0.3995, 0.8788, 0.4567, 0.3812, 0.4833, 0.5450, 0.5700],
  23650. [0.6245, 0.4100, 0.7763, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  23651. [0.6153, 0.4119, 0.8462, 0.3833, 0.3600, 0.3200, 0.5106, 0.5563]]],
  23652. device='cuda:0')
  23653. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  23654. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  23655. loss_train: 0.04425763736071531
  23656. step: 59
  23657. running loss: 0.0007501294467917849
  23658. Train Steps: 59/90 Loss: 0.0008 torch.Size([8, 600, 800])
  23659. torch.Size([8, 8])
  23660. tensor([[0.6161, 0.4055, 0.8675, 0.3867, 0.3713, 0.4033, 0.5195, 0.5162],
  23661. [0.6228, 0.4004, 0.8750, 0.5250, 0.3825, 0.5233, 0.6362, 0.5000],
  23662. [0.6087, 0.3951, 0.8387, 0.5833, 0.4188, 0.4933, 0.5146, 0.4830],
  23663. [0.6218, 0.4185, 0.7338, 0.2650, 0.4625, 0.1950, 0.5687, 0.5800],
  23664. [0.6167, 0.4048, 0.6831, 0.3639, 0.3763, 0.3017, 0.5700, 0.5883],
  23665. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896],
  23666. [0.6286, 0.4078, 0.8063, 0.2267, 0.4788, 0.1533, 0.5953, 0.4913],
  23667. [ nan, nan, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819]],
  23668. device='cuda:0', dtype=torch.float64)
  23669. predictions are: tensor([[0.6232, 0.4094, 0.9060, 0.3644, 0.3718, 0.3913, 0.5427, 0.5234],
  23670. [0.5923, 0.3678, 0.8974, 0.4886, 0.3832, 0.5251, 0.6586, 0.5065],
  23671. [0.5933, 0.3822, 0.8689, 0.5705, 0.4047, 0.4963, 0.5346, 0.5147],
  23672. [0.6270, 0.4227, 0.7658, 0.2528, 0.4602, 0.1894, 0.5709, 0.5945],
  23673. [0.5592, 0.3721, 0.7411, 0.3281, 0.3730, 0.3106, 0.5833, 0.5873],
  23674. [0.6357, 0.4168, 0.8609, 0.3741, 0.3467, 0.3067, 0.5425, 0.5905],
  23675. [0.6143, 0.3925, 0.8168, 0.2518, 0.4762, 0.1560, 0.5998, 0.5089],
  23676. [0.0199, 0.0254, 0.7184, 0.2600, 0.3754, 0.2183, 0.5676, 0.5825]],
  23677. device='cuda:0', grad_fn=<AddmmBackward>)
  23678. landmarks are: tensor([[[0.6161, 0.4055, 0.8675, 0.3867, 0.3713, 0.4033, 0.5195, 0.5162],
  23679. [0.6228, 0.4004, 0.8750, 0.5250, 0.3825, 0.5233, 0.6363, 0.5000],
  23680. [0.6087, 0.3951, 0.8388, 0.5833, 0.4187, 0.4933, 0.5146, 0.4830],
  23681. [0.6218, 0.4185, 0.7337, 0.2650, 0.4625, 0.1950, 0.5688, 0.5800],
  23682. [0.6167, 0.4048, 0.6831, 0.3639, 0.3762, 0.3017, 0.5700, 0.5883],
  23683. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896],
  23684. [0.6286, 0.4078, 0.8062, 0.2267, 0.4787, 0.1533, 0.5953, 0.4913],
  23685. [0.0000, 0.0000, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819]]],
  23686. device='cuda:0')
  23687. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  23688. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  23689. loss_train: 0.0447188821999589
  23690. step: 60
  23691. running loss: 0.0007453147033326483
  23692. Train Steps: 60/90 Loss: 0.0007 torch.Size([8, 600, 800])
  23693. torch.Size([8, 8])
  23694. tensor([[0.6043, 0.4022, 0.6887, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136],
  23695. [0.6111, 0.3995, 0.8788, 0.4567, 0.3813, 0.4833, 0.5450, 0.5700],
  23696. [0.6214, 0.4112, 0.7838, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  23697. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6038, 0.6167],
  23698. [0.6127, 0.4118, 0.8650, 0.5083, 0.4088, 0.5367, 0.5300, 0.5456],
  23699. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  23700. [0.6321, 0.4048, 0.8738, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  23701. [0.6267, 0.4065, 0.8313, 0.2467, 0.4788, 0.1733, 0.6312, 0.5133]],
  23702. device='cuda:0', dtype=torch.float64)
  23703. predictions are: tensor([[0.5633, 0.3762, 0.6976, 0.2267, 0.3762, 0.2533, 0.5589, 0.5271],
  23704. [0.5295, 0.3590, 0.8761, 0.4640, 0.3860, 0.4623, 0.5525, 0.5737],
  23705. [0.5826, 0.3891, 0.7838, 0.2237, 0.3603, 0.2857, 0.5810, 0.5268],
  23706. [0.6006, 0.4065, 0.8024, 0.3331, 0.3506, 0.3795, 0.6129, 0.6237],
  23707. [0.6266, 0.4274, 0.8681, 0.4975, 0.4155, 0.5203, 0.5503, 0.5702],
  23708. [0.5805, 0.3821, 0.8804, 0.4771, 0.3678, 0.5073, 0.6005, 0.5190],
  23709. [0.5910, 0.3802, 0.8822, 0.5528, 0.3864, 0.4164, 0.6233, 0.4992],
  23710. [0.5723, 0.3710, 0.8342, 0.2592, 0.4869, 0.1630, 0.6223, 0.5278]],
  23711. device='cuda:0', grad_fn=<AddmmBackward>)
  23712. landmarks are: tensor([[[0.6043, 0.4022, 0.6888, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136],
  23713. [0.6111, 0.3995, 0.8788, 0.4567, 0.3812, 0.4833, 0.5450, 0.5700],
  23714. [0.6214, 0.4112, 0.7837, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083],
  23715. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6037, 0.6167],
  23716. [0.6127, 0.4118, 0.8650, 0.5083, 0.4087, 0.5367, 0.5300, 0.5456],
  23717. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  23718. [0.6321, 0.4048, 0.8737, 0.5617, 0.3875, 0.4417, 0.6361, 0.4927],
  23719. [0.6266, 0.4065, 0.8313, 0.2467, 0.4787, 0.1733, 0.6313, 0.5133]]],
  23720. device='cuda:0')
  23721. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  23722. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  23723. loss_train: 0.04516495233110618
  23724. step: 61
  23725. running loss: 0.0007404090546082981
  23726. Train Steps: 61/90 Loss: 0.0007 torch.Size([8, 600, 800])
  23727. torch.Size([8, 8])
  23728. tensor([[0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  23729. [0.6176, 0.3911, 0.8738, 0.4217, 0.3488, 0.4033, 0.6025, 0.4817],
  23730. [0.6276, 0.4095, 0.8237, 0.2250, 0.4662, 0.1783, 0.6171, 0.4869],
  23731. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6063, 0.5617],
  23732. [0.6229, 0.4107, 0.8137, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  23733. [0.6147, 0.4107, 0.8137, 0.3333, 0.3750, 0.2683, 0.5006, 0.5412],
  23734. [0.6200, 0.3978, 0.8900, 0.4550, 0.3775, 0.5200, 0.6150, 0.5367],
  23735. [0.6193, 0.3930, 0.8949, 0.4437, 0.3852, 0.5435, 0.6263, 0.5263]],
  23736. device='cuda:0', dtype=torch.float64)
  23737. predictions are: tensor([[0.5905, 0.3962, 0.7859, 0.2966, 0.4024, 0.2382, 0.5886, 0.5458],
  23738. [0.5662, 0.3733, 0.8546, 0.4214, 0.3753, 0.3908, 0.5882, 0.5191],
  23739. [0.5349, 0.3653, 0.8111, 0.2279, 0.4760, 0.1968, 0.6072, 0.5061],
  23740. [0.6284, 0.4320, 0.8794, 0.4852, 0.3632, 0.4742, 0.5870, 0.5755],
  23741. [0.5406, 0.3842, 0.7860, 0.2888, 0.4766, 0.1790, 0.5583, 0.5668],
  23742. [0.5116, 0.3594, 0.7696, 0.3321, 0.3682, 0.2809, 0.4983, 0.5764],
  23743. [0.6265, 0.4087, 0.8644, 0.4585, 0.3857, 0.5276, 0.5999, 0.5688],
  23744. [0.5494, 0.3629, 0.8488, 0.4420, 0.3872, 0.5560, 0.6158, 0.5347]],
  23745. device='cuda:0', grad_fn=<AddmmBackward>)
  23746. landmarks are: tensor([[[0.6223, 0.3990, 0.8037, 0.2800, 0.4000, 0.2283, 0.5864, 0.5208],
  23747. [0.6176, 0.3911, 0.8737, 0.4217, 0.3487, 0.4033, 0.6025, 0.4817],
  23748. [0.6276, 0.4095, 0.8238, 0.2250, 0.4663, 0.1783, 0.6171, 0.4869],
  23749. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6062, 0.5617],
  23750. [0.6229, 0.4107, 0.8138, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  23751. [0.6147, 0.4107, 0.8138, 0.3333, 0.3750, 0.2683, 0.5006, 0.5412],
  23752. [0.6199, 0.3978, 0.8900, 0.4550, 0.3775, 0.5200, 0.6150, 0.5367],
  23753. [0.6193, 0.3930, 0.8949, 0.4437, 0.3852, 0.5435, 0.6263, 0.5263]]],
  23754. device='cuda:0')
  23755. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  23756. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  23757. loss_train: 0.04607175434648525
  23758. step: 62
  23759. running loss: 0.0007430928120400847
  23760.  
  23761. Train Steps: 62/90 Loss: 0.0007 torch.Size([8, 600, 800])
  23762. torch.Size([8, 8])
  23763. tensor([[0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  23764. [0.6179, 0.3998, 0.8396, 0.3505, 0.3552, 0.3768, 0.6064, 0.5329],
  23765. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  23766. [0.6129, 0.4063, 0.8738, 0.5250, 0.4313, 0.4733, 0.5230, 0.5874],
  23767. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  23768. [ nan, nan, 0.6900, 0.1917, 0.3937, 0.2367, 0.5240, 0.5246],
  23769. [0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  23770. [0.6361, 0.4076, 0.8862, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297]],
  23771. device='cuda:0', dtype=torch.float64)
  23772. predictions are: tensor([[0.6707, 0.4670, 0.8677, 0.4491, 0.3788, 0.3431, 0.5712, 0.5661],
  23773. [0.6315, 0.4356, 0.8234, 0.3544, 0.3526, 0.3590, 0.5785, 0.5125],
  23774. [0.5871, 0.4064, 0.7431, 0.2339, 0.3850, 0.2712, 0.6034, 0.5493],
  23775. [0.6304, 0.4422, 0.8382, 0.5349, 0.4207, 0.4679, 0.5292, 0.5819],
  23776. [0.5833, 0.4111, 0.7829, 0.3156, 0.3633, 0.2995, 0.4954, 0.5502],
  23777. [0.0056, 0.0311, 0.6748, 0.2038, 0.3992, 0.2276, 0.5228, 0.5308],
  23778. [0.6268, 0.4436, 0.8411, 0.5354, 0.4784, 0.4864, 0.5279, 0.5470],
  23779. [0.6584, 0.4570, 0.8667, 0.5405, 0.3740, 0.4484, 0.6698, 0.5254]],
  23780. device='cuda:0', grad_fn=<AddmmBackward>)
  23781. landmarks are: tensor([[[0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  23782. [0.6179, 0.3998, 0.8396, 0.3505, 0.3552, 0.3768, 0.6064, 0.5329],
  23783. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  23784. [0.6130, 0.4063, 0.8737, 0.5250, 0.4313, 0.4733, 0.5230, 0.5874],
  23785. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  23786. [0.0000, 0.0000, 0.6900, 0.1917, 0.3938, 0.2367, 0.5240, 0.5246],
  23787. [0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  23788. [0.6361, 0.4076, 0.8863, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297]]],
  23789. device='cuda:0')
  23790. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  23791. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  23792. loss_train: 0.04644800083769951
  23793. step: 63
  23794. running loss: 0.0007372698545666589
  23795. Train Steps: 63/90 Loss: 0.0007 torch.Size([8, 600, 800])
  23796. torch.Size([8, 8])
  23797. tensor([[0.6147, 0.4026, 0.6600, 0.2467, 0.4088, 0.2150, 0.5489, 0.5773],
  23798. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  23799. [0.6228, 0.4119, 0.7938, 0.2233, 0.4674, 0.1773, 0.6188, 0.5433],
  23800. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  23801. [0.6168, 0.4029, 0.8523, 0.3417, 0.3588, 0.5000, 0.6125, 0.5400],
  23802. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  23803. [0.6179, 0.4118, 0.7278, 0.4237, 0.3588, 0.3400, 0.5675, 0.5917],
  23804. [0.6262, 0.4163, 0.8850, 0.5183, 0.3763, 0.4150, 0.6025, 0.5500]],
  23805. device='cuda:0', dtype=torch.float64)
  23806. predictions are: tensor([[0.5616, 0.3987, 0.6612, 0.2353, 0.4208, 0.1947, 0.5107, 0.5745],
  23807. [0.5573, 0.3817, 0.7776, 0.2782, 0.3594, 0.3186, 0.5920, 0.5305],
  23808. [0.5528, 0.3848, 0.7682, 0.2134, 0.4712, 0.1687, 0.5880, 0.5412],
  23809. [0.5727, 0.3796, 0.8123, 0.5452, 0.4452, 0.4979, 0.5041, 0.4830],
  23810. [0.6293, 0.4308, 0.8392, 0.3437, 0.3631, 0.4830, 0.5959, 0.5178],
  23811. [0.6417, 0.4228, 0.8615, 0.4075, 0.3550, 0.4194, 0.5687, 0.5197],
  23812. [0.5581, 0.3868, 0.7555, 0.3847, 0.3587, 0.3180, 0.5133, 0.5807],
  23813. [0.6269, 0.4197, 0.8552, 0.4713, 0.3764, 0.3860, 0.5693, 0.5327]],
  23814. device='cuda:0', grad_fn=<AddmmBackward>)
  23815. landmarks are: tensor([[[0.6147, 0.4026, 0.6600, 0.2467, 0.4087, 0.2150, 0.5489, 0.5773],
  23816. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  23817. [0.6228, 0.4119, 0.7937, 0.2233, 0.4674, 0.1773, 0.6187, 0.5433],
  23818. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  23819. [0.6168, 0.4029, 0.8523, 0.3417, 0.3587, 0.5000, 0.6125, 0.5400],
  23820. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  23821. [0.6179, 0.4118, 0.7278, 0.4237, 0.3587, 0.3400, 0.5675, 0.5917],
  23822. [0.6262, 0.4163, 0.8850, 0.5183, 0.3762, 0.4150, 0.6025, 0.5500]]],
  23823. device='cuda:0')
  23824. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  23825. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  23826. loss_train: 0.0470950579183409
  23827. step: 64
  23828. running loss: 0.0007358602799740765
  23829. Train Steps: 64/90 Loss: 0.0007 torch.Size([8, 600, 800])
  23830. torch.Size([8, 8])
  23831. tensor([[0.6085, 0.4005, 0.8400, 0.4317, 0.3763, 0.4750, 0.5476, 0.5058],
  23832. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  23833. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896],
  23834. [0.6202, 0.4079, 0.8025, 0.2500, 0.3763, 0.3217, 0.6125, 0.5533],
  23835. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  23836. [0.6346, 0.4086, 0.7938, 0.5500, 0.3962, 0.4867, 0.7343, 0.5702],
  23837. [0.6222, 0.3937, 0.8350, 0.5617, 0.4138, 0.4600, 0.5800, 0.5233],
  23838. [0.6164, 0.4076, 0.8838, 0.4117, 0.3713, 0.5550, 0.6238, 0.5350]],
  23839. device='cuda:0', dtype=torch.float64)
  23840. predictions are: tensor([[0.6138, 0.4214, 0.8354, 0.4241, 0.3648, 0.4590, 0.5131, 0.4924],
  23841. [0.6428, 0.4388, 0.8917, 0.4322, 0.4117, 0.3551, 0.6836, 0.5497],
  23842. [0.5849, 0.4031, 0.7995, 0.3712, 0.3463, 0.3027, 0.4893, 0.5656],
  23843. [0.6113, 0.4225, 0.7709, 0.2444, 0.3745, 0.3282, 0.5668, 0.5318],
  23844. [0.6188, 0.4291, 0.8725, 0.4718, 0.3644, 0.4302, 0.6124, 0.5090],
  23845. [0.6344, 0.4364, 0.7759, 0.5476, 0.3937, 0.4836, 0.6670, 0.5534],
  23846. [0.6023, 0.3985, 0.8128, 0.5589, 0.4047, 0.4458, 0.5374, 0.5205],
  23847. [0.6651, 0.4503, 0.8545, 0.4076, 0.3714, 0.5456, 0.5912, 0.5054]],
  23848. device='cuda:0', grad_fn=<AddmmBackward>)
  23849. landmarks are: tensor([[[0.6084, 0.4005, 0.8400, 0.4317, 0.3762, 0.4750, 0.5476, 0.5058],
  23850. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  23851. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896],
  23852. [0.6202, 0.4079, 0.8025, 0.2500, 0.3762, 0.3217, 0.6125, 0.5533],
  23853. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  23854. [0.6346, 0.4086, 0.7937, 0.5500, 0.3963, 0.4867, 0.7343, 0.5702],
  23855. [0.6222, 0.3937, 0.8350, 0.5617, 0.4137, 0.4600, 0.5800, 0.5233],
  23856. [0.6164, 0.4076, 0.8838, 0.4117, 0.3713, 0.5550, 0.6237, 0.5350]]],
  23857. device='cuda:0')
  23858. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  23859. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  23860. loss_train: 0.04768331356171984
  23861. step: 65
  23862. running loss: 0.0007335894394110745
  23863. Train Steps: 65/90 Loss: 0.0007 torch.Size([8, 600, 800])
  23864. torch.Size([8, 8])
  23865. tensor([[0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  23866. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  23867. [0.6147, 0.4112, 0.7988, 0.3200, 0.3775, 0.2767, 0.5150, 0.5550],
  23868. [0.6120, 0.4014, 0.6863, 0.2817, 0.3700, 0.2783, 0.5513, 0.5667],
  23869. [0.6260, 0.4120, 0.8013, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895],
  23870. [0.6272, 0.4045, 0.8538, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  23871. [0.6205, 0.4016, 0.8350, 0.2717, 0.3987, 0.2550, 0.5787, 0.5133],
  23872. [0.6264, 0.4071, 0.9038, 0.3867, 0.3663, 0.3917, 0.6338, 0.5283]],
  23873. device='cuda:0', dtype=torch.float64)
  23874. predictions are: tensor([[0.6108, 0.4151, 0.8757, 0.4770, 0.4467, 0.4963, 0.5417, 0.5156],
  23875. [0.6673, 0.4383, 0.8964, 0.4764, 0.4516, 0.5921, 0.6215, 0.5118],
  23876. [0.5872, 0.3994, 0.7957, 0.3206, 0.3719, 0.3037, 0.5088, 0.5784],
  23877. [0.5984, 0.3938, 0.6967, 0.2951, 0.3714, 0.3026, 0.5459, 0.5644],
  23878. [0.5771, 0.3856, 0.8076, 0.2430, 0.4911, 0.1712, 0.6498, 0.5274],
  23879. [0.5973, 0.4015, 0.8380, 0.5838, 0.3782, 0.4484, 0.5942, 0.4922],
  23880. [0.5895, 0.3874, 0.8228, 0.2661, 0.4062, 0.2829, 0.5728, 0.5299],
  23881. [0.6101, 0.4115, 0.8977, 0.3835, 0.3710, 0.4011, 0.6480, 0.5377]],
  23882. device='cuda:0', grad_fn=<AddmmBackward>)
  23883. landmarks are: tensor([[[0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  23884. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  23885. [0.6147, 0.4112, 0.7987, 0.3200, 0.3775, 0.2767, 0.5150, 0.5550],
  23886. [0.6120, 0.4013, 0.6862, 0.2817, 0.3700, 0.2783, 0.5512, 0.5667],
  23887. [0.6259, 0.4120, 0.8012, 0.2350, 0.4888, 0.1533, 0.6281, 0.4895],
  23888. [0.6271, 0.4045, 0.8537, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  23889. [0.6205, 0.4015, 0.8350, 0.2717, 0.3988, 0.2550, 0.5788, 0.5133],
  23890. [0.6264, 0.4071, 0.9038, 0.3867, 0.3663, 0.3917, 0.6338, 0.5283]]],
  23891. device='cuda:0')
  23892. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  23893. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  23894. loss_train: 0.04797610554669518
  23895. step: 66
  23896. running loss: 0.0007269106901014421
  23897.  
  23898. Train Steps: 66/90 Loss: 0.0007 torch.Size([8, 600, 800])
  23899. torch.Size([8, 8])
  23900. tensor([[0.6264, 0.3972, 0.8853, 0.4771, 0.3853, 0.4511, 0.6293, 0.5334],
  23901. [0.6300, 0.4133, 0.8538, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  23902. [0.6204, 0.4007, 0.7838, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  23903. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  23904. [0.6122, 0.4006, 0.8850, 0.4217, 0.4088, 0.5517, 0.6063, 0.5517],
  23905. [0.6212, 0.4033, 0.8938, 0.4167, 0.3813, 0.4267, 0.5613, 0.5583],
  23906. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  23907. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378]],
  23908. device='cuda:0', dtype=torch.float64)
  23909. predictions are: tensor([[0.6038, 0.3741, 0.9009, 0.5030, 0.3543, 0.5153, 0.6121, 0.5218],
  23910. [0.6826, 0.4435, 0.8553, 0.2553, 0.5435, 0.2630, 0.7351, 0.5675],
  23911. [0.6452, 0.4126, 0.7562, 0.2424, 0.4441, 0.2027, 0.5805, 0.5247],
  23912. [0.5573, 0.3575, 0.7552, 0.2212, 0.4434, 0.2005, 0.5919, 0.5020],
  23913. [0.6784, 0.4411, 0.8808, 0.4649, 0.4005, 0.5945, 0.5945, 0.5313],
  23914. [0.6025, 0.4080, 0.8856, 0.4498, 0.3573, 0.4614, 0.5600, 0.5423],
  23915. [0.6007, 0.3935, 0.7706, 0.2596, 0.4448, 0.2150, 0.5805, 0.5448],
  23916. [0.6657, 0.4195, 0.8535, 0.5146, 0.4174, 0.5708, 0.7015, 0.5428]],
  23917. device='cuda:0', grad_fn=<AddmmBackward>)
  23918. landmarks are: tensor([[[0.6264, 0.3972, 0.8853, 0.4771, 0.3853, 0.4511, 0.6293, 0.5334],
  23919. [0.6300, 0.4133, 0.8537, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  23920. [0.6204, 0.4007, 0.7837, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  23921. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  23922. [0.6122, 0.4006, 0.8850, 0.4217, 0.4087, 0.5517, 0.6062, 0.5517],
  23923. [0.6212, 0.4033, 0.8938, 0.4167, 0.3812, 0.4267, 0.5612, 0.5583],
  23924. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  23925. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378]]],
  23926. device='cuda:0')
  23927. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  23928. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  23929. loss_train: 0.048750259753433056
  23930. step: 67
  23931. running loss: 0.0007276158172154187
  23932. Train Steps: 67/90 Loss: 0.0007 torch.Size([8, 600, 800])
  23933. torch.Size([8, 8])
  23934. tensor([[0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  23935. [0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  23936. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  23937. [0.6110, 0.4047, 0.8700, 0.4483, 0.3713, 0.3967, 0.5088, 0.5517],
  23938. [0.6178, 0.4012, 0.8900, 0.4495, 0.3891, 0.5617, 0.5972, 0.4879],
  23939. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  23940. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  23941. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290]],
  23942. device='cuda:0', dtype=torch.float64)
  23943. predictions are: tensor([[0.6498, 0.4061, 0.8830, 0.4685, 0.3808, 0.4473, 0.6150, 0.5376],
  23944. [0.6717, 0.4231, 0.8865, 0.5102, 0.4039, 0.5551, 0.6107, 0.5116],
  23945. [0.6743, 0.4255, 0.8788, 0.4651, 0.3708, 0.3492, 0.5595, 0.5616],
  23946. [0.6536, 0.4194, 0.8888, 0.4669, 0.3753, 0.4371, 0.5224, 0.5394],
  23947. [0.7025, 0.4403, 0.8933, 0.4476, 0.3991, 0.5874, 0.6223, 0.4826],
  23948. [0.6191, 0.3752, 0.9168, 0.3248, 0.5109, 0.2552, 0.7435, 0.5421],
  23949. [0.6457, 0.3986, 0.7897, 0.2415, 0.4592, 0.1989, 0.6085, 0.5358],
  23950. [0.6341, 0.3853, 0.7629, 0.2232, 0.4380, 0.2254, 0.6620, 0.5276]],
  23951. device='cuda:0', grad_fn=<AddmmBackward>)
  23952. landmarks are: tensor([[[0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  23953. [0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  23954. [0.6201, 0.4116, 0.8725, 0.4733, 0.3700, 0.3217, 0.5386, 0.5767],
  23955. [0.6110, 0.4047, 0.8700, 0.4483, 0.3713, 0.3967, 0.5088, 0.5517],
  23956. [0.6178, 0.4012, 0.8900, 0.4495, 0.3891, 0.5617, 0.5972, 0.4879],
  23957. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  23958. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  23959. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290]]],
  23960. device='cuda:0')
  23961. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  23962. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  23963. loss_train: 0.049255486417678185
  23964. step: 68
  23965. running loss: 0.0007243453884952675
  23966. Train Steps: 68/90 Loss: 0.0007 torch.Size([8, 600, 800])
  23967. torch.Size([8, 8])
  23968. tensor([[0.6204, 0.4049, 0.7975, 0.2700, 0.3937, 0.2567, 0.5700, 0.5183],
  23969. [0.6168, 0.4029, 0.8523, 0.3417, 0.3588, 0.5000, 0.6125, 0.5400],
  23970. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  23971. [0.6271, 0.4040, 0.9138, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  23972. [0.6236, 0.3966, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  23973. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  23974. [0.6307, 0.4045, 0.8025, 0.5833, 0.3775, 0.4867, 0.6892, 0.5459],
  23975. [0.6329, 0.4196, 0.9238, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748]],
  23976. device='cuda:0', dtype=torch.float64)
  23977. predictions are: tensor([[0.6066, 0.3856, 0.8011, 0.2484, 0.4016, 0.2708, 0.5804, 0.5224],
  23978. [0.6193, 0.3879, 0.8731, 0.3425, 0.3602, 0.5105, 0.6248, 0.5186],
  23979. [0.6137, 0.3860, 0.7856, 0.2496, 0.3973, 0.2793, 0.5676, 0.5596],
  23980. [0.6431, 0.3960, 0.9349, 0.3631, 0.4764, 0.2546, 0.7226, 0.5344],
  23981. [0.6240, 0.3707, 0.9125, 0.4803, 0.3718, 0.4266, 0.6152, 0.5204],
  23982. [0.6428, 0.3944, 0.7469, 0.2015, 0.4372, 0.2012, 0.6081, 0.5121],
  23983. [0.6284, 0.3823, 0.8317, 0.5589, 0.4057, 0.4849, 0.6951, 0.5196],
  23984. [0.6600, 0.4035, 0.9298, 0.4591, 0.4420, 0.2887, 0.7233, 0.5457]],
  23985. device='cuda:0', grad_fn=<AddmmBackward>)
  23986. landmarks are: tensor([[[0.6204, 0.4049, 0.7975, 0.2700, 0.3938, 0.2567, 0.5700, 0.5183],
  23987. [0.6168, 0.4029, 0.8523, 0.3417, 0.3587, 0.5000, 0.6125, 0.5400],
  23988. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  23989. [0.6271, 0.4040, 0.9137, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  23990. [0.6236, 0.3965, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  23991. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  23992. [0.6307, 0.4045, 0.8025, 0.5833, 0.3775, 0.4867, 0.6892, 0.5459],
  23993. [0.6329, 0.4196, 0.9237, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748]]],
  23994. device='cuda:0')
  23995. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  23996. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  23997. loss_train: 0.04951172370056156
  23998. step: 69
  23999. running loss: 0.0007175612130516168
  24000. Train Steps: 69/90 Loss: 0.0007 torch.Size([8, 600, 800])
  24001. torch.Size([8, 8])
  24002. tensor([[0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  24003. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  24004. [0.6099, 0.4030, 0.8638, 0.5117, 0.4983, 0.4965, 0.5086, 0.5388],
  24005. [0.6198, 0.4101, 0.8838, 0.5283, 0.3763, 0.5267, 0.5913, 0.5567],
  24006. [0.6034, 0.4011, 0.7350, 0.2533, 0.3438, 0.3367, 0.5516, 0.5084],
  24007. [0.6202, 0.4066, 0.8398, 0.2648, 0.3925, 0.2627, 0.5845, 0.5124],
  24008. [0.6364, 0.4165, 0.9088, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  24009. [0.6125, 0.4076, 0.8488, 0.3883, 0.3700, 0.3683, 0.5026, 0.5505]],
  24010. device='cuda:0', dtype=torch.float64)
  24011. predictions are: tensor([[0.6369, 0.4063, 0.9166, 0.4760, 0.3783, 0.4141, 0.6935, 0.5250],
  24012. [0.6562, 0.3967, 0.9160, 0.4163, 0.3726, 0.3441, 0.6424, 0.5450],
  24013. [0.6723, 0.4198, 0.8920, 0.5082, 0.5132, 0.4629, 0.5744, 0.5426],
  24014. [0.6825, 0.4162, 0.9109, 0.5298, 0.3909, 0.5015, 0.6304, 0.5498],
  24015. [0.6299, 0.3838, 0.7454, 0.2568, 0.3605, 0.3079, 0.6139, 0.5104],
  24016. [0.6160, 0.3820, 0.8326, 0.2532, 0.4049, 0.2346, 0.6360, 0.5246],
  24017. [0.6479, 0.4071, 0.9284, 0.4265, 0.4301, 0.2949, 0.6847, 0.5333],
  24018. [0.6430, 0.3917, 0.8725, 0.3664, 0.3732, 0.3443, 0.5314, 0.5447]],
  24019. device='cuda:0', grad_fn=<AddmmBackward>)
  24020. landmarks are: tensor([[[0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  24021. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  24022. [0.6098, 0.4030, 0.8637, 0.5117, 0.4983, 0.4965, 0.5086, 0.5388],
  24023. [0.6198, 0.4101, 0.8838, 0.5283, 0.3762, 0.5267, 0.5913, 0.5567],
  24024. [0.6033, 0.4011, 0.7350, 0.2533, 0.3438, 0.3367, 0.5516, 0.5084],
  24025. [0.6202, 0.4066, 0.8398, 0.2648, 0.3925, 0.2627, 0.5845, 0.5124],
  24026. [0.6364, 0.4165, 0.9087, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  24027. [0.6125, 0.4076, 0.8487, 0.3883, 0.3700, 0.3683, 0.5026, 0.5505]]],
  24028. device='cuda:0')
  24029. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  24030. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  24031. loss_train: 0.05016214041097555
  24032. step: 70
  24033. running loss: 0.0007166020058710793
  24034.  
  24035. Train Steps: 70/90 Loss: 0.0007 torch.Size([8, 600, 800])
  24036. torch.Size([8, 8])
  24037. tensor([[ nan, nan, 0.7268, 0.2333, 0.4125, 0.1933, 0.5112, 0.5383],
  24038. [0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050],
  24039. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  24040. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  24041. [0.6135, 0.3994, 0.7913, 0.3050, 0.3625, 0.3050, 0.5837, 0.5050],
  24042. [0.6137, 0.4038, 0.8563, 0.4050, 0.3813, 0.2550, 0.5106, 0.4954],
  24043. [0.6102, 0.4020, 0.8638, 0.3717, 0.3625, 0.5017, 0.6038, 0.5500],
  24044. [0.6248, 0.4185, 0.8500, 0.5767, 0.4463, 0.4550, 0.5613, 0.5917]],
  24045. device='cuda:0', dtype=torch.float64)
  24046. predictions are: tensor([[0.2846, 0.1741, 0.7245, 0.1989, 0.4228, 0.1664, 0.5666, 0.5444],
  24047. [0.7737, 0.4870, 0.8862, 0.4056, 0.3876, 0.4681, 0.6411, 0.6069],
  24048. [0.6626, 0.4021, 0.8599, 0.2900, 0.3500, 0.3739, 0.6576, 0.5374],
  24049. [0.6829, 0.4202, 0.8933, 0.5592, 0.3822, 0.4282, 0.6746, 0.5033],
  24050. [0.6728, 0.4183, 0.8102, 0.3008, 0.3831, 0.2867, 0.6244, 0.5226],
  24051. [0.7139, 0.4413, 0.8702, 0.4020, 0.4185, 0.2396, 0.5509, 0.5229],
  24052. [0.6474, 0.4059, 0.8866, 0.3840, 0.3860, 0.4751, 0.6363, 0.5533],
  24053. [0.7377, 0.4589, 0.8799, 0.5809, 0.4689, 0.4086, 0.5932, 0.6058]],
  24054. device='cuda:0', grad_fn=<AddmmBackward>)
  24055. landmarks are: tensor([[[0.0000, 0.0000, 0.7268, 0.2333, 0.4125, 0.1933, 0.5113, 0.5383],
  24056. [0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050],
  24057. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  24058. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  24059. [0.6135, 0.3994, 0.7912, 0.3050, 0.3625, 0.3050, 0.5838, 0.5050],
  24060. [0.6137, 0.4038, 0.8562, 0.4050, 0.3812, 0.2550, 0.5106, 0.4954],
  24061. [0.6102, 0.4020, 0.8637, 0.3717, 0.3625, 0.5017, 0.6037, 0.5500],
  24062. [0.6248, 0.4185, 0.8500, 0.5767, 0.4462, 0.4550, 0.5612, 0.5917]]],
  24063. device='cuda:0')
  24064. loss_train_step before backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  24065. loss_train_step after backward: tensor(0.0032, device='cuda:0', grad_fn=<MseLossBackward>)
  24066. loss_train: 0.053396602292195894
  24067. step: 71
  24068. running loss: 0.0007520648210168436
  24069. Train Steps: 71/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24070. torch.Size([8, 8])
  24071. tensor([[ nan, nan, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  24072. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783],
  24073. [0.6277, 0.4118, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  24074. [0.6329, 0.4196, 0.9238, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748],
  24075. [0.6336, 0.4191, 0.8938, 0.5167, 0.3937, 0.3517, 0.7343, 0.5748],
  24076. [0.6307, 0.4029, 0.8988, 0.4817, 0.3937, 0.3500, 0.7311, 0.5378],
  24077. [0.6250, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6088, 0.5183],
  24078. [0.6168, 0.4029, 0.8523, 0.3417, 0.3588, 0.5000, 0.6125, 0.5400]],
  24079. device='cuda:0', dtype=torch.float64)
  24080. predictions are: tensor([[0.1801, 0.1107, 0.7775, 0.2522, 0.3755, 0.2760, 0.5482, 0.5658],
  24081. [0.6555, 0.4322, 0.8262, 0.3429, 0.3661, 0.2883, 0.5107, 0.5873],
  24082. [0.6422, 0.4133, 0.8809, 0.3643, 0.3938, 0.2512, 0.6069, 0.5259],
  24083. [0.6714, 0.4308, 0.9102, 0.4524, 0.4319, 0.2764, 0.7034, 0.5672],
  24084. [0.6379, 0.4200, 0.8942, 0.5023, 0.4042, 0.3349, 0.7128, 0.5673],
  24085. [0.6437, 0.4143, 0.8999, 0.4609, 0.3997, 0.3374, 0.7002, 0.5466],
  24086. [0.6936, 0.4358, 0.8862, 0.4660, 0.4581, 0.5420, 0.6077, 0.5310],
  24087. [0.6167, 0.4021, 0.8470, 0.3338, 0.3548, 0.4915, 0.5990, 0.5446]],
  24088. device='cuda:0', grad_fn=<AddmmBackward>)
  24089. landmarks are: tensor([[[0.0000, 0.0000, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  24090. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783],
  24091. [0.6277, 0.4117, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  24092. [0.6329, 0.4196, 0.9237, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748],
  24093. [0.6336, 0.4191, 0.8938, 0.5167, 0.3938, 0.3517, 0.7343, 0.5748],
  24094. [0.6307, 0.4029, 0.8988, 0.4817, 0.3938, 0.3500, 0.7311, 0.5378],
  24095. [0.6251, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6087, 0.5183],
  24096. [0.6168, 0.4029, 0.8523, 0.3417, 0.3587, 0.5000, 0.6125, 0.5400]]],
  24097. device='cuda:0')
  24098. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  24099. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  24100. loss_train: 0.05442776322888676
  24101. step: 72
  24102. running loss: 0.0007559411559567605
  24103. Train Steps: 72/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24104. torch.Size([8, 8])
  24105. tensor([[0.6185, 0.4080, 0.8625, 0.3483, 0.3788, 0.2650, 0.5320, 0.5272],
  24106. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  24107. [ nan, nan, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  24108. [0.6304, 0.4024, 0.8925, 0.4800, 0.3937, 0.4817, 0.7485, 0.5297],
  24109. [0.6157, 0.4102, 0.8513, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  24110. [0.6175, 0.3957, 0.8700, 0.4817, 0.4662, 0.5133, 0.5800, 0.5517],
  24111. [0.6135, 0.3994, 0.7913, 0.3050, 0.3625, 0.3050, 0.5837, 0.5050],
  24112. [0.6259, 0.4156, 0.8812, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960]],
  24113. device='cuda:0', dtype=torch.float64)
  24114. predictions are: tensor([[0.6175, 0.4065, 0.8519, 0.3611, 0.3844, 0.2618, 0.5318, 0.5447],
  24115. [0.6762, 0.4368, 0.8453, 0.2408, 0.4535, 0.2396, 0.6948, 0.5665],
  24116. [0.1374, 0.0842, 0.7511, 0.2508, 0.3693, 0.2625, 0.5106, 0.5620],
  24117. [0.6211, 0.3957, 0.8926, 0.4968, 0.3888, 0.4682, 0.6999, 0.5258],
  24118. [0.6164, 0.4151, 0.8564, 0.3892, 0.3551, 0.3569, 0.5156, 0.5850],
  24119. [0.6030, 0.4068, 0.8657, 0.4823, 0.4596, 0.5002, 0.5685, 0.5478],
  24120. [0.6145, 0.3968, 0.7989, 0.3191, 0.3619, 0.3048, 0.5876, 0.5153],
  24121. [0.6064, 0.3935, 0.8939, 0.3244, 0.4601, 0.1780, 0.6137, 0.5080]],
  24122. device='cuda:0', grad_fn=<AddmmBackward>)
  24123. landmarks are: tensor([[[0.6186, 0.4080, 0.8625, 0.3483, 0.3787, 0.2650, 0.5320, 0.5272],
  24124. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  24125. [0.0000, 0.0000, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  24126. [0.6304, 0.4024, 0.8925, 0.4800, 0.3938, 0.4817, 0.7485, 0.5297],
  24127. [0.6157, 0.4102, 0.8512, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  24128. [0.6175, 0.3957, 0.8700, 0.4817, 0.4663, 0.5133, 0.5800, 0.5517],
  24129. [0.6135, 0.3994, 0.7912, 0.3050, 0.3625, 0.3050, 0.5838, 0.5050],
  24130. [0.6259, 0.4156, 0.8813, 0.3183, 0.4775, 0.1867, 0.6219, 0.4960]]],
  24131. device='cuda:0')
  24132. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  24133. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  24134. loss_train: 0.05502122991310898
  24135. step: 73
  24136. running loss: 0.0007537154782617668
  24137. Train Steps: 73/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24138. torch.Size([8, 8])
  24139. tensor([[0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  24140. [0.6182, 0.3972, 0.8552, 0.5914, 0.3683, 0.4181, 0.5688, 0.5378],
  24141. [0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5413, 0.5433],
  24142. [0.6185, 0.4079, 0.8838, 0.4617, 0.4838, 0.5650, 0.6175, 0.5850],
  24143. [0.6339, 0.4102, 0.8588, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089],
  24144. [0.6201, 0.4151, 0.8588, 0.5467, 0.3700, 0.3950, 0.5637, 0.5933],
  24145. [ nan, nan, 0.9088, 0.3783, 0.4562, 0.2617, 0.6741, 0.5575],
  24146. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986]],
  24147. device='cuda:0', dtype=torch.float64)
  24148. predictions are: tensor([[0.5918, 0.3869, 0.8965, 0.4814, 0.3500, 0.4986, 0.6200, 0.5045],
  24149. [0.6109, 0.3999, 0.8538, 0.5933, 0.3659, 0.4532, 0.5646, 0.5465],
  24150. [0.6070, 0.4145, 0.8305, 0.3525, 0.3421, 0.3375, 0.5324, 0.5634],
  24151. [0.6302, 0.4250, 0.8856, 0.4422, 0.4552, 0.5652, 0.6061, 0.5706],
  24152. [0.6146, 0.4082, 0.8555, 0.3182, 0.4298, 0.2219, 0.6388, 0.5193],
  24153. [0.6136, 0.4235, 0.8473, 0.5513, 0.3558, 0.4063, 0.5634, 0.5982],
  24154. [0.2057, 0.1608, 0.9028, 0.3820, 0.4332, 0.2708, 0.6609, 0.5754],
  24155. [0.6079, 0.4167, 0.8930, 0.3856, 0.4246, 0.2257, 0.6196, 0.5225]],
  24156. device='cuda:0', grad_fn=<AddmmBackward>)
  24157. landmarks are: tensor([[[0.6205, 0.4004, 0.8938, 0.4883, 0.3663, 0.5000, 0.6357, 0.4947],
  24158. [0.6182, 0.3972, 0.8552, 0.5914, 0.3683, 0.4181, 0.5688, 0.5378],
  24159. [0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5412, 0.5433],
  24160. [0.6184, 0.4079, 0.8838, 0.4617, 0.4837, 0.5650, 0.6175, 0.5850],
  24161. [0.6339, 0.4102, 0.8587, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089],
  24162. [0.6202, 0.4151, 0.8587, 0.5467, 0.3700, 0.3950, 0.5638, 0.5933],
  24163. [0.0000, 0.0000, 0.9087, 0.3783, 0.4563, 0.2617, 0.6741, 0.5575],
  24164. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986]]],
  24165. device='cuda:0')
  24166. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  24167. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  24168. loss_train: 0.05629452761786524
  24169. step: 74
  24170. running loss: 0.0007607368597008817
  24171.  
  24172. Train Steps: 74/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24173. torch.Size([8, 8])
  24174. tensor([[0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  24175. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6138, 0.5333],
  24176. [ nan, nan, 0.7612, 0.3250, 0.4037, 0.2533, 0.5438, 0.5767],
  24177. [0.6229, 0.4107, 0.8137, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  24178. [0.6241, 0.4143, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  24179. [0.6177, 0.4086, 0.8738, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700],
  24180. [0.6201, 0.4151, 0.8588, 0.5467, 0.3700, 0.3950, 0.5637, 0.5933],
  24181. [0.6199, 0.4093, 0.7913, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700]],
  24182. device='cuda:0', dtype=torch.float64)
  24183. predictions are: tensor([[ 0.5973, 0.3953, 0.8694, 0.4266, 0.3525, 0.4234, 0.5897, 0.5375],
  24184. [ 0.5847, 0.4172, 0.8889, 0.4558, 0.3766, 0.4846, 0.6057, 0.5288],
  24185. [-0.0367, -0.0179, 0.7660, 0.3021, 0.4063, 0.2443, 0.5496, 0.5709],
  24186. [ 0.5803, 0.4091, 0.8079, 0.2855, 0.4667, 0.1643, 0.5777, 0.5374],
  24187. [ 0.6129, 0.4125, 0.8918, 0.4804, 0.4105, 0.5255, 0.6131, 0.5526],
  24188. [ 0.6146, 0.4240, 0.8785, 0.4078, 0.3761, 0.5326, 0.6223, 0.5452],
  24189. [ 0.6074, 0.4237, 0.8493, 0.5601, 0.3660, 0.3887, 0.5702, 0.5769],
  24190. [ 0.5856, 0.4142, 0.7920, 0.2483, 0.4149, 0.2620, 0.5998, 0.5645]],
  24191. device='cuda:0', grad_fn=<AddmmBackward>)
  24192. landmarks are: tensor([[[0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  24193. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6137, 0.5333],
  24194. [0.0000, 0.0000, 0.7613, 0.3250, 0.4038, 0.2533, 0.5437, 0.5767],
  24195. [0.6229, 0.4107, 0.8138, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  24196. [0.6241, 0.4142, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  24197. [0.6177, 0.4085, 0.8737, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700],
  24198. [0.6202, 0.4151, 0.8587, 0.5467, 0.3700, 0.3950, 0.5638, 0.5933],
  24199. [0.6198, 0.4093, 0.7912, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700]]],
  24200. device='cuda:0')
  24201. loss_train_step before backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  24202. loss_train_step after backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  24203. loss_train: 0.05648698155710008
  24204. step: 75
  24205. running loss: 0.0007531597540946678
  24206. Train Steps: 75/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24207. torch.Size([8, 8])
  24208. tensor([[0.6208, 0.4082, 0.8538, 0.3067, 0.3588, 0.3717, 0.6112, 0.5517],
  24209. [0.6275, 0.4024, 0.8600, 0.2283, 0.5350, 0.1800, 0.7074, 0.5413],
  24210. [0.6284, 0.4127, 0.8538, 0.5867, 0.4363, 0.5083, 0.6038, 0.5433],
  24211. [0.6200, 0.3993, 0.8519, 0.4923, 0.3962, 0.4717, 0.6013, 0.5433],
  24212. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  24213. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167],
  24214. [ nan, nan, 0.8488, 0.2300, 0.5563, 0.2100, 0.7390, 0.5679],
  24215. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383]],
  24216. device='cuda:0', dtype=torch.float64)
  24217. predictions are: tensor([[ 0.5860, 0.3923, 0.8425, 0.3362, 0.3515, 0.3928, 0.5738, 0.5406],
  24218. [ 0.6200, 0.4110, 0.8593, 0.2493, 0.5200, 0.1874, 0.6884, 0.5160],
  24219. [ 0.5664, 0.3781, 0.8495, 0.6168, 0.4182, 0.5069, 0.5700, 0.5070],
  24220. [ 0.5546, 0.3733, 0.8446, 0.5169, 0.3833, 0.4792, 0.5760, 0.5295],
  24221. [ 0.6115, 0.4124, 0.8515, 0.4124, 0.3485, 0.3957, 0.5713, 0.5496],
  24222. [ 0.5548, 0.3851, 0.7570, 0.3027, 0.3813, 0.2792, 0.5600, 0.6106],
  24223. [-0.0340, -0.0110, 0.8601, 0.2618, 0.5085, 0.2280, 0.7064, 0.5195],
  24224. [ 0.5388, 0.3658, 0.8529, 0.5009, 0.4163, 0.4864, 0.5184, 0.5185]],
  24225. device='cuda:0', grad_fn=<AddmmBackward>)
  24226. landmarks are: tensor([[[0.6208, 0.4082, 0.8537, 0.3067, 0.3587, 0.3717, 0.6112, 0.5517],
  24227. [0.6275, 0.4024, 0.8600, 0.2283, 0.5350, 0.1800, 0.7074, 0.5413],
  24228. [0.6284, 0.4127, 0.8537, 0.5867, 0.4363, 0.5083, 0.6037, 0.5433],
  24229. [0.6200, 0.3993, 0.8519, 0.4923, 0.3963, 0.4717, 0.6012, 0.5433],
  24230. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  24231. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167],
  24232. [0.0000, 0.0000, 0.8487, 0.2300, 0.5562, 0.2100, 0.7390, 0.5679],
  24233. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383]]],
  24234. device='cuda:0')
  24235. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  24236. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  24237. loss_train: 0.05719406365824398
  24238. step: 76
  24239. running loss: 0.0007525534691874208
  24240. Train Steps: 76/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24241. torch.Size([8, 8])
  24242. tensor([[0.6361, 0.4076, 0.8862, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297],
  24243. [0.6185, 0.4042, 0.7700, 0.2250, 0.4062, 0.2117, 0.5763, 0.5150],
  24244. [0.6038, 0.3946, 0.8413, 0.4883, 0.3563, 0.4550, 0.5266, 0.4693],
  24245. [0.6250, 0.4106, 0.8700, 0.3717, 0.3588, 0.4967, 0.6038, 0.5167],
  24246. [0.6142, 0.3982, 0.8650, 0.4883, 0.3912, 0.4317, 0.5315, 0.5350],
  24247. [0.6175, 0.3957, 0.8700, 0.4817, 0.4662, 0.5133, 0.5800, 0.5517],
  24248. [0.6100, 0.4016, 0.8600, 0.5067, 0.4612, 0.5233, 0.5086, 0.5519],
  24249. [0.6120, 0.4014, 0.6863, 0.2817, 0.3700, 0.2783, 0.5513, 0.5667]],
  24250. device='cuda:0', dtype=torch.float64)
  24251. predictions are: tensor([[0.5419, 0.3606, 0.9036, 0.5570, 0.3769, 0.4674, 0.6729, 0.5537],
  24252. [0.5144, 0.3454, 0.7958, 0.2320, 0.4242, 0.1981, 0.5915, 0.5206],
  24253. [0.5326, 0.3526, 0.8689, 0.5011, 0.3797, 0.4656, 0.5266, 0.5109],
  24254. [0.5822, 0.3942, 0.8984, 0.3880, 0.3711, 0.5033, 0.6150, 0.5528],
  24255. [0.5335, 0.3495, 0.8842, 0.5127, 0.3977, 0.4380, 0.5443, 0.5262],
  24256. [0.5281, 0.3587, 0.8922, 0.4874, 0.4806, 0.4987, 0.5777, 0.5571],
  24257. [0.5011, 0.3317, 0.8862, 0.5422, 0.4830, 0.5021, 0.5015, 0.5560],
  24258. [0.5466, 0.3579, 0.7138, 0.2882, 0.3758, 0.2871, 0.5510, 0.5647]],
  24259. device='cuda:0', grad_fn=<AddmmBackward>)
  24260. landmarks are: tensor([[[0.6361, 0.4076, 0.8863, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297],
  24261. [0.6184, 0.4042, 0.7700, 0.2250, 0.4062, 0.2117, 0.5763, 0.5150],
  24262. [0.6038, 0.3946, 0.8413, 0.4883, 0.3562, 0.4550, 0.5266, 0.4693],
  24263. [0.6250, 0.4105, 0.8700, 0.3717, 0.3587, 0.4967, 0.6037, 0.5167],
  24264. [0.6143, 0.3982, 0.8650, 0.4883, 0.3913, 0.4317, 0.5315, 0.5350],
  24265. [0.6175, 0.3957, 0.8700, 0.4817, 0.4663, 0.5133, 0.5800, 0.5517],
  24266. [0.6100, 0.4016, 0.8600, 0.5067, 0.4613, 0.5233, 0.5086, 0.5519],
  24267. [0.6120, 0.4013, 0.6862, 0.2817, 0.3700, 0.2783, 0.5512, 0.5667]]],
  24268. device='cuda:0')
  24269. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  24270. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  24271. loss_train: 0.05860871459299233
  24272. step: 77
  24273. running loss: 0.000761152137571329
  24274. Train Steps: 77/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24275. torch.Size([8, 8])
  24276. tensor([[0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726],
  24277. [0.6196, 0.4094, 0.7562, 0.2817, 0.3937, 0.3183, 0.6013, 0.6183],
  24278. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  24279. [0.6200, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913],
  24280. [0.6107, 0.4013, 0.8700, 0.4650, 0.5049, 0.5176, 0.5850, 0.5550],
  24281. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864],
  24282. [0.6197, 0.4091, 0.8800, 0.4783, 0.3538, 0.4767, 0.5950, 0.5550],
  24283. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395]],
  24284. device='cuda:0', dtype=torch.float64)
  24285. predictions are: tensor([[0.6068, 0.3924, 0.7669, 0.2712, 0.3886, 0.2821, 0.5187, 0.4813],
  24286. [0.5069, 0.3377, 0.7737, 0.2816, 0.4305, 0.3110, 0.6087, 0.6188],
  24287. [0.5604, 0.3725, 0.8664, 0.3686, 0.3614, 0.3860, 0.5699, 0.5144],
  24288. [0.4626, 0.2921, 0.8657, 0.5072, 0.4164, 0.4981, 0.5736, 0.4907],
  24289. [0.5271, 0.3523, 0.8900, 0.4604, 0.5042, 0.4929, 0.5649, 0.5500],
  24290. [0.5475, 0.3650, 0.8750, 0.4847, 0.3954, 0.4822, 0.7045, 0.5844],
  24291. [0.5740, 0.3863, 0.8746, 0.4546, 0.3666, 0.4736, 0.5707, 0.5636],
  24292. [0.4890, 0.3238, 0.8717, 0.5288, 0.3855, 0.3552, 0.5640, 0.5294]],
  24293. device='cuda:0', grad_fn=<AddmmBackward>)
  24294. landmarks are: tensor([[[0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726],
  24295. [0.6196, 0.4094, 0.7563, 0.2817, 0.3938, 0.3183, 0.6012, 0.6183],
  24296. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  24297. [0.6199, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913],
  24298. [0.6107, 0.4013, 0.8700, 0.4650, 0.5049, 0.5176, 0.5850, 0.5550],
  24299. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864],
  24300. [0.6197, 0.4091, 0.8800, 0.4783, 0.3537, 0.4767, 0.5950, 0.5550],
  24301. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395]]],
  24302. device='cuda:0')
  24303. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  24304. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  24305. loss_train: 0.060351279433234595
  24306. step: 78
  24307. running loss: 0.0007737343517081358
  24308.  
  24309. Train Steps: 78/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24310. torch.Size([8, 8])
  24311. tensor([[ nan, nan, 0.8363, 0.3317, 0.3563, 0.3367, 0.5329, 0.5142],
  24312. [0.6224, 0.4179, 0.8700, 0.5683, 0.4037, 0.4683, 0.5650, 0.5600],
  24313. [0.6125, 0.4035, 0.7825, 0.3100, 0.3463, 0.4900, 0.5832, 0.5637],
  24314. [0.6200, 0.4059, 0.8700, 0.4900, 0.4163, 0.5000, 0.6162, 0.5467],
  24315. [0.6233, 0.4091, 0.8100, 0.2950, 0.3563, 0.3883, 0.6013, 0.5200],
  24316. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  24317. [ nan, nan, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  24318. [0.6189, 0.4049, 0.8888, 0.4417, 0.4213, 0.5200, 0.5988, 0.5633]],
  24319. device='cuda:0', dtype=torch.float64)
  24320. predictions are: tensor([[-0.0143, -0.0024, 0.8038, 0.3344, 0.3643, 0.3234, 0.5648, 0.5117],
  24321. [ 0.6862, 0.4534, 0.8565, 0.5897, 0.4113, 0.4675, 0.5926, 0.5485],
  24322. [ 0.6735, 0.4393, 0.7570, 0.3208, 0.3560, 0.4950, 0.5838, 0.5183],
  24323. [ 0.6936, 0.4571, 0.8300, 0.4819, 0.4362, 0.5289, 0.6201, 0.5342],
  24324. [ 0.6809, 0.4478, 0.7915, 0.3032, 0.3672, 0.4096, 0.6177, 0.5297],
  24325. [ 0.6235, 0.4041, 0.8658, 0.5046, 0.3969, 0.4643, 0.5237, 0.5457],
  24326. [-0.0387, -0.0200, 0.8594, 0.2558, 0.5169, 0.2219, 0.7540, 0.5521],
  24327. [ 0.6800, 0.4418, 0.8547, 0.4494, 0.4336, 0.5273, 0.6129, 0.5608]],
  24328. device='cuda:0', grad_fn=<AddmmBackward>)
  24329. landmarks are: tensor([[[0.0000, 0.0000, 0.8363, 0.3317, 0.3562, 0.3367, 0.5329, 0.5142],
  24330. [0.6224, 0.4179, 0.8700, 0.5683, 0.4038, 0.4683, 0.5650, 0.5600],
  24331. [0.6125, 0.4035, 0.7825, 0.3100, 0.3462, 0.4900, 0.5832, 0.5637],
  24332. [0.6199, 0.4059, 0.8700, 0.4900, 0.4162, 0.5000, 0.6162, 0.5467],
  24333. [0.6233, 0.4091, 0.8100, 0.2950, 0.3562, 0.3883, 0.6012, 0.5200],
  24334. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  24335. [0.0000, 0.0000, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  24336. [0.6189, 0.4049, 0.8888, 0.4417, 0.4212, 0.5200, 0.5987, 0.5633]]],
  24337. device='cuda:0')
  24338. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  24339. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  24340. loss_train: 0.06108348035195377
  24341. step: 79
  24342. running loss: 0.0007732086120500476
  24343. Train Steps: 79/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24344. torch.Size([8, 8])
  24345. tensor([[0.6201, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  24346. [0.6296, 0.4008, 0.9150, 0.4317, 0.4263, 0.3050, 0.7256, 0.5413],
  24347. [0.6177, 0.4086, 0.8738, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700],
  24348. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  24349. [0.6200, 0.4070, 0.8938, 0.4183, 0.3538, 0.4567, 0.6175, 0.5400],
  24350. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6188, 0.5283],
  24351. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  24352. [0.6249, 0.4142, 0.8350, 0.3283, 0.3613, 0.3700, 0.6188, 0.5400]],
  24353. device='cuda:0', dtype=torch.float64)
  24354. predictions are: tensor([[0.5224, 0.3419, 0.7550, 0.2004, 0.4276, 0.2089, 0.5920, 0.4913],
  24355. [0.5525, 0.3680, 0.9134, 0.4232, 0.4112, 0.3087, 0.7105, 0.5515],
  24356. [0.5902, 0.4045, 0.8558, 0.3962, 0.3823, 0.5594, 0.6005, 0.5678],
  24357. [0.5608, 0.3880, 0.8351, 0.3319, 0.3604, 0.4914, 0.5836, 0.5657],
  24358. [0.6617, 0.4348, 0.8837, 0.4245, 0.3588, 0.4808, 0.6075, 0.5278],
  24359. [0.5215, 0.3512, 0.8832, 0.3702, 0.4036, 0.2688, 0.6154, 0.5268],
  24360. [0.5126, 0.3345, 0.8717, 0.4826, 0.4417, 0.5923, 0.5912, 0.5300],
  24361. [0.5494, 0.3744, 0.8171, 0.3323, 0.3662, 0.3569, 0.6079, 0.5482]],
  24362. device='cuda:0', grad_fn=<AddmmBackward>)
  24363. landmarks are: tensor([[[0.6202, 0.4055, 0.7717, 0.1919, 0.4250, 0.2079, 0.5871, 0.5044],
  24364. [0.6296, 0.4008, 0.9150, 0.4317, 0.4263, 0.3050, 0.7256, 0.5413],
  24365. [0.6177, 0.4085, 0.8737, 0.3950, 0.3775, 0.5600, 0.6225, 0.5700],
  24366. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  24367. [0.6200, 0.4070, 0.8938, 0.4183, 0.3537, 0.4567, 0.6175, 0.5400],
  24368. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6187, 0.5283],
  24369. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  24370. [0.6249, 0.4142, 0.8350, 0.3283, 0.3613, 0.3700, 0.6187, 0.5400]]],
  24371. device='cuda:0')
  24372. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  24373. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  24374. loss_train: 0.06227331746777054
  24375. step: 80
  24376. running loss: 0.0007784164683471318
  24377. Train Steps: 80/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24378. torch.Size([8, 8])
  24379. tensor([[0.6185, 0.4042, 0.7700, 0.2250, 0.4062, 0.2117, 0.5763, 0.5150],
  24380. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  24381. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  24382. [0.6350, 0.4043, 0.8738, 0.5650, 0.3850, 0.4750, 0.6401, 0.4950],
  24383. [0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133],
  24384. [0.6296, 0.4008, 0.9150, 0.4317, 0.4263, 0.3050, 0.7256, 0.5413],
  24385. [0.6286, 0.4055, 0.9000, 0.4717, 0.3763, 0.4683, 0.7018, 0.5494],
  24386. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268]],
  24387. device='cuda:0', dtype=torch.float64)
  24388. predictions are: tensor([[0.5728, 0.3780, 0.7630, 0.2122, 0.4002, 0.2459, 0.5436, 0.5268],
  24389. [0.5629, 0.3804, 0.8455, 0.2477, 0.4786, 0.1995, 0.5931, 0.5053],
  24390. [0.5817, 0.3744, 0.8778, 0.3963, 0.3435, 0.3911, 0.5819, 0.5604],
  24391. [0.6186, 0.3939, 0.8592, 0.5448, 0.3678, 0.5465, 0.6132, 0.5131],
  24392. [0.6066, 0.3892, 0.8570, 0.5147, 0.3561, 0.5063, 0.6153, 0.5325],
  24393. [0.5814, 0.3799, 0.9132, 0.3978, 0.3947, 0.3442, 0.6963, 0.5596],
  24394. [0.5945, 0.3818, 0.8924, 0.4424, 0.3485, 0.5178, 0.6996, 0.5596],
  24395. [0.5771, 0.3658, 0.8505, 0.4618, 0.4108, 0.5537, 0.5980, 0.5515]],
  24396. device='cuda:0', grad_fn=<AddmmBackward>)
  24397. landmarks are: tensor([[[0.6184, 0.4042, 0.7700, 0.2250, 0.4062, 0.2117, 0.5763, 0.5150],
  24398. [0.6258, 0.4143, 0.8525, 0.2617, 0.4950, 0.1667, 0.6219, 0.4967],
  24399. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  24400. [0.6350, 0.4043, 0.8737, 0.5650, 0.3850, 0.4750, 0.6401, 0.4950],
  24401. [0.6280, 0.4055, 0.8600, 0.5317, 0.3800, 0.4700, 0.6275, 0.5133],
  24402. [0.6296, 0.4008, 0.9150, 0.4317, 0.4263, 0.3050, 0.7256, 0.5413],
  24403. [0.6286, 0.4055, 0.9000, 0.4717, 0.3762, 0.4683, 0.7018, 0.5494],
  24404. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268]]],
  24405. device='cuda:0')
  24406. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  24407. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  24408. loss_train: 0.06305107554362621
  24409. step: 81
  24410. running loss: 0.000778408340044768
  24411. Train Steps: 81/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24412. torch.Size([8, 8])
  24413. tensor([[0.6154, 0.4112, 0.7037, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  24414. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6138, 0.5333],
  24415. [0.6196, 0.4094, 0.7562, 0.2817, 0.3937, 0.3183, 0.6013, 0.6183],
  24416. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  24417. [0.6117, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103],
  24418. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  24419. [0.6185, 0.4067, 0.8838, 0.4450, 0.4037, 0.4733, 0.5213, 0.5142],
  24420. [0.6219, 0.4114, 0.8175, 0.2817, 0.3925, 0.2783, 0.5900, 0.5350]],
  24421. device='cuda:0', dtype=torch.float64)
  24422. predictions are: tensor([[0.5925, 0.3868, 0.7185, 0.2207, 0.4239, 0.1918, 0.5743, 0.5804],
  24423. [0.5980, 0.3963, 0.9158, 0.4424, 0.3656, 0.5126, 0.6424, 0.5339],
  24424. [0.6732, 0.4341, 0.7695, 0.2681, 0.3920, 0.3232, 0.6471, 0.6118],
  24425. [0.4580, 0.2868, 0.7231, 0.2085, 0.4095, 0.2476, 0.5547, 0.5508],
  24426. [0.5601, 0.3514, 0.6649, 0.2035, 0.3818, 0.2585, 0.5571, 0.5226],
  24427. [0.5739, 0.3568, 0.7445, 0.1915, 0.4272, 0.2050, 0.6112, 0.5174],
  24428. [0.5811, 0.3766, 0.8925, 0.4364, 0.3828, 0.4687, 0.5463, 0.5253],
  24429. [0.5821, 0.3855, 0.8140, 0.2603, 0.3948, 0.2667, 0.6166, 0.5354]],
  24430. device='cuda:0', grad_fn=<AddmmBackward>)
  24431. landmarks are: tensor([[[0.6154, 0.4112, 0.7038, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  24432. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6137, 0.5333],
  24433. [0.6196, 0.4094, 0.7563, 0.2817, 0.3938, 0.3183, 0.6012, 0.6183],
  24434. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  24435. [0.6116, 0.4018, 0.6562, 0.1967, 0.3738, 0.2550, 0.5280, 0.5103],
  24436. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  24437. [0.6185, 0.4067, 0.8838, 0.4450, 0.4038, 0.4733, 0.5213, 0.5142],
  24438. [0.6219, 0.4114, 0.8175, 0.2817, 0.3925, 0.2783, 0.5900, 0.5350]]],
  24439. device='cuda:0')
  24440. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  24441. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  24442. loss_train: 0.06413436621369328
  24443. step: 82
  24444. running loss: 0.000782126417240162
  24445.  
  24446. Train Steps: 82/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24447. torch.Size([8, 8])
  24448. tensor([[0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6138, 0.5333],
  24449. [0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6262, 0.5367],
  24450. [0.6216, 0.4167, 0.8588, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667],
  24451. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  24452. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  24453. [0.6156, 0.4125, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  24454. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  24455. [0.6127, 0.4118, 0.8650, 0.5083, 0.4088, 0.5367, 0.5300, 0.5456]],
  24456. device='cuda:0', dtype=torch.float64)
  24457. predictions are: tensor([[0.6515, 0.4266, 0.8918, 0.4172, 0.3766, 0.4620, 0.6430, 0.5268],
  24458. [0.6456, 0.4085, 0.8798, 0.4478, 0.4039, 0.5626, 0.6127, 0.5239],
  24459. [0.7003, 0.4583, 0.8623, 0.5334, 0.3712, 0.4893, 0.6281, 0.5466],
  24460. [0.6903, 0.4388, 0.7631, 0.3358, 0.3392, 0.3826, 0.5472, 0.5373],
  24461. [0.6365, 0.4038, 0.8620, 0.3614, 0.3483, 0.3893, 0.6020, 0.5475],
  24462. [0.6166, 0.4032, 0.8847, 0.4609, 0.4272, 0.5328, 0.6282, 0.5048],
  24463. [0.7030, 0.4594, 0.7801, 0.2450, 0.4328, 0.1511, 0.6083, 0.5073],
  24464. [0.6425, 0.4281, 0.8641, 0.4693, 0.4036, 0.5050, 0.5713, 0.5506]],
  24465. device='cuda:0', grad_fn=<AddmmBackward>)
  24466. landmarks are: tensor([[[0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6137, 0.5333],
  24467. [0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6263, 0.5367],
  24468. [0.6216, 0.4167, 0.8587, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667],
  24469. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  24470. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  24471. [0.6155, 0.4124, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  24472. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  24473. [0.6127, 0.4118, 0.8650, 0.5083, 0.4087, 0.5367, 0.5300, 0.5456]]],
  24474. device='cuda:0')
  24475. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  24476. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  24477. loss_train: 0.06510589212120976
  24478. step: 83
  24479. running loss: 0.0007844083388097562
  24480. Train Steps: 83/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24481. torch.Size([8, 8])
  24482. tensor([[ nan, nan, 0.7525, 0.2291, 0.3838, 0.3017, 0.6050, 0.5667],
  24483. [0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767],
  24484. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  24485. [0.6229, 0.4066, 0.8513, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  24486. [0.6286, 0.4086, 0.8408, 0.2801, 0.4163, 0.2800, 0.6725, 0.5393],
  24487. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  24488. [0.6236, 0.3966, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  24489. [0.6200, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183]],
  24490. device='cuda:0', dtype=torch.float64)
  24491. predictions are: tensor([[0.1093, 0.0623, 0.7614, 0.2376, 0.4039, 0.2795, 0.6019, 0.5619],
  24492. [0.6374, 0.4181, 0.8749, 0.4911, 0.3949, 0.4465, 0.5443, 0.5621],
  24493. [0.7407, 0.4638, 0.8816, 0.4757, 0.3898, 0.5048, 0.6563, 0.4767],
  24494. [0.7346, 0.4880, 0.8432, 0.5665, 0.4289, 0.4709, 0.6206, 0.5230],
  24495. [0.7766, 0.4909, 0.8456, 0.2801, 0.4178, 0.2757, 0.6832, 0.5210],
  24496. [0.6897, 0.4571, 0.8626, 0.4530, 0.4290, 0.4688, 0.5380, 0.4883],
  24497. [0.6902, 0.4332, 0.8677, 0.4743, 0.3426, 0.3980, 0.5872, 0.5126],
  24498. [0.7270, 0.4772, 0.8036, 0.2919, 0.3839, 0.2818, 0.6136, 0.5013]],
  24499. device='cuda:0', grad_fn=<AddmmBackward>)
  24500. landmarks are: tensor([[[0.0000, 0.0000, 0.7525, 0.2291, 0.3837, 0.3017, 0.6050, 0.5667],
  24501. [0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767],
  24502. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  24503. [0.6229, 0.4066, 0.8512, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  24504. [0.6286, 0.4086, 0.8408, 0.2801, 0.4162, 0.2800, 0.6725, 0.5393],
  24505. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  24506. [0.6236, 0.3965, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  24507. [0.6201, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183]]],
  24508. device='cuda:0')
  24509. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  24510. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  24511. loss_train: 0.06704556952172425
  24512. step: 84
  24513. running loss: 0.0007981615419252886
  24514. Train Steps: 84/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24515. torch.Size([8, 8])
  24516. tensor([[0.6273, 0.4143, 0.8750, 0.5700, 0.3987, 0.4717, 0.6013, 0.5467],
  24517. [0.6263, 0.4030, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  24518. [0.6267, 0.4080, 0.8438, 0.2633, 0.4763, 0.1800, 0.6259, 0.5240],
  24519. [ nan, nan, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  24520. [0.6114, 0.4018, 0.7213, 0.1967, 0.3763, 0.2700, 0.5875, 0.5533],
  24521. [0.6185, 0.4079, 0.8838, 0.4617, 0.4838, 0.5650, 0.6175, 0.5850],
  24522. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  24523. [0.6202, 0.4053, 0.8638, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133]],
  24524. device='cuda:0', dtype=torch.float64)
  24525. predictions are: tensor([[0.7320, 0.4737, 0.8611, 0.5556, 0.3902, 0.4397, 0.5956, 0.5358],
  24526. [0.7077, 0.4610, 0.8938, 0.4741, 0.3591, 0.5033, 0.6193, 0.4845],
  24527. [0.7522, 0.4808, 0.8343, 0.2710, 0.4589, 0.1764, 0.6038, 0.5168],
  24528. [0.0814, 0.0406, 0.8817, 0.3188, 0.4925, 0.2216, 0.6584, 0.5543],
  24529. [0.6683, 0.4413, 0.7192, 0.2160, 0.3753, 0.2583, 0.5710, 0.5413],
  24530. [0.7060, 0.4613, 0.8823, 0.4435, 0.4619, 0.5227, 0.6069, 0.5665],
  24531. [0.6551, 0.4286, 0.8835, 0.4538, 0.3918, 0.5613, 0.6043, 0.5227],
  24532. [0.7188, 0.4664, 0.8615, 0.5288, 0.4406, 0.4852, 0.5693, 0.5099]],
  24533. device='cuda:0', grad_fn=<AddmmBackward>)
  24534. landmarks are: tensor([[[0.6273, 0.4143, 0.8750, 0.5700, 0.3988, 0.4717, 0.6012, 0.5467],
  24535. [0.6263, 0.4029, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  24536. [0.6267, 0.4080, 0.8438, 0.2633, 0.4762, 0.1800, 0.6259, 0.5240],
  24537. [0.0000, 0.0000, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  24538. [0.6114, 0.4018, 0.7212, 0.1967, 0.3762, 0.2700, 0.5875, 0.5533],
  24539. [0.6184, 0.4079, 0.8838, 0.4617, 0.4837, 0.5650, 0.6175, 0.5850],
  24540. [0.6234, 0.4023, 0.8888, 0.4633, 0.3975, 0.5767, 0.6400, 0.5183],
  24541. [0.6202, 0.4053, 0.8637, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133]]],
  24542. device='cuda:0')
  24543. loss_train_step before backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  24544. loss_train_step after backward: tensor(0.0015, device='cuda:0', grad_fn=<MseLossBackward>)
  24545. loss_train: 0.06853765701816883
  24546. step: 85
  24547. running loss: 0.0008063253766843392
  24548. Train Steps: 85/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24549. torch.Size([8, 8])
  24550. tensor([[0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  24551. [0.6202, 0.4064, 0.7879, 0.2179, 0.4567, 0.1725, 0.5955, 0.5478],
  24552. [0.6286, 0.3977, 0.9038, 0.4733, 0.3900, 0.4150, 0.7074, 0.5320],
  24553. [0.6200, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913],
  24554. [0.6100, 0.4016, 0.8600, 0.5067, 0.4612, 0.5233, 0.5086, 0.5519],
  24555. [0.6173, 0.4114, 0.7325, 0.2500, 0.4213, 0.1917, 0.5338, 0.5700],
  24556. [0.6058, 0.3978, 0.8287, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461],
  24557. [0.6308, 0.3990, 0.8688, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133]],
  24558. device='cuda:0', dtype=torch.float64)
  24559. predictions are: tensor([[0.6508, 0.4336, 0.8535, 0.2768, 0.4638, 0.2824, 0.6922, 0.5507],
  24560. [0.6845, 0.4468, 0.7882, 0.2625, 0.4845, 0.1763, 0.5718, 0.5383],
  24561. [0.6801, 0.4236, 0.9062, 0.4683, 0.3671, 0.4290, 0.6874, 0.5161],
  24562. [0.6959, 0.4352, 0.8825, 0.5497, 0.4140, 0.4984, 0.5772, 0.4890],
  24563. [0.6354, 0.4205, 0.8702, 0.5328, 0.4810, 0.5111, 0.4839, 0.5492],
  24564. [0.5374, 0.3567, 0.7500, 0.2449, 0.4344, 0.2038, 0.5264, 0.5679],
  24565. [0.6251, 0.4128, 0.8308, 0.3663, 0.3539, 0.4017, 0.5425, 0.5324],
  24566. [0.6461, 0.3940, 0.8751, 0.5405, 0.4116, 0.4967, 0.6169, 0.5035]],
  24567. device='cuda:0', grad_fn=<AddmmBackward>)
  24568. landmarks are: tensor([[[0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  24569. [0.6202, 0.4064, 0.7879, 0.2179, 0.4567, 0.1725, 0.5955, 0.5478],
  24570. [0.6286, 0.3977, 0.9038, 0.4733, 0.3900, 0.4150, 0.7074, 0.5320],
  24571. [0.6199, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913],
  24572. [0.6100, 0.4016, 0.8600, 0.5067, 0.4613, 0.5233, 0.5086, 0.5519],
  24573. [0.6173, 0.4114, 0.7325, 0.2500, 0.4212, 0.1917, 0.5337, 0.5700],
  24574. [0.6058, 0.3978, 0.8288, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461],
  24575. [0.6308, 0.3990, 0.8687, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133]]],
  24576. device='cuda:0')
  24577. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  24578. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  24579. loss_train: 0.06918519355531316
  24580. step: 86
  24581. running loss: 0.0008044789948292229
  24582.  
  24583. Train Steps: 86/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24584. torch.Size([8, 8])
  24585. tensor([[0.6353, 0.4128, 0.9138, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  24586. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  24587. [0.6156, 0.4125, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  24588. [0.6336, 0.4154, 0.8900, 0.2767, 0.4988, 0.2867, 0.7422, 0.5540],
  24589. [0.6127, 0.4084, 0.8700, 0.4467, 0.3987, 0.4317, 0.5013, 0.5471],
  24590. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  24591. [0.6263, 0.4030, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  24592. [ nan, nan, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633]],
  24593. device='cuda:0', dtype=torch.float64)
  24594. predictions are: tensor([[ 0.6749, 0.4588, 0.8992, 0.3963, 0.4897, 0.3346, 0.7159, 0.6081],
  24595. [ 0.6886, 0.4441, 0.8908, 0.4452, 0.3752, 0.3636, 0.5976, 0.5500],
  24596. [ 0.6656, 0.4560, 0.8947, 0.5214, 0.4696, 0.5912, 0.5820, 0.5216],
  24597. [ 0.6618, 0.4416, 0.8677, 0.3127, 0.5169, 0.3047, 0.7389, 0.5555],
  24598. [ 0.6127, 0.4198, 0.8645, 0.4757, 0.4118, 0.4427, 0.4806, 0.5549],
  24599. [ 0.6725, 0.4273, 0.7533, 0.2409, 0.4704, 0.1748, 0.5897, 0.4989],
  24600. [ 0.6451, 0.4203, 0.8935, 0.5194, 0.3952, 0.5392, 0.6293, 0.4988],
  24601. [-0.0591, -0.0411, 0.7639, 0.3055, 0.4165, 0.2820, 0.4991, 0.5784]],
  24602. device='cuda:0', grad_fn=<AddmmBackward>)
  24603. landmarks are: tensor([[[0.6353, 0.4128, 0.9137, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991],
  24604. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  24605. [0.6155, 0.4124, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  24606. [0.6336, 0.4154, 0.8900, 0.2767, 0.4988, 0.2867, 0.7422, 0.5540],
  24607. [0.6127, 0.4084, 0.8700, 0.4467, 0.3988, 0.4317, 0.5013, 0.5471],
  24608. [0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  24609. [0.6263, 0.4029, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  24610. [0.0000, 0.0000, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633]]],
  24611. device='cuda:0')
  24612. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  24613. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  24614. loss_train: 0.06988749319862109
  24615. step: 87
  24616. running loss: 0.0008033045195243803
  24617. Train Steps: 87/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24618. torch.Size([8, 8])
  24619. tensor([[0.6274, 0.4003, 0.8638, 0.5967, 0.3688, 0.4900, 0.6108, 0.4661],
  24620. [0.6293, 0.4024, 0.8750, 0.5000, 0.4012, 0.5733, 0.7121, 0.5633],
  24621. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  24622. [0.6332, 0.4165, 0.9100, 0.3350, 0.4188, 0.3683, 0.7438, 0.5528],
  24623. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373],
  24624. [0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  24625. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  24626. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901]],
  24627. device='cuda:0', dtype=torch.float64)
  24628. predictions are: tensor([[0.6093, 0.3767, 0.8861, 0.6041, 0.4109, 0.4834, 0.6146, 0.5040],
  24629. [0.5823, 0.3717, 0.9175, 0.5160, 0.4458, 0.5912, 0.6775, 0.5757],
  24630. [0.6102, 0.4081, 0.7195, 0.2583, 0.4323, 0.2579, 0.5604, 0.5746],
  24631. [0.5733, 0.3835, 0.9435, 0.3738, 0.4507, 0.3760, 0.7062, 0.5696],
  24632. [0.6482, 0.4254, 0.8759, 0.5979, 0.4299, 0.4589, 0.5743, 0.5528],
  24633. [0.6330, 0.4260, 0.9245, 0.4645, 0.4103, 0.3689, 0.5681, 0.5909],
  24634. [0.6020, 0.3900, 0.8795, 0.5706, 0.4186, 0.5017, 0.7201, 0.5703],
  24635. [0.5623, 0.3640, 0.8323, 0.2971, 0.4104, 0.2730, 0.5071, 0.5186]],
  24636. device='cuda:0', grad_fn=<AddmmBackward>)
  24637. landmarks are: tensor([[[0.6274, 0.4003, 0.8637, 0.5967, 0.3688, 0.4900, 0.6108, 0.4661],
  24638. [0.6293, 0.4024, 0.8750, 0.5000, 0.4013, 0.5733, 0.7121, 0.5633],
  24639. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  24640. [0.6332, 0.4165, 0.9100, 0.3350, 0.4187, 0.3683, 0.7438, 0.5528],
  24641. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373],
  24642. [0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  24643. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  24644. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901]]],
  24645. device='cuda:0')
  24646. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  24647. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  24648. loss_train: 0.0706424161180621
  24649. step: 88
  24650. running loss: 0.0008027547286143421
  24651. Train Steps: 88/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24652. torch.Size([8, 8])
  24653. tensor([[0.6265, 0.4251, 0.7113, 0.3550, 0.4375, 0.2117, 0.5587, 0.6118],
  24654. [ nan, nan, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  24655. [ nan, nan, 0.7525, 0.2291, 0.3838, 0.3017, 0.6050, 0.5667],
  24656. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  24657. [0.6132, 0.4066, 0.7259, 0.2402, 0.3588, 0.3300, 0.6000, 0.5600],
  24658. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  24659. [0.6102, 0.4005, 0.8688, 0.5100, 0.4813, 0.5400, 0.5404, 0.5064],
  24660. [0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050]],
  24661. device='cuda:0', dtype=torch.float64)
  24662. predictions are: tensor([[0.7329, 0.4968, 0.7483, 0.3786, 0.4358, 0.2171, 0.5631, 0.5937],
  24663. [0.0535, 0.0381, 0.8644, 0.2943, 0.5412, 0.2195, 0.6699, 0.5556],
  24664. [0.1255, 0.0737, 0.7721, 0.2515, 0.4036, 0.3109, 0.5987, 0.5628],
  24665. [0.7190, 0.4711, 0.9182, 0.5764, 0.3963, 0.4302, 0.5831, 0.5388],
  24666. [0.6366, 0.4122, 0.7466, 0.2615, 0.3732, 0.3417, 0.6117, 0.5708],
  24667. [0.6869, 0.4533, 0.7324, 0.2783, 0.3922, 0.2976, 0.6281, 0.5651],
  24668. [0.6615, 0.4455, 0.8908, 0.5147, 0.4825, 0.5266, 0.5687, 0.4876],
  24669. [0.6953, 0.4555, 0.9070, 0.4323, 0.3842, 0.5011, 0.6273, 0.5798]],
  24670. device='cuda:0', grad_fn=<AddmmBackward>)
  24671. landmarks are: tensor([[[0.6265, 0.4251, 0.7113, 0.3550, 0.4375, 0.2117, 0.5587, 0.6118],
  24672. [0.0000, 0.0000, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  24673. [0.0000, 0.0000, 0.7525, 0.2291, 0.3837, 0.3017, 0.6050, 0.5667],
  24674. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  24675. [0.6132, 0.4066, 0.7259, 0.2402, 0.3587, 0.3300, 0.6000, 0.5600],
  24676. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  24677. [0.6102, 0.4005, 0.8687, 0.5100, 0.4812, 0.5400, 0.5404, 0.5064],
  24678. [0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050]]],
  24679. device='cuda:0')
  24680. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  24681. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  24682. loss_train: 0.07201781774347182
  24683. step: 89
  24684. running loss: 0.0008091889634097958
  24685. Train Steps: 89/90 Loss: 0.0008 torch.Size([8, 600, 800])
  24686. torch.Size([8, 8])
  24687. tensor([[0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  24688. [ nan, nan, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  24689. [0.6131, 0.4037, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  24690. [0.6034, 0.4011, 0.7350, 0.2533, 0.3438, 0.3367, 0.5516, 0.5084],
  24691. [0.6296, 0.4008, 0.9150, 0.4317, 0.4263, 0.3050, 0.7256, 0.5413],
  24692. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  24693. [0.6127, 0.4084, 0.8700, 0.4467, 0.3987, 0.4317, 0.5013, 0.5471],
  24694. [0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633]],
  24695. device='cuda:0', dtype=torch.float64)
  24696. predictions are: tensor([[ 0.6060, 0.3917, 0.6771, 0.3172, 0.3757, 0.2812, 0.5570, 0.5812],
  24697. [-0.0581, -0.0375, 0.8625, 0.2447, 0.5502, 0.2356, 0.7400, 0.5764],
  24698. [ 0.5796, 0.3699, 0.6941, 0.2917, 0.3760, 0.2843, 0.5499, 0.5678],
  24699. [ 0.5950, 0.3760, 0.7373, 0.2582, 0.3706, 0.3330, 0.5954, 0.5064],
  24700. [ 0.6301, 0.3950, 0.9330, 0.4419, 0.4240, 0.3126, 0.7530, 0.5388],
  24701. [ 0.6633, 0.4413, 0.8922, 0.4290, 0.3791, 0.3910, 0.5547, 0.5141],
  24702. [ 0.5718, 0.3831, 0.8884, 0.4709, 0.4058, 0.4407, 0.5156, 0.5442],
  24703. [ 0.5806, 0.3847, 0.9144, 0.5505, 0.3841, 0.4862, 0.5937, 0.5728]],
  24704. device='cuda:0', grad_fn=<AddmmBackward>)
  24705. landmarks are: tensor([[[0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  24706. [0.0000, 0.0000, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  24707. [0.6131, 0.4036, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  24708. [0.6033, 0.4011, 0.7350, 0.2533, 0.3438, 0.3367, 0.5516, 0.5084],
  24709. [0.6296, 0.4008, 0.9150, 0.4317, 0.4263, 0.3050, 0.7256, 0.5413],
  24710. [0.6160, 0.4086, 0.8775, 0.3983, 0.3750, 0.3783, 0.5128, 0.4882],
  24711. [0.6127, 0.4084, 0.8700, 0.4467, 0.3988, 0.4317, 0.5013, 0.5471],
  24712. [0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633]]],
  24713. device='cuda:0')
  24714. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  24715. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  24716. loss_train: 0.0725348553241929
  24717. step: 90
  24718. running loss: 0.0008059428369354767
  24719.  
  24720. Valid Steps: 10/10 Loss: nan 6.2924
  24721. --------------------------------------------------
  24722. Epoch: 8 Train Loss: 0.0008 Valid Loss: nan
  24723. --------------------------------------------------
  24724. size of train loader is: 90
  24725. torch.Size([8, 600, 800])
  24726. torch.Size([8, 8])
  24727. tensor([[0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  24728. [0.6266, 0.4067, 0.8588, 0.2867, 0.4300, 0.2850, 0.6325, 0.5267],
  24729. [ nan, nan, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  24730. [0.6201, 0.3970, 0.8413, 0.4950, 0.4413, 0.5183, 0.6088, 0.5400],
  24731. [0.6092, 0.4001, 0.8638, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  24732. [0.6179, 0.3998, 0.8396, 0.3505, 0.3552, 0.3768, 0.6064, 0.5329],
  24733. [0.6091, 0.3997, 0.8314, 0.4334, 0.3788, 0.4550, 0.5213, 0.5656],
  24734. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167]],
  24735. device='cuda:0', dtype=torch.float64)
  24736. predictions are: tensor([[ 0.5809, 0.3892, 0.8763, 0.5048, 0.3556, 0.4534, 0.5933, 0.5725],
  24737. [ 0.5585, 0.3613, 0.8566, 0.2598, 0.4254, 0.2554, 0.6410, 0.5312],
  24738. [-0.0281, -0.0072, 0.8385, 0.2609, 0.5153, 0.1945, 0.6909, 0.5716],
  24739. [ 0.6296, 0.4013, 0.8372, 0.4857, 0.4138, 0.5036, 0.6315, 0.5534],
  24740. [ 0.5775, 0.3675, 0.8493, 0.4625, 0.4122, 0.5282, 0.5725, 0.5064],
  24741. [ 0.6405, 0.4147, 0.8342, 0.3335, 0.3355, 0.3601, 0.6054, 0.5336],
  24742. [ 0.5747, 0.3587, 0.8181, 0.4326, 0.3595, 0.4352, 0.5605, 0.5659],
  24743. [ 0.5893, 0.3826, 0.8734, 0.4997, 0.3901, 0.4934, 0.6083, 0.5265]],
  24744. device='cuda:0', grad_fn=<AddmmBackward>)
  24745. landmarks are: tensor([[[0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  24746. [0.6266, 0.4067, 0.8587, 0.2867, 0.4300, 0.2850, 0.6325, 0.5267],
  24747. [0.0000, 0.0000, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621],
  24748. [0.6201, 0.3970, 0.8413, 0.4950, 0.4412, 0.5183, 0.6087, 0.5400],
  24749. [0.6092, 0.4001, 0.8637, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  24750. [0.6179, 0.3998, 0.8396, 0.3505, 0.3552, 0.3768, 0.6064, 0.5329],
  24751. [0.6091, 0.3997, 0.8314, 0.4334, 0.3787, 0.4550, 0.5213, 0.5656],
  24752. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167]]],
  24753. device='cuda:0')
  24754. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  24755. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  24756. loss_train: 0.0004860513436142355
  24757. step: 1
  24758. running loss: 0.0004860513436142355
  24759. Train Steps: 1/90 Loss: 0.0005 torch.Size([8, 600, 800])
  24760. torch.Size([8, 8])
  24761. tensor([[0.6127, 0.4115, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  24762. [0.6228, 0.4119, 0.7938, 0.2233, 0.4674, 0.1773, 0.6188, 0.5433],
  24763. [0.6248, 0.4032, 0.7738, 0.1900, 0.4813, 0.1400, 0.5941, 0.4904],
  24764. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  24765. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  24766. [0.6199, 0.4071, 0.7600, 0.2117, 0.4037, 0.2767, 0.6138, 0.5550],
  24767. [0.6339, 0.4118, 0.7988, 0.5800, 0.3912, 0.4583, 0.7343, 0.5760],
  24768. [0.6115, 0.3998, 0.7063, 0.2383, 0.4037, 0.1950, 0.5320, 0.4993]],
  24769. device='cuda:0', dtype=torch.float64)
  24770. predictions are: tensor([[0.4761, 0.3098, 0.7138, 0.2520, 0.3472, 0.2872, 0.5280, 0.5661],
  24771. [0.5404, 0.3577, 0.7892, 0.2008, 0.4592, 0.1701, 0.6062, 0.5569],
  24772. [0.5778, 0.3732, 0.7722, 0.2057, 0.4636, 0.1201, 0.5917, 0.4919],
  24773. [0.5289, 0.3369, 0.8543, 0.4854, 0.4132, 0.4918, 0.5679, 0.5458],
  24774. [0.4861, 0.3146, 0.8276, 0.4051, 0.3561, 0.4725, 0.5445, 0.5689],
  24775. [0.5397, 0.3539, 0.7541, 0.1826, 0.4015, 0.2716, 0.5978, 0.5660],
  24776. [0.5419, 0.3511, 0.7921, 0.5433, 0.3567, 0.4557, 0.6956, 0.5577],
  24777. [0.5144, 0.3353, 0.7089, 0.2007, 0.3883, 0.1958, 0.5168, 0.5069]],
  24778. device='cuda:0', grad_fn=<AddmmBackward>)
  24779. landmarks are: tensor([[[0.6127, 0.4114, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  24780. [0.6228, 0.4119, 0.7937, 0.2233, 0.4674, 0.1773, 0.6187, 0.5433],
  24781. [0.6248, 0.4032, 0.7738, 0.1900, 0.4812, 0.1400, 0.5941, 0.4904],
  24782. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  24783. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  24784. [0.6199, 0.4071, 0.7600, 0.2117, 0.4038, 0.2767, 0.6137, 0.5550],
  24785. [0.6339, 0.4118, 0.7987, 0.5800, 0.3913, 0.4583, 0.7343, 0.5760],
  24786. [0.6115, 0.3998, 0.7063, 0.2383, 0.4038, 0.1950, 0.5320, 0.4993]]],
  24787. device='cuda:0')
  24788. loss_train_step before backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  24789. loss_train_step after backward: tensor(0.0020, device='cuda:0', grad_fn=<MseLossBackward>)
  24790. loss_train: 0.00245886217453517
  24791. step: 2
  24792. running loss: 0.001229431087267585
  24793. Train Steps: 2/90 Loss: 0.0012 torch.Size([8, 600, 800])
  24794. torch.Size([8, 8])
  24795. tensor([[0.6215, 0.4119, 0.7688, 0.2300, 0.4200, 0.2283, 0.5925, 0.5317],
  24796. [0.6200, 0.4118, 0.8287, 0.4017, 0.3775, 0.2833, 0.5391, 0.5799],
  24797. [ nan, nan, 0.6412, 0.1900, 0.4238, 0.1883, 0.5487, 0.5700],
  24798. [0.6201, 0.4017, 0.8871, 0.4621, 0.3517, 0.4675, 0.5999, 0.5106],
  24799. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  24800. [0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  24801. [0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  24802. [0.6304, 0.4024, 0.8925, 0.4800, 0.3937, 0.4817, 0.7485, 0.5297]],
  24803. device='cuda:0', dtype=torch.float64)
  24804. predictions are: tensor([[0.4858, 0.3245, 0.7570, 0.2105, 0.4402, 0.2475, 0.5965, 0.5613],
  24805. [0.5709, 0.3834, 0.7998, 0.3829, 0.3771, 0.2972, 0.5274, 0.5805],
  24806. [0.0874, 0.0654, 0.6780, 0.1980, 0.4254, 0.2052, 0.5490, 0.5937],
  24807. [0.5820, 0.3664, 0.8662, 0.4281, 0.3647, 0.4916, 0.5814, 0.5280],
  24808. [0.5865, 0.3777, 0.8605, 0.4721, 0.3877, 0.5724, 0.6234, 0.5273],
  24809. [0.6304, 0.4000, 0.8916, 0.4334, 0.3792, 0.4218, 0.5842, 0.5127],
  24810. [0.5845, 0.3798, 0.8789, 0.4629, 0.3794, 0.4046, 0.5080, 0.5814],
  24811. [0.6068, 0.3876, 0.8614, 0.4630, 0.4006, 0.4813, 0.7141, 0.5286]],
  24812. device='cuda:0', grad_fn=<AddmmBackward>)
  24813. landmarks are: tensor([[[0.6215, 0.4119, 0.7688, 0.2300, 0.4200, 0.2283, 0.5925, 0.5317],
  24814. [0.6200, 0.4118, 0.8288, 0.4017, 0.3775, 0.2833, 0.5391, 0.5799],
  24815. [0.0000, 0.0000, 0.6413, 0.1900, 0.4238, 0.1883, 0.5487, 0.5700],
  24816. [0.6201, 0.4017, 0.8871, 0.4621, 0.3517, 0.4675, 0.5999, 0.5106],
  24817. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  24818. [0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  24819. [0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  24820. [0.6304, 0.4024, 0.8925, 0.4800, 0.3938, 0.4817, 0.7485, 0.5297]]],
  24821. device='cuda:0')
  24822. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  24823. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  24824. loss_train: 0.0034466659708414227
  24825. step: 3
  24826. running loss: 0.0011488886569471408
  24827. Train Steps: 3/90 Loss: 0.0011 torch.Size([8, 600, 800])
  24828. torch.Size([8, 8])
  24829. tensor([[0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  24830. [0.6175, 0.4091, 0.7863, 0.2800, 0.3638, 0.3583, 0.6188, 0.5433],
  24831. [ nan, nan, 0.8525, 0.2217, 0.5413, 0.2367, 0.7367, 0.5482],
  24832. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  24833. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  24834. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  24835. [0.6293, 0.4097, 0.8800, 0.2517, 0.5262, 0.2600, 0.7430, 0.5378],
  24836. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400]],
  24837. device='cuda:0', dtype=torch.float64)
  24838. predictions are: tensor([[0.5710, 0.3786, 0.8432, 0.3172, 0.3542, 0.2946, 0.5317, 0.5189],
  24839. [0.5667, 0.3979, 0.7542, 0.2634, 0.3320, 0.3626, 0.5838, 0.5578],
  24840. [0.0320, 0.0410, 0.8086, 0.2045, 0.5000, 0.2239, 0.6892, 0.5512],
  24841. [0.6161, 0.4025, 0.8187, 0.4901, 0.4073, 0.4758, 0.5308, 0.5540],
  24842. [0.5866, 0.3672, 0.8515, 0.4788, 0.3924, 0.5120, 0.6089, 0.4946],
  24843. [0.5726, 0.3708, 0.8269, 0.4869, 0.4068, 0.5184, 0.6734, 0.5451],
  24844. [0.6295, 0.4060, 0.8382, 0.2308, 0.4869, 0.2442, 0.6998, 0.5377],
  24845. [0.5761, 0.3858, 0.7478, 0.2148, 0.4222, 0.1888, 0.5715, 0.5357]],
  24846. device='cuda:0', grad_fn=<AddmmBackward>)
  24847. landmarks are: tensor([[[0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117],
  24848. [0.6175, 0.4091, 0.7862, 0.2800, 0.3638, 0.3583, 0.6187, 0.5433],
  24849. [0.0000, 0.0000, 0.8525, 0.2217, 0.5412, 0.2367, 0.7367, 0.5482],
  24850. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  24851. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  24852. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  24853. [0.6293, 0.4097, 0.8800, 0.2517, 0.5263, 0.2600, 0.7430, 0.5378],
  24854. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400]]],
  24855. device='cuda:0')
  24856. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  24857. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  24858. loss_train: 0.004258358589140698
  24859. step: 4
  24860. running loss: 0.0010645896472851746
  24861.  
  24862. Train Steps: 4/90 Loss: 0.0011 torch.Size([8, 600, 800])
  24863. torch.Size([8, 8])
  24864. tensor([[0.6252, 0.4158, 0.8988, 0.4083, 0.3788, 0.4783, 0.6225, 0.5633],
  24865. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  24866. [0.6200, 0.4070, 0.8938, 0.4183, 0.3538, 0.4567, 0.6175, 0.5400],
  24867. [0.6346, 0.4092, 0.7712, 0.5917, 0.4037, 0.4767, 0.7343, 0.5725],
  24868. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  24869. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  24870. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  24871. [0.6361, 0.4076, 0.8862, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297]],
  24872. device='cuda:0', dtype=torch.float64)
  24873. predictions are: tensor([[0.6286, 0.4358, 0.8769, 0.3787, 0.3642, 0.4453, 0.5894, 0.5461],
  24874. [0.5941, 0.4048, 0.8591, 0.4995, 0.4092, 0.5502, 0.5732, 0.5008],
  24875. [0.4940, 0.3278, 0.8859, 0.3949, 0.3396, 0.4385, 0.5975, 0.5139],
  24876. [0.6161, 0.4102, 0.7761, 0.5220, 0.3656, 0.4566, 0.6900, 0.5708],
  24877. [0.5571, 0.3729, 0.8705, 0.4630, 0.3933, 0.4678, 0.5163, 0.5224],
  24878. [0.5966, 0.3890, 0.8608, 0.4651, 0.4166, 0.4937, 0.6023, 0.5137],
  24879. [0.5856, 0.3920, 0.8522, 0.2076, 0.5237, 0.1951, 0.7348, 0.5356],
  24880. [0.5518, 0.3843, 0.8716, 0.5228, 0.3479, 0.4377, 0.6359, 0.5273]],
  24881. device='cuda:0', grad_fn=<AddmmBackward>)
  24882. landmarks are: tensor([[[0.6252, 0.4158, 0.8988, 0.4083, 0.3787, 0.4783, 0.6225, 0.5633],
  24883. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  24884. [0.6200, 0.4070, 0.8938, 0.4183, 0.3537, 0.4567, 0.6175, 0.5400],
  24885. [0.6346, 0.4092, 0.7713, 0.5917, 0.4038, 0.4767, 0.7343, 0.5725],
  24886. [0.6193, 0.4165, 0.8838, 0.4700, 0.4150, 0.4867, 0.5427, 0.5261],
  24887. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  24888. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  24889. [0.6361, 0.4076, 0.8863, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297]]],
  24890. device='cuda:0')
  24891. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  24892. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  24893. loss_train: 0.005290690081892535
  24894. step: 5
  24895. running loss: 0.001058138016378507
  24896. Train Steps: 5/90 Loss: 0.0011 torch.Size([8, 600, 800])
  24897. torch.Size([8, 8])
  24898. tensor([[0.6223, 0.4028, 0.8988, 0.4200, 0.3763, 0.5733, 0.6375, 0.5167],
  24899. [ nan, nan, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733],
  24900. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  24901. [0.6230, 0.4113, 0.7213, 0.1983, 0.4325, 0.2367, 0.6262, 0.5400],
  24902. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  24903. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  24904. [0.6115, 0.3998, 0.7063, 0.2383, 0.4037, 0.1950, 0.5320, 0.4993],
  24905. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290]],
  24906. device='cuda:0', dtype=torch.float64)
  24907. predictions are: tensor([[0.5628, 0.3675, 0.9012, 0.4321, 0.4039, 0.5772, 0.6425, 0.5173],
  24908. [0.1123, 0.0837, 0.6882, 0.2115, 0.4484, 0.1851, 0.5827, 0.5779],
  24909. [0.6376, 0.4141, 0.8935, 0.5407, 0.4235, 0.5582, 0.7549, 0.5473],
  24910. [0.6193, 0.4202, 0.7262, 0.2125, 0.4546, 0.2231, 0.6338, 0.5522],
  24911. [0.6757, 0.4559, 0.7992, 0.3261, 0.3754, 0.2913, 0.5048, 0.5636],
  24912. [0.6086, 0.4059, 0.8714, 0.4880, 0.3842, 0.5257, 0.6139, 0.5058],
  24913. [0.6429, 0.4291, 0.7146, 0.2307, 0.4096, 0.1962, 0.5411, 0.4932],
  24914. [0.5703, 0.3764, 0.8226, 0.3077, 0.3508, 0.3935, 0.6427, 0.5397]],
  24915. device='cuda:0', grad_fn=<AddmmBackward>)
  24916. landmarks are: tensor([[[0.6223, 0.4028, 0.8988, 0.4200, 0.3762, 0.5733, 0.6375, 0.5167],
  24917. [0.0000, 0.0000, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733],
  24918. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  24919. [0.6230, 0.4113, 0.7212, 0.1983, 0.4325, 0.2367, 0.6263, 0.5400],
  24920. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646],
  24921. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  24922. [0.6115, 0.3998, 0.7063, 0.2383, 0.4038, 0.1950, 0.5320, 0.4993],
  24923. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290]]],
  24924. device='cuda:0')
  24925. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  24926. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  24927. loss_train: 0.006010964134475216
  24928. step: 6
  24929. running loss: 0.0010018273557458695
  24930. Train Steps: 6/90 Loss: 0.0010 torch.Size([8, 600, 800])
  24931. torch.Size([8, 8])
  24932. tensor([[0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  24933. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  24934. [0.6133, 0.4066, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358],
  24935. [0.6091, 0.3997, 0.8314, 0.4334, 0.3788, 0.4550, 0.5213, 0.5656],
  24936. [0.6197, 0.4090, 0.7825, 0.2500, 0.4200, 0.2483, 0.5988, 0.5667],
  24937. [0.6246, 0.4028, 0.8738, 0.4867, 0.4088, 0.5667, 0.6362, 0.5200],
  24938. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  24939. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116]],
  24940. device='cuda:0', dtype=torch.float64)
  24941. predictions are: tensor([[0.6033, 0.3968, 0.8633, 0.4681, 0.3756, 0.3958, 0.5996, 0.5504],
  24942. [0.6511, 0.4366, 0.8907, 0.5408, 0.3869, 0.4642, 0.6706, 0.4991],
  24943. [0.6189, 0.4157, 0.7080, 0.2688, 0.3831, 0.2361, 0.5732, 0.5136],
  24944. [0.6097, 0.4049, 0.8417, 0.4520, 0.3871, 0.4563, 0.5520, 0.5556],
  24945. [0.6564, 0.4354, 0.7826, 0.2350, 0.4024, 0.2438, 0.6246, 0.5608],
  24946. [0.6491, 0.4279, 0.8879, 0.4915, 0.4421, 0.5618, 0.6730, 0.5224],
  24947. [0.6068, 0.4018, 0.7448, 0.1839, 0.4090, 0.2412, 0.6519, 0.5410],
  24948. [0.5664, 0.3761, 0.8622, 0.4227, 0.4177, 0.5919, 0.6115, 0.5155]],
  24949. device='cuda:0', grad_fn=<AddmmBackward>)
  24950. landmarks are: tensor([[[0.6125, 0.4010, 0.8650, 0.4567, 0.3663, 0.3900, 0.5600, 0.5567],
  24951. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  24952. [0.6133, 0.4065, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358],
  24953. [0.6091, 0.3997, 0.8314, 0.4334, 0.3787, 0.4550, 0.5213, 0.5656],
  24954. [0.6197, 0.4090, 0.7825, 0.2500, 0.4200, 0.2483, 0.5987, 0.5667],
  24955. [0.6246, 0.4028, 0.8737, 0.4867, 0.4087, 0.5667, 0.6363, 0.5200],
  24956. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  24957. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116]]],
  24958. device='cuda:0')
  24959. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  24960. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  24961. loss_train: 0.006409419496776536
  24962. step: 7
  24963. running loss: 0.0009156313566823623
  24964. Train Steps: 7/90 Loss: 0.0009 torch.Size([8, 600, 800])
  24965. torch.Size([8, 8])
  24966. tensor([[0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  24967. [0.6275, 0.4111, 0.8463, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  24968. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  24969. [0.6185, 0.4129, 0.8900, 0.4567, 0.3937, 0.5417, 0.5734, 0.5110],
  24970. [0.6196, 0.4088, 0.8888, 0.4583, 0.4500, 0.5683, 0.6138, 0.5883],
  24971. [0.6284, 0.4127, 0.8538, 0.5867, 0.4363, 0.5083, 0.6038, 0.5433],
  24972. [0.6167, 0.4048, 0.6831, 0.3639, 0.3763, 0.3017, 0.5700, 0.5883],
  24973. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450]],
  24974. device='cuda:0', dtype=torch.float64)
  24975. predictions are: tensor([[0.6184, 0.3991, 0.8760, 0.3445, 0.3885, 0.2568, 0.6579, 0.4637],
  24976. [0.6569, 0.4371, 0.8507, 0.2505, 0.4504, 0.2141, 0.6291, 0.4659],
  24977. [0.6265, 0.4320, 0.8225, 0.5594, 0.4027, 0.4503, 0.5940, 0.5949],
  24978. [0.6272, 0.4296, 0.8969, 0.4458, 0.3914, 0.5446, 0.5957, 0.4927],
  24979. [0.6098, 0.4292, 0.8974, 0.4494, 0.4426, 0.5651, 0.6509, 0.5711],
  24980. [0.5869, 0.3878, 0.8679, 0.5695, 0.4349, 0.5059, 0.6164, 0.5257],
  24981. [0.5942, 0.3922, 0.7184, 0.3401, 0.3705, 0.3252, 0.5932, 0.5651],
  24982. [0.6090, 0.4050, 0.8767, 0.4538, 0.3760, 0.4140, 0.6232, 0.5329]],
  24983. device='cuda:0', grad_fn=<AddmmBackward>)
  24984. landmarks are: tensor([[[0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  24985. [0.6275, 0.4111, 0.8462, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  24986. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  24987. [0.6186, 0.4129, 0.8900, 0.4567, 0.3938, 0.5417, 0.5734, 0.5110],
  24988. [0.6196, 0.4088, 0.8888, 0.4583, 0.4500, 0.5683, 0.6137, 0.5883],
  24989. [0.6284, 0.4127, 0.8537, 0.5867, 0.4363, 0.5083, 0.6037, 0.5433],
  24990. [0.6167, 0.4048, 0.6831, 0.3639, 0.3762, 0.3017, 0.5700, 0.5883],
  24991. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450]]],
  24992. device='cuda:0')
  24993. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  24994. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  24995. loss_train: 0.006727247644448653
  24996. step: 8
  24997. running loss: 0.0008409059555560816
  24998.  
  24999. Train Steps: 8/90 Loss: 0.0008 torch.Size([8, 600, 800])
  25000. torch.Size([8, 8])
  25001. tensor([[0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  25002. [0.6275, 0.4111, 0.8463, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  25003. [0.6196, 0.4088, 0.8888, 0.4583, 0.4500, 0.5683, 0.6138, 0.5883],
  25004. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  25005. [0.6154, 0.4112, 0.7037, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  25006. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  25007. [0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  25008. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600]],
  25009. device='cuda:0', dtype=torch.float64)
  25010. predictions are: tensor([[0.6183, 0.4030, 0.7435, 0.3145, 0.3552, 0.3331, 0.5759, 0.5422],
  25011. [0.6342, 0.4077, 0.8421, 0.2665, 0.4672, 0.2127, 0.6182, 0.4593],
  25012. [0.6046, 0.4084, 0.8912, 0.4746, 0.4667, 0.5795, 0.6430, 0.5635],
  25013. [0.5526, 0.3554, 0.9060, 0.4546, 0.4237, 0.3305, 0.6833, 0.5396],
  25014. [0.6539, 0.4260, 0.7077, 0.2388, 0.4306, 0.1791, 0.5493, 0.5390],
  25015. [0.6156, 0.4022, 0.7569, 0.2314, 0.3848, 0.2848, 0.6123, 0.5242],
  25016. [0.6136, 0.4001, 0.8375, 0.4258, 0.3736, 0.5034, 0.6086, 0.5390],
  25017. [0.6474, 0.4279, 0.8735, 0.5599, 0.4168, 0.5013, 0.5937, 0.5653]],
  25018. device='cuda:0', grad_fn=<AddmmBackward>)
  25019. landmarks are: tensor([[[0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  25020. [0.6275, 0.4111, 0.8462, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  25021. [0.6196, 0.4088, 0.8888, 0.4583, 0.4500, 0.5683, 0.6137, 0.5883],
  25022. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  25023. [0.6154, 0.4112, 0.7038, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  25024. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  25025. [0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  25026. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600]]],
  25027. device='cuda:0')
  25028. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  25029. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  25030. loss_train: 0.0071166820416692644
  25031. step: 9
  25032. running loss: 0.0007907424490743628
  25033. Train Steps: 9/90 Loss: 0.0008 torch.Size([8, 600, 800])
  25034. torch.Size([8, 8])
  25035. tensor([[0.6064, 0.3953, 0.8738, 0.4417, 0.3663, 0.4683, 0.5511, 0.5416],
  25036. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  25037. [0.6250, 0.4110, 0.7238, 0.2067, 0.4263, 0.1883, 0.5625, 0.5633],
  25038. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5463, 0.5800],
  25039. [0.6201, 0.4064, 0.8688, 0.5050, 0.4225, 0.5100, 0.6138, 0.5500],
  25040. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  25041. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373],
  25042. [0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947]],
  25043. device='cuda:0', dtype=torch.float64)
  25044. predictions are: tensor([[0.6942, 0.4339, 0.8907, 0.4814, 0.3835, 0.5011, 0.5572, 0.5364],
  25045. [0.6969, 0.4367, 0.8986, 0.5011, 0.3749, 0.4080, 0.6448, 0.5017],
  25046. [0.5561, 0.3609, 0.7203, 0.2376, 0.4244, 0.2072, 0.5682, 0.5440],
  25047. [0.6394, 0.4293, 0.7587, 0.3248, 0.3745, 0.2947, 0.5382, 0.5660],
  25048. [0.6396, 0.4080, 0.8953, 0.5133, 0.4375, 0.5452, 0.6052, 0.5679],
  25049. [0.5969, 0.4074, 0.7429, 0.3088, 0.4482, 0.2296, 0.5628, 0.5750],
  25050. [0.6409, 0.4072, 0.8550, 0.6127, 0.4090, 0.4920, 0.5700, 0.5451],
  25051. [0.6599, 0.4126, 0.7889, 0.1861, 0.4329, 0.2702, 0.6305, 0.5054]],
  25052. device='cuda:0', grad_fn=<AddmmBackward>)
  25053. landmarks are: tensor([[[0.6064, 0.3952, 0.8737, 0.4417, 0.3663, 0.4683, 0.5511, 0.5416],
  25054. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  25055. [0.6250, 0.4110, 0.7237, 0.2067, 0.4263, 0.1883, 0.5625, 0.5633],
  25056. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5462, 0.5800],
  25057. [0.6201, 0.4064, 0.8687, 0.5050, 0.4225, 0.5100, 0.6137, 0.5500],
  25058. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  25059. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373],
  25060. [0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947]]],
  25061. device='cuda:0')
  25062. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  25063. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  25064. loss_train: 0.007774444384267554
  25065. step: 10
  25066. running loss: 0.0007774444384267554
  25067. Train Steps: 10/90 Loss: 0.0008 torch.Size([8, 600, 800])
  25068. torch.Size([8, 8])
  25069. tensor([[0.6057, 0.4011, 0.8750, 0.4267, 0.4400, 0.5800, 0.5845, 0.5585],
  25070. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  25071. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323],
  25072. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  25073. [0.6068, 0.3963, 0.8650, 0.4317, 0.4037, 0.5083, 0.5253, 0.4999],
  25074. [0.6182, 0.4099, 0.7812, 0.3000, 0.3937, 0.2367, 0.5325, 0.5750],
  25075. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  25076. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749]],
  25077. device='cuda:0', dtype=torch.float64)
  25078. predictions are: tensor([[0.6385, 0.4033, 0.8655, 0.4518, 0.4351, 0.5511, 0.6074, 0.5391],
  25079. [0.6870, 0.4239, 0.8564, 0.5643, 0.3862, 0.4419, 0.5884, 0.5476],
  25080. [0.5826, 0.3651, 0.7670, 0.2289, 0.4291, 0.1928, 0.6284, 0.5399],
  25081. [0.6836, 0.4302, 0.8857, 0.5144, 0.3720, 0.4463, 0.5233, 0.5645],
  25082. [0.6096, 0.3817, 0.8424, 0.4521, 0.3899, 0.4873, 0.5573, 0.5129],
  25083. [0.6418, 0.4088, 0.7759, 0.3049, 0.4048, 0.2246, 0.5228, 0.5798],
  25084. [0.6431, 0.4098, 0.8683, 0.4738, 0.4422, 0.4891, 0.5520, 0.5079],
  25085. [0.6509, 0.4179, 0.7390, 0.3151, 0.3359, 0.3168, 0.5603, 0.5707]],
  25086. device='cuda:0', grad_fn=<AddmmBackward>)
  25087. landmarks are: tensor([[[0.6057, 0.4011, 0.8750, 0.4267, 0.4400, 0.5800, 0.5845, 0.5585],
  25088. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  25089. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323],
  25090. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  25091. [0.6068, 0.3963, 0.8650, 0.4317, 0.4038, 0.5083, 0.5253, 0.4999],
  25092. [0.6182, 0.4099, 0.7812, 0.3000, 0.3938, 0.2367, 0.5325, 0.5750],
  25093. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  25094. [0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749]]],
  25095. device='cuda:0')
  25096. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  25097. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  25098. loss_train: 0.008245526551036164
  25099. step: 11
  25100. running loss: 0.0007495933228214694
  25101. Train Steps: 11/90 Loss: 0.0007 torch.Size([8, 600, 800])
  25102. torch.Size([8, 8])
  25103. tensor([[0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  25104. [0.6299, 0.4303, 0.7963, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  25105. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373],
  25106. [0.6250, 0.4131, 0.8688, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  25107. [0.6275, 0.4003, 0.9100, 0.3783, 0.4388, 0.3133, 0.7058, 0.5343],
  25108. [ nan, nan, 0.7525, 0.2291, 0.3838, 0.3017, 0.6050, 0.5667],
  25109. [0.6187, 0.4104, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  25110. [0.6186, 0.4060, 0.8750, 0.5050, 0.3538, 0.4367, 0.5813, 0.6083]],
  25111. device='cuda:0', dtype=torch.float64)
  25112. predictions are: tensor([[0.6789, 0.4254, 0.8802, 0.5339, 0.3951, 0.4785, 0.5689, 0.5207],
  25113. [0.6904, 0.4412, 0.7863, 0.3883, 0.4789, 0.2596, 0.4998, 0.6308],
  25114. [0.6794, 0.4258, 0.8299, 0.5935, 0.3994, 0.4723, 0.5408, 0.5457],
  25115. [0.6904, 0.4317, 0.8464, 0.3005, 0.4299, 0.2478, 0.5632, 0.5375],
  25116. [0.6913, 0.4495, 0.8708, 0.3862, 0.4187, 0.3195, 0.6650, 0.5311],
  25117. [0.0830, 0.0339, 0.7411, 0.2390, 0.3734, 0.2989, 0.5673, 0.5606],
  25118. [0.6843, 0.4467, 0.6902, 0.2188, 0.3999, 0.2534, 0.5375, 0.5657],
  25119. [0.7003, 0.4519, 0.8651, 0.5073, 0.3672, 0.4434, 0.5544, 0.6115]],
  25120. device='cuda:0', grad_fn=<AddmmBackward>)
  25121. landmarks are: tensor([[[0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  25122. [0.6299, 0.4303, 0.7962, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  25123. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373],
  25124. [0.6250, 0.4131, 0.8687, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  25125. [0.6275, 0.4003, 0.9100, 0.3783, 0.4387, 0.3133, 0.7058, 0.5343],
  25126. [0.0000, 0.0000, 0.7525, 0.2291, 0.3837, 0.3017, 0.6050, 0.5667],
  25127. [0.6187, 0.4103, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  25128. [0.6186, 0.4060, 0.8750, 0.5050, 0.3537, 0.4367, 0.5813, 0.6083]]],
  25129. device='cuda:0')
  25130. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  25131.  
  25132. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  25133. loss_train: 0.009264791355235502
  25134. step: 12
  25135. running loss: 0.0007720659462696252
  25136. Train Steps: 12/90 Loss: 0.0008 torch.Size([8, 600, 800])
  25137. torch.Size([8, 8])
  25138. tensor([[0.6249, 0.4142, 0.8350, 0.3283, 0.3613, 0.3700, 0.6188, 0.5400],
  25139. [0.6113, 0.4088, 0.6859, 0.2208, 0.4363, 0.1700, 0.5188, 0.5533],
  25140. [0.6250, 0.4131, 0.8688, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  25141. [0.6040, 0.4002, 0.7338, 0.2267, 0.3975, 0.2100, 0.5231, 0.4778],
  25142. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  25143. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  25144. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667],
  25145. [0.6286, 0.3977, 0.9038, 0.4733, 0.3900, 0.4150, 0.7074, 0.5320]],
  25146. device='cuda:0', dtype=torch.float64)
  25147. predictions are: tensor([[0.6438, 0.4170, 0.8441, 0.3354, 0.3635, 0.3598, 0.5761, 0.5718],
  25148. [0.5178, 0.3205, 0.6909, 0.2413, 0.4453, 0.1691, 0.4824, 0.5813],
  25149. [0.6444, 0.4104, 0.8591, 0.3085, 0.4397, 0.2368, 0.5545, 0.5576],
  25150. [0.6467, 0.4118, 0.7465, 0.2338, 0.4028, 0.2046, 0.5104, 0.5042],
  25151. [0.6424, 0.4062, 0.8770, 0.4743, 0.4497, 0.5087, 0.5798, 0.5563],
  25152. [0.6998, 0.4514, 0.8549, 0.5256, 0.4192, 0.5506, 0.6577, 0.6106],
  25153. [0.5969, 0.3950, 0.8450, 0.3362, 0.3615, 0.3658, 0.5247, 0.6014],
  25154. [0.6412, 0.4060, 0.8985, 0.4726, 0.3843, 0.4225, 0.6617, 0.5409]],
  25155. device='cuda:0', grad_fn=<AddmmBackward>)
  25156. landmarks are: tensor([[[0.6249, 0.4142, 0.8350, 0.3283, 0.3613, 0.3700, 0.6187, 0.5400],
  25157. [0.6113, 0.4088, 0.6859, 0.2208, 0.4363, 0.1700, 0.5188, 0.5533],
  25158. [0.6250, 0.4131, 0.8687, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  25159. [0.6040, 0.4002, 0.7337, 0.2267, 0.3975, 0.2100, 0.5231, 0.4778],
  25160. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  25161. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  25162. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667],
  25163. [0.6286, 0.3977, 0.9038, 0.4733, 0.3900, 0.4150, 0.7074, 0.5320]]],
  25164. device='cuda:0')
  25165. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  25166. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  25167. loss_train: 0.010133300122106448
  25168. step: 13
  25169. running loss: 0.000779484624777419
  25170. Train Steps: 13/90 Loss: 0.0008 torch.Size([8, 600, 800])
  25171. torch.Size([8, 8])
  25172. tensor([[0.6265, 0.4091, 0.8950, 0.3533, 0.3600, 0.3967, 0.6295, 0.4901],
  25173. [0.6168, 0.4029, 0.8523, 0.3417, 0.3588, 0.5000, 0.6125, 0.5400],
  25174. [0.6193, 0.4050, 0.7313, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  25175. [0.6082, 0.4024, 0.8738, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  25176. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  25177. [0.6164, 0.4076, 0.8838, 0.4117, 0.3713, 0.5550, 0.6238, 0.5350],
  25178. [0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950],
  25179. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167]],
  25180. device='cuda:0', dtype=torch.float64)
  25181. predictions are: tensor([[0.6662, 0.4234, 0.9031, 0.3694, 0.3580, 0.3724, 0.6157, 0.5550],
  25182. [0.6267, 0.4100, 0.8555, 0.3439, 0.3576, 0.4699, 0.6097, 0.5852],
  25183. [0.6212, 0.4056, 0.7296, 0.2370, 0.4127, 0.2023, 0.5632, 0.5981],
  25184. [0.5978, 0.3986, 0.8732, 0.4130, 0.3557, 0.3778, 0.5035, 0.5433],
  25185. [0.6166, 0.4000, 0.6922, 0.2149, 0.4322, 0.1552, 0.5333, 0.5524],
  25186. [0.6210, 0.3970, 0.8759, 0.4106, 0.3713, 0.5390, 0.6254, 0.5757],
  25187. [0.6666, 0.4193, 0.8550, 0.5351, 0.4326, 0.4919, 0.5749, 0.5434],
  25188. [0.6432, 0.4139, 0.8767, 0.5163, 0.4053, 0.4915, 0.5790, 0.5741]],
  25189. device='cuda:0', grad_fn=<AddmmBackward>)
  25190. landmarks are: tensor([[[0.6265, 0.4091, 0.8950, 0.3533, 0.3600, 0.3967, 0.6295, 0.4901],
  25191. [0.6168, 0.4029, 0.8523, 0.3417, 0.3587, 0.5000, 0.6125, 0.5400],
  25192. [0.6193, 0.4050, 0.7312, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  25193. [0.6082, 0.4024, 0.8737, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  25194. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  25195. [0.6164, 0.4076, 0.8838, 0.4117, 0.3713, 0.5550, 0.6237, 0.5350],
  25196. [0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950],
  25197. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167]]],
  25198. device='cuda:0')
  25199. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  25200. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  25201. loss_train: 0.010566788056166843
  25202. step: 14
  25203. running loss: 0.0007547705754404888
  25204. Train Steps: 14/90 Loss: 0.0008 torch.Size([8, 600, 800])
  25205. torch.Size([8, 8])
  25206. tensor([[0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153],
  25207. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  25208. [0.6267, 0.4080, 0.8438, 0.2633, 0.4763, 0.1800, 0.6259, 0.5240],
  25209. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  25210. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  25211. [0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  25212. [0.6259, 0.4133, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  25213. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540]],
  25214. device='cuda:0', dtype=torch.float64)
  25215. predictions are: tensor([[0.6224, 0.4058, 0.8545, 0.4940, 0.4377, 0.5267, 0.6175, 0.5340],
  25216. [0.5924, 0.3953, 0.8547, 0.4568, 0.4137, 0.4933, 0.5830, 0.5806],
  25217. [0.5722, 0.3748, 0.8269, 0.2542, 0.4515, 0.2007, 0.6299, 0.5232],
  25218. [0.5919, 0.3963, 0.8611, 0.2738, 0.3905, 0.2842, 0.5890, 0.5384],
  25219. [0.5859, 0.3965, 0.8591, 0.4376, 0.3848, 0.4553, 0.5460, 0.5158],
  25220. [0.6042, 0.4017, 0.8832, 0.4621, 0.3735, 0.4036, 0.5302, 0.5805],
  25221. [0.5670, 0.3899, 0.8042, 0.2198, 0.4908, 0.1637, 0.6256, 0.5006],
  25222. [0.6143, 0.4162, 0.8901, 0.4287, 0.3958, 0.3150, 0.6744, 0.5654]],
  25223. device='cuda:0', grad_fn=<AddmmBackward>)
  25224. landmarks are: tensor([[[0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153],
  25225. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  25226. [0.6267, 0.4080, 0.8438, 0.2633, 0.4762, 0.1800, 0.6259, 0.5240],
  25227. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  25228. [0.6086, 0.3998, 0.8788, 0.4450, 0.4025, 0.4650, 0.5306, 0.5103],
  25229. [0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  25230. [0.6259, 0.4132, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  25231. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540]]],
  25232. device='cuda:0')
  25233. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  25234. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  25235. loss_train: 0.0108872129349038
  25236. step: 15
  25237. running loss: 0.0007258141956602534
  25238. Train Steps: 15/90 Loss: 0.0007 torch.Size([8, 600, 800])
  25239. torch.Size([8, 8])
  25240. tensor([[0.6193, 0.4079, 0.7288, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  25241. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  25242. [ nan, nan, 0.8888, 0.3100, 0.5262, 0.2817, 0.7145, 0.6003],
  25243. [0.6259, 0.4133, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  25244. [0.6277, 0.4057, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  25245. [0.6098, 0.3991, 0.8638, 0.4717, 0.4263, 0.4967, 0.5212, 0.5650],
  25246. [0.6182, 0.3930, 0.8841, 0.3892, 0.3556, 0.4967, 0.6222, 0.5279],
  25247. [0.6353, 0.4128, 0.9138, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991]],
  25248. device='cuda:0', dtype=torch.float64)
  25249. predictions are: tensor([[0.6015, 0.4170, 0.7197, 0.2435, 0.4179, 0.2499, 0.5913, 0.6008],
  25250. [0.6438, 0.4261, 0.8888, 0.4595, 0.4274, 0.5851, 0.6316, 0.5029],
  25251. [0.0754, 0.0515, 0.8658, 0.2984, 0.4927, 0.2718, 0.7197, 0.5535],
  25252. [0.6106, 0.4292, 0.8134, 0.2247, 0.4864, 0.1641, 0.6272, 0.4792],
  25253. [0.6354, 0.4273, 0.8239, 0.2510, 0.4067, 0.1870, 0.6059, 0.4785],
  25254. [0.6247, 0.4177, 0.8501, 0.4693, 0.4100, 0.5023, 0.5431, 0.5304],
  25255. [0.6193, 0.3932, 0.8688, 0.3921, 0.3346, 0.4652, 0.6326, 0.5106],
  25256. [0.6273, 0.4316, 0.8989, 0.3456, 0.4486, 0.3181, 0.7250, 0.5764]],
  25257. device='cuda:0', grad_fn=<AddmmBackward>)
  25258. landmarks are: tensor([[[0.6193, 0.4078, 0.7287, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  25259. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  25260. [0.0000, 0.0000, 0.8888, 0.3100, 0.5263, 0.2817, 0.7145, 0.6003],
  25261. [0.6259, 0.4132, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  25262. [0.6277, 0.4056, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  25263. [0.6098, 0.3991, 0.8637, 0.4717, 0.4263, 0.4967, 0.5213, 0.5650],
  25264. [0.6182, 0.3930, 0.8841, 0.3892, 0.3556, 0.4967, 0.6222, 0.5279],
  25265. [0.6353, 0.4128, 0.9137, 0.3533, 0.4688, 0.3250, 0.7145, 0.5991]]],
  25266. device='cuda:0')
  25267. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  25268. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  25269. loss_train: 0.011273512180196121
  25270. step: 16
  25271. running loss: 0.0007045945112622576
  25272.  
  25273. Train Steps: 16/90 Loss: 0.0007 torch.Size([8, 600, 800])
  25274. torch.Size([8, 8])
  25275. tensor([[0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6262, 0.5367],
  25276. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  25277. [0.6261, 0.4045, 0.8865, 0.5369, 0.3895, 0.4859, 0.6683, 0.5249],
  25278. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  25279. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  25280. [0.6128, 0.4022, 0.8738, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  25281. [0.6167, 0.4048, 0.6831, 0.3639, 0.3763, 0.3017, 0.5700, 0.5883],
  25282. [0.6179, 0.4082, 0.6688, 0.2667, 0.3588, 0.3317, 0.5750, 0.5783]],
  25283. device='cuda:0', dtype=torch.float64)
  25284. predictions are: tensor([[0.5943, 0.3980, 0.9217, 0.4521, 0.4279, 0.5748, 0.6315, 0.5061],
  25285. [0.5945, 0.3896, 0.8918, 0.3888, 0.3856, 0.4484, 0.6115, 0.5012],
  25286. [0.5628, 0.3809, 0.9031, 0.5169, 0.3854, 0.4670, 0.6718, 0.5036],
  25287. [0.5827, 0.3990, 0.7920, 0.2506, 0.3870, 0.2685, 0.5760, 0.5261],
  25288. [0.5897, 0.4110, 0.8753, 0.2631, 0.4937, 0.1924, 0.6866, 0.5137],
  25289. [0.5579, 0.3951, 0.8914, 0.4911, 0.5017, 0.4916, 0.5551, 0.4862],
  25290. [0.5632, 0.3852, 0.7218, 0.3323, 0.3831, 0.2869, 0.6009, 0.5554],
  25291. [0.6234, 0.4210, 0.6913, 0.2689, 0.3568, 0.2974, 0.5792, 0.5452]],
  25292. device='cuda:0', grad_fn=<AddmmBackward>)
  25293. landmarks are: tensor([[[0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6263, 0.5367],
  25294. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  25295. [0.6261, 0.4045, 0.8865, 0.5369, 0.3895, 0.4859, 0.6683, 0.5249],
  25296. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  25297. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  25298. [0.6128, 0.4022, 0.8737, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  25299. [0.6167, 0.4048, 0.6831, 0.3639, 0.3762, 0.3017, 0.5700, 0.5883],
  25300. [0.6179, 0.4082, 0.6687, 0.2667, 0.3587, 0.3317, 0.5750, 0.5783]]],
  25301. device='cuda:0')
  25302. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  25303. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  25304. loss_train: 0.011899679695488885
  25305. step: 17
  25306. running loss: 0.0006999811585581698
  25307. Train Steps: 17/90 Loss: 0.0007 torch.Size([8, 600, 800])
  25308. torch.Size([8, 8])
  25309. tensor([[0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297],
  25310. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6362, 0.5200],
  25311. [0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  25312. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  25313. [0.6160, 0.4093, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  25314. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  25315. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290],
  25316. [0.6296, 0.4076, 0.8400, 0.5583, 0.3700, 0.4367, 0.6876, 0.5494]],
  25317. device='cuda:0', dtype=torch.float64)
  25318. predictions are: tensor([[0.5759, 0.3881, 0.8742, 0.5237, 0.3796, 0.4402, 0.6483, 0.5152],
  25319. [0.5830, 0.3891, 0.8870, 0.3364, 0.3729, 0.5032, 0.6474, 0.4874],
  25320. [0.6178, 0.4078, 0.8628, 0.5105, 0.4526, 0.5175, 0.6762, 0.5152],
  25321. [0.6367, 0.4283, 0.8008, 0.3185, 0.3739, 0.3252, 0.6114, 0.4945],
  25322. [0.5898, 0.3955, 0.8429, 0.4416, 0.3837, 0.4432, 0.5401, 0.5572],
  25323. [0.5827, 0.3843, 0.8635, 0.4610, 0.4559, 0.4643, 0.5379, 0.5216],
  25324. [0.6468, 0.4418, 0.7687, 0.1860, 0.4526, 0.1878, 0.6652, 0.5080],
  25325. [0.6178, 0.4185, 0.8519, 0.5210, 0.3906, 0.4284, 0.6882, 0.5288]],
  25326. device='cuda:0', grad_fn=<AddmmBackward>)
  25327. landmarks are: tensor([[[0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297],
  25328. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6363, 0.5200],
  25329. [0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  25330. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  25331. [0.6160, 0.4092, 0.8314, 0.4417, 0.3675, 0.4583, 0.5250, 0.5808],
  25332. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  25333. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290],
  25334. [0.6296, 0.4076, 0.8400, 0.5583, 0.3700, 0.4367, 0.6876, 0.5494]]],
  25335. device='cuda:0')
  25336. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  25337. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  25338. loss_train: 0.012323376664426178
  25339. step: 18
  25340. running loss: 0.0006846320369125655
  25341. Train Steps: 18/90 Loss: 0.0007 torch.Size([8, 600, 800])
  25342. torch.Size([8, 8])
  25343. tensor([[0.6286, 0.4078, 0.8063, 0.2267, 0.4788, 0.1533, 0.5953, 0.4913],
  25344. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  25345. [0.6275, 0.4024, 0.8500, 0.5383, 0.3912, 0.4883, 0.6288, 0.5100],
  25346. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290],
  25347. [ nan, nan, 0.8488, 0.2300, 0.5563, 0.2100, 0.7390, 0.5679],
  25348. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  25349. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  25350. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540]],
  25351. device='cuda:0', dtype=torch.float64)
  25352. predictions are: tensor([[0.5797, 0.3699, 0.8009, 0.2417, 0.4690, 0.1885, 0.6042, 0.4845],
  25353. [0.5711, 0.3744, 0.8285, 0.5885, 0.4178, 0.4771, 0.5702, 0.5147],
  25354. [0.6353, 0.3908, 0.8537, 0.5339, 0.4054, 0.5166, 0.6063, 0.4817],
  25355. [0.6587, 0.4327, 0.7637, 0.2214, 0.4382, 0.2271, 0.6492, 0.5154],
  25356. [0.0846, 0.0291, 0.8546, 0.2416, 0.5450, 0.2666, 0.7472, 0.5351],
  25357. [0.6410, 0.4245, 0.7340, 0.1888, 0.4015, 0.2636, 0.6165, 0.5394],
  25358. [0.6169, 0.4116, 0.8886, 0.3509, 0.3904, 0.3113, 0.5855, 0.5371],
  25359. [0.6510, 0.4297, 0.9147, 0.4437, 0.4104, 0.3270, 0.6731, 0.5441]],
  25360. device='cuda:0', grad_fn=<AddmmBackward>)
  25361. landmarks are: tensor([[[0.6286, 0.4078, 0.8062, 0.2267, 0.4787, 0.1533, 0.5953, 0.4913],
  25362. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  25363. [0.6275, 0.4024, 0.8500, 0.5383, 0.3913, 0.4883, 0.6288, 0.5100],
  25364. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290],
  25365. [0.0000, 0.0000, 0.8487, 0.2300, 0.5562, 0.2100, 0.7390, 0.5679],
  25366. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  25367. [0.6203, 0.4072, 0.8892, 0.3523, 0.3783, 0.3017, 0.5898, 0.5478],
  25368. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540]]],
  25369. device='cuda:0')
  25370. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  25371. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  25372. loss_train: 0.012789542262908071
  25373. step: 19
  25374. running loss: 0.0006731338033109511
  25375. Train Steps: 19/90 Loss: 0.0007 torch.Size([8, 600, 800])
  25376. torch.Size([8, 8])
  25377. tensor([[0.6213, 0.4131, 0.8438, 0.3550, 0.3513, 0.4400, 0.5716, 0.5123],
  25378. [0.6346, 0.4086, 0.7938, 0.5500, 0.3962, 0.4867, 0.7343, 0.5702],
  25379. [0.6266, 0.4067, 0.8588, 0.2867, 0.4300, 0.2850, 0.6325, 0.5267],
  25380. [ nan, nan, 0.8463, 0.2550, 0.5850, 0.2133, 0.7129, 0.6072],
  25381. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5988, 0.5667],
  25382. [0.6064, 0.3953, 0.8738, 0.4417, 0.3663, 0.4683, 0.5511, 0.5416],
  25383. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  25384. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967]],
  25385. device='cuda:0', dtype=torch.float64)
  25386. predictions are: tensor([[0.6144, 0.4006, 0.8502, 0.3668, 0.3638, 0.4422, 0.5844, 0.4963],
  25387. [0.6249, 0.4046, 0.8022, 0.5589, 0.3946, 0.4815, 0.7036, 0.5593],
  25388. [0.6223, 0.4128, 0.8770, 0.2965, 0.4358, 0.2833, 0.6508, 0.5138],
  25389. [0.0960, 0.0492, 0.8665, 0.2693, 0.5726, 0.2490, 0.7453, 0.5857],
  25390. [0.6294, 0.4175, 0.8569, 0.3660, 0.3810, 0.3529, 0.6021, 0.5679],
  25391. [0.6125, 0.3823, 0.8703, 0.4571, 0.3776, 0.4691, 0.5518, 0.5235],
  25392. [0.6238, 0.4022, 0.8976, 0.3624, 0.3795, 0.4454, 0.6836, 0.5143],
  25393. [0.6391, 0.4209, 0.8698, 0.5133, 0.3780, 0.3417, 0.6381, 0.5009]],
  25394. device='cuda:0', grad_fn=<AddmmBackward>)
  25395. landmarks are: tensor([[[0.6213, 0.4131, 0.8438, 0.3550, 0.3512, 0.4400, 0.5716, 0.5123],
  25396. [0.6346, 0.4086, 0.7937, 0.5500, 0.3963, 0.4867, 0.7343, 0.5702],
  25397. [0.6266, 0.4067, 0.8587, 0.2867, 0.4300, 0.2850, 0.6325, 0.5267],
  25398. [0.0000, 0.0000, 0.8462, 0.2550, 0.5850, 0.2133, 0.7129, 0.6072],
  25399. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5987, 0.5667],
  25400. [0.6064, 0.3952, 0.8737, 0.4417, 0.3663, 0.4683, 0.5511, 0.5416],
  25401. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  25402. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967]]],
  25403. device='cuda:0')
  25404. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  25405. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  25406. loss_train: 0.013119590788846835
  25407. step: 20
  25408. running loss: 0.0006559795394423418
  25409.  
  25410. Train Steps: 20/90 Loss: 0.0007 torch.Size([8, 600, 800])
  25411. torch.Size([8, 8])
  25412. tensor([[0.6250, 0.4131, 0.8688, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  25413. [0.6272, 0.4045, 0.8538, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  25414. [0.6200, 0.4059, 0.8700, 0.4900, 0.4163, 0.5000, 0.6162, 0.5467],
  25415. [0.6255, 0.4017, 0.8688, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  25416. [0.6124, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  25417. [0.6200, 0.4118, 0.8287, 0.4017, 0.3775, 0.2833, 0.5391, 0.5799],
  25418. [0.6308, 0.3990, 0.8688, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  25419. [0.6147, 0.4026, 0.6600, 0.2467, 0.4088, 0.2150, 0.5489, 0.5773]],
  25420. device='cuda:0', dtype=torch.float64)
  25421. predictions are: tensor([[0.6296, 0.4074, 0.8948, 0.3214, 0.4477, 0.2529, 0.6697, 0.5555],
  25422. [0.6054, 0.3959, 0.8685, 0.5957, 0.3754, 0.4682, 0.6372, 0.4977],
  25423. [0.5450, 0.3340, 0.8945, 0.4939, 0.4314, 0.5339, 0.6315, 0.5598],
  25424. [0.5763, 0.3613, 0.8921, 0.3362, 0.3809, 0.3682, 0.6608, 0.5188],
  25425. [0.6269, 0.4156, 0.7311, 0.3101, 0.3836, 0.3070, 0.5511, 0.5822],
  25426. [0.5670, 0.3752, 0.8318, 0.4177, 0.3885, 0.3006, 0.5604, 0.5971],
  25427. [0.6255, 0.3709, 0.8807, 0.5453, 0.3992, 0.5174, 0.6570, 0.5380],
  25428. [0.5794, 0.3761, 0.6974, 0.2540, 0.4204, 0.2369, 0.5726, 0.6080]],
  25429. device='cuda:0', grad_fn=<AddmmBackward>)
  25430. landmarks are: tensor([[[0.6250, 0.4131, 0.8687, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  25431. [0.6271, 0.4045, 0.8537, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  25432. [0.6199, 0.4059, 0.8700, 0.4900, 0.4162, 0.5000, 0.6162, 0.5467],
  25433. [0.6255, 0.4017, 0.8687, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  25434. [0.6123, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  25435. [0.6200, 0.4118, 0.8288, 0.4017, 0.3775, 0.2833, 0.5391, 0.5799],
  25436. [0.6308, 0.3990, 0.8687, 0.5183, 0.3950, 0.4983, 0.6388, 0.5133],
  25437. [0.6147, 0.4026, 0.6600, 0.2467, 0.4087, 0.2150, 0.5489, 0.5773]]],
  25438. device='cuda:0')
  25439. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  25440. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  25441. loss_train: 0.013877765712095425
  25442. step: 21
  25443. running loss: 0.0006608459862902583
  25444. Train Steps: 21/90 Loss: 0.0007 torch.Size([8, 600, 800])
  25445. torch.Size([8, 8])
  25446. tensor([[0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  25447. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  25448. [0.6241, 0.4143, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  25449. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050],
  25450. [0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717],
  25451. [0.6325, 0.4165, 0.9000, 0.4617, 0.3813, 0.4900, 0.7485, 0.5447],
  25452. [0.6275, 0.4024, 0.8500, 0.5383, 0.3912, 0.4883, 0.6288, 0.5100],
  25453. [0.6178, 0.4059, 0.7525, 0.2250, 0.4313, 0.1783, 0.5404, 0.5240]],
  25454. device='cuda:0', dtype=torch.float64)
  25455. predictions are: tensor([[0.5650, 0.3551, 0.8819, 0.5313, 0.4009, 0.5310, 0.5723, 0.5404],
  25456. [0.5824, 0.3737, 0.9007, 0.3910, 0.4633, 0.2645, 0.6485, 0.5586],
  25457. [0.6109, 0.3947, 0.8851, 0.4863, 0.3972, 0.5470, 0.6387, 0.5904],
  25458. [0.5914, 0.3826, 0.8671, 0.5249, 0.4320, 0.5236, 0.5967, 0.5469],
  25459. [0.6290, 0.4103, 0.7343, 0.2679, 0.3852, 0.2887, 0.5911, 0.6182],
  25460. [0.5890, 0.3860, 0.9081, 0.4825, 0.3769, 0.5022, 0.7166, 0.5768],
  25461. [0.6258, 0.3809, 0.8503, 0.5562, 0.3911, 0.5099, 0.6123, 0.5244],
  25462. [0.5949, 0.3755, 0.7443, 0.2458, 0.4314, 0.1738, 0.5385, 0.5607]],
  25463. device='cuda:0', grad_fn=<AddmmBackward>)
  25464. landmarks are: tensor([[[0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  25465. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  25466. [0.6241, 0.4142, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  25467. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050],
  25468. [0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717],
  25469. [0.6325, 0.4165, 0.9000, 0.4617, 0.3812, 0.4900, 0.7485, 0.5447],
  25470. [0.6275, 0.4024, 0.8500, 0.5383, 0.3913, 0.4883, 0.6288, 0.5100],
  25471. [0.6178, 0.4059, 0.7525, 0.2250, 0.4313, 0.1783, 0.5404, 0.5240]]],
  25472. device='cuda:0')
  25473. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  25474. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  25475. loss_train: 0.014404586370801553
  25476. step: 22
  25477. running loss: 0.0006547539259455251
  25478. Train Steps: 22/90 Loss: 0.0007 torch.Size([8, 600, 800])
  25479. torch.Size([8, 8])
  25480. tensor([[0.6289, 0.4032, 0.8419, 0.5446, 0.4075, 0.5017, 0.6312, 0.5117],
  25481. [0.6266, 0.4067, 0.8588, 0.2867, 0.4300, 0.2850, 0.6325, 0.5267],
  25482. [0.6124, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  25483. [0.6125, 0.3999, 0.8750, 0.4883, 0.4750, 0.4700, 0.5533, 0.5617],
  25484. [0.6090, 0.4010, 0.7838, 0.3483, 0.3538, 0.3783, 0.5462, 0.5077],
  25485. [0.6250, 0.4236, 0.8638, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  25486. [ nan, nan, 0.8850, 0.3000, 0.5363, 0.2250, 0.7343, 0.5771],
  25487. [0.6159, 0.4085, 0.6900, 0.2283, 0.4088, 0.1950, 0.5123, 0.5397]],
  25488. device='cuda:0', dtype=torch.float64)
  25489. predictions are: tensor([[0.6460, 0.3959, 0.8257, 0.5581, 0.3873, 0.4902, 0.6624, 0.5208],
  25490. [0.6283, 0.4119, 0.8619, 0.2948, 0.4182, 0.2608, 0.6523, 0.5370],
  25491. [0.6427, 0.4274, 0.6905, 0.2963, 0.3651, 0.2692, 0.5266, 0.5680],
  25492. [0.6403, 0.4086, 0.8583, 0.4989, 0.4620, 0.4748, 0.5454, 0.5695],
  25493. [0.6413, 0.4226, 0.7729, 0.3604, 0.3415, 0.3635, 0.5549, 0.5128],
  25494. [0.6557, 0.4392, 0.8538, 0.3977, 0.3870, 0.3068, 0.5797, 0.5940],
  25495. [0.0340, 0.0126, 0.8629, 0.2940, 0.5285, 0.2054, 0.7630, 0.5743],
  25496. [0.5121, 0.3272, 0.6654, 0.2389, 0.4041, 0.1704, 0.5190, 0.5577]],
  25497. device='cuda:0', grad_fn=<AddmmBackward>)
  25498. landmarks are: tensor([[[0.6289, 0.4031, 0.8419, 0.5446, 0.4075, 0.5017, 0.6313, 0.5117],
  25499. [0.6266, 0.4067, 0.8587, 0.2867, 0.4300, 0.2850, 0.6325, 0.5267],
  25500. [0.6123, 0.4083, 0.6954, 0.3069, 0.3650, 0.2750, 0.5163, 0.5485],
  25501. [0.6125, 0.3999, 0.8750, 0.4883, 0.4750, 0.4700, 0.5533, 0.5617],
  25502. [0.6090, 0.4010, 0.7837, 0.3483, 0.3537, 0.3783, 0.5462, 0.5077],
  25503. [0.6250, 0.4236, 0.8637, 0.3767, 0.4050, 0.3150, 0.5649, 0.5799],
  25504. [0.0000, 0.0000, 0.8850, 0.3000, 0.5362, 0.2250, 0.7343, 0.5771],
  25505. [0.6159, 0.4085, 0.6900, 0.2283, 0.4087, 0.1950, 0.5123, 0.5397]]],
  25506. device='cuda:0')
  25507. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  25508. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  25509. loss_train: 0.014939425949705765
  25510. step: 23
  25511. running loss: 0.0006495402586828593
  25512. Train Steps: 23/90 Loss: 0.0006 torch.Size([8, 600, 800])
  25513. torch.Size([8, 8])
  25514. tensor([[0.6144, 0.4032, 0.8563, 0.3283, 0.3525, 0.4200, 0.5775, 0.5583],
  25515. [0.6185, 0.4079, 0.8838, 0.4617, 0.4838, 0.5650, 0.6175, 0.5850],
  25516. [0.6188, 0.4099, 0.7400, 0.2433, 0.3962, 0.2750, 0.6162, 0.5467],
  25517. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  25518. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  25519. [0.6175, 0.3957, 0.8700, 0.4817, 0.4662, 0.5133, 0.5800, 0.5517],
  25520. [0.6198, 0.4101, 0.8838, 0.5283, 0.3763, 0.5267, 0.5913, 0.5567],
  25521. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391]],
  25522. device='cuda:0', dtype=torch.float64)
  25523. predictions are: tensor([[0.6195, 0.3847, 0.8599, 0.3413, 0.3459, 0.3976, 0.5567, 0.5486],
  25524. [0.6105, 0.3927, 0.8786, 0.4613, 0.4847, 0.5375, 0.6117, 0.5935],
  25525. [0.6538, 0.4311, 0.7202, 0.2353, 0.3908, 0.2772, 0.5970, 0.5718],
  25526. [0.6367, 0.4093, 0.8734, 0.4899, 0.4099, 0.5058, 0.6015, 0.5389],
  25527. [0.6352, 0.3884, 0.8787, 0.5097, 0.3985, 0.4984, 0.6342, 0.4904],
  25528. [0.6424, 0.4127, 0.8495, 0.4789, 0.4571, 0.4931, 0.5678, 0.5521],
  25529. [0.6069, 0.3848, 0.8498, 0.5368, 0.3684, 0.4988, 0.5818, 0.5520],
  25530. [0.6352, 0.3954, 0.8017, 0.2887, 0.4095, 0.2181, 0.5965, 0.5479]],
  25531. device='cuda:0', grad_fn=<AddmmBackward>)
  25532. landmarks are: tensor([[[0.6144, 0.4032, 0.8562, 0.3283, 0.3525, 0.4200, 0.5775, 0.5583],
  25533. [0.6184, 0.4079, 0.8838, 0.4617, 0.4837, 0.5650, 0.6175, 0.5850],
  25534. [0.6188, 0.4099, 0.7400, 0.2433, 0.3963, 0.2750, 0.6162, 0.5467],
  25535. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  25536. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  25537. [0.6175, 0.3957, 0.8700, 0.4817, 0.4663, 0.5133, 0.5800, 0.5517],
  25538. [0.6198, 0.4101, 0.8838, 0.5283, 0.3762, 0.5267, 0.5913, 0.5567],
  25539. [0.6201, 0.4027, 0.8029, 0.2728, 0.4042, 0.2310, 0.5980, 0.5391]]],
  25540. device='cuda:0')
  25541. loss_train_step before backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  25542. loss_train_step after backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  25543. loss_train: 0.015169626043643802
  25544. step: 24
  25545. running loss: 0.0006320677518184917
  25546.  
  25547. Train Steps: 24/90 Loss: 0.0006 torch.Size([8, 600, 800])
  25548. torch.Size([8, 8])
  25549. tensor([[0.6163, 0.4006, 0.8788, 0.4683, 0.3663, 0.4883, 0.5887, 0.5017],
  25550. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  25551. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864],
  25552. [0.6075, 0.4000, 0.8513, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  25553. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167],
  25554. [0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5988, 0.5700],
  25555. [0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  25556. [0.6115, 0.4081, 0.6725, 0.2433, 0.4088, 0.1933, 0.5167, 0.5544]],
  25557. device='cuda:0', dtype=torch.float64)
  25558. predictions are: tensor([[0.5821, 0.3851, 0.8686, 0.4562, 0.3777, 0.4854, 0.5571, 0.5129],
  25559. [0.5867, 0.3747, 0.8297, 0.5226, 0.4066, 0.5395, 0.6983, 0.5600],
  25560. [0.6231, 0.4036, 0.8612, 0.5096, 0.3793, 0.4699, 0.6989, 0.5680],
  25561. [0.6329, 0.3993, 0.8323, 0.5190, 0.4639, 0.5174, 0.5098, 0.4993],
  25562. [0.6596, 0.4292, 0.8100, 0.3271, 0.3733, 0.2925, 0.5534, 0.5063],
  25563. [0.6017, 0.3856, 0.7856, 0.2735, 0.3719, 0.3316, 0.5911, 0.5630],
  25564. [0.6342, 0.4121, 0.8748, 0.4167, 0.3754, 0.5315, 0.5385, 0.4856],
  25565. [0.6165, 0.3957, 0.6680, 0.2330, 0.4212, 0.1719, 0.5023, 0.5374]],
  25566. device='cuda:0', grad_fn=<AddmmBackward>)
  25567. landmarks are: tensor([[[0.6163, 0.4006, 0.8788, 0.4683, 0.3663, 0.4883, 0.5888, 0.5017],
  25568. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  25569. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864],
  25570. [0.6075, 0.4000, 0.8512, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  25571. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167],
  25572. [0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5987, 0.5700],
  25573. [0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  25574. [0.6115, 0.4081, 0.6725, 0.2433, 0.4087, 0.1933, 0.5167, 0.5544]]],
  25575. device='cuda:0')
  25576. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  25577. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  25578. loss_train: 0.015483570605283603
  25579. step: 25
  25580. running loss: 0.0006193428242113441
  25581. Train Steps: 25/90 Loss: 0.0006 torch.Size([8, 600, 800])
  25582. torch.Size([8, 8])
  25583. tensor([[0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  25584. [0.6126, 0.4073, 0.8750, 0.5133, 0.3800, 0.4333, 0.4986, 0.5378],
  25585. [0.6339, 0.4118, 0.7988, 0.5800, 0.3912, 0.4583, 0.7343, 0.5760],
  25586. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  25587. [ nan, nan, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733],
  25588. [0.6222, 0.3937, 0.8350, 0.5617, 0.4138, 0.4600, 0.5800, 0.5233],
  25589. [0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090],
  25590. [0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767]],
  25591. device='cuda:0', dtype=torch.float64)
  25592. predictions are: tensor([[ 0.6852, 0.4632, 0.8482, 0.2990, 0.3883, 0.3077, 0.5581, 0.5466],
  25593. [ 0.6258, 0.4264, 0.8568, 0.5108, 0.3806, 0.4525, 0.5085, 0.5110],
  25594. [ 0.6581, 0.4255, 0.7901, 0.5492, 0.3743, 0.4760, 0.7085, 0.5539],
  25595. [ 0.7054, 0.4497, 0.8871, 0.4031, 0.3478, 0.4665, 0.5962, 0.4949],
  25596. [-0.0766, -0.0439, 0.6816, 0.1964, 0.4734, 0.1699, 0.5512, 0.5810],
  25597. [ 0.6778, 0.4353, 0.8170, 0.5715, 0.4026, 0.4640, 0.5526, 0.5143],
  25598. [ 0.6876, 0.4435, 0.8741, 0.3700, 0.3973, 0.3027, 0.6253, 0.4798],
  25599. [ 0.6833, 0.4596, 0.8834, 0.4272, 0.3690, 0.3819, 0.5770, 0.5657]],
  25600. device='cuda:0', grad_fn=<AddmmBackward>)
  25601. landmarks are: tensor([[[0.6236, 0.4081, 0.8575, 0.3000, 0.3713, 0.3033, 0.5550, 0.5633],
  25602. [0.6126, 0.4073, 0.8750, 0.5133, 0.3800, 0.4333, 0.4986, 0.5378],
  25603. [0.6339, 0.4118, 0.7987, 0.5800, 0.3913, 0.4583, 0.7343, 0.5760],
  25604. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  25605. [0.0000, 0.0000, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733],
  25606. [0.6222, 0.3937, 0.8350, 0.5617, 0.4137, 0.4600, 0.5800, 0.5233],
  25607. [0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090],
  25608. [0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767]]],
  25609. device='cuda:0')
  25610. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  25611. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  25612. loss_train: 0.016324585973052308
  25613. step: 26
  25614. running loss: 0.0006278686912712426
  25615. Train Steps: 26/90 Loss: 0.0006 torch.Size([8, 600, 800])
  25616. torch.Size([8, 8])
  25617. tensor([[0.6286, 0.3977, 0.9038, 0.4733, 0.3900, 0.4150, 0.7074, 0.5320],
  25618. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  25619. [0.6275, 0.4157, 0.8337, 0.5800, 0.3763, 0.4200, 0.5547, 0.6125],
  25620. [0.6305, 0.3983, 0.8950, 0.4833, 0.3688, 0.4683, 0.6375, 0.5117],
  25621. [0.6097, 0.4024, 0.8488, 0.3717, 0.3875, 0.5517, 0.5836, 0.5591],
  25622. [0.6129, 0.4069, 0.8750, 0.5067, 0.3875, 0.4233, 0.5235, 0.5881],
  25623. [0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  25624. [0.6197, 0.4091, 0.8800, 0.4783, 0.3538, 0.4767, 0.5950, 0.5550]],
  25625. device='cuda:0', dtype=torch.float64)
  25626. predictions are: tensor([[0.6205, 0.3967, 0.9026, 0.4544, 0.3932, 0.4375, 0.6945, 0.5092],
  25627. [0.6907, 0.4562, 0.8337, 0.5767, 0.3931, 0.5079, 0.6054, 0.4719],
  25628. [0.6233, 0.4279, 0.8179, 0.5471, 0.3836, 0.4258, 0.5543, 0.6103],
  25629. [0.6515, 0.4255, 0.8997, 0.4725, 0.3777, 0.4727, 0.6153, 0.5013],
  25630. [0.6181, 0.4017, 0.8575, 0.3643, 0.4084, 0.5700, 0.5951, 0.5194],
  25631. [0.6669, 0.4463, 0.8475, 0.5062, 0.3903, 0.4386, 0.4891, 0.5614],
  25632. [0.6693, 0.4533, 0.8795, 0.4952, 0.3682, 0.4686, 0.5514, 0.5426],
  25633. [0.6131, 0.4099, 0.8660, 0.4607, 0.3811, 0.4951, 0.5762, 0.5430]],
  25634. device='cuda:0', grad_fn=<AddmmBackward>)
  25635. landmarks are: tensor([[[0.6286, 0.3977, 0.9038, 0.4733, 0.3900, 0.4150, 0.7074, 0.5320],
  25636. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  25637. [0.6275, 0.4157, 0.8338, 0.5800, 0.3762, 0.4200, 0.5547, 0.6125],
  25638. [0.6305, 0.3983, 0.8950, 0.4833, 0.3688, 0.4683, 0.6375, 0.5117],
  25639. [0.6097, 0.4024, 0.8487, 0.3717, 0.3875, 0.5517, 0.5836, 0.5591],
  25640. [0.6129, 0.4069, 0.8750, 0.5067, 0.3875, 0.4233, 0.5235, 0.5881],
  25641. [0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633],
  25642. [0.6197, 0.4091, 0.8800, 0.4783, 0.3537, 0.4767, 0.5950, 0.5550]]],
  25643. device='cuda:0')
  25644. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  25645. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  25646. loss_train: 0.016808000102173537
  25647. step: 27
  25648. running loss: 0.0006225185223027236
  25649. Train Steps: 27/90 Loss: 0.0006 torch.Size([8, 600, 800])
  25650. torch.Size([8, 8])
  25651. tensor([[0.6072, 0.4029, 0.7037, 0.2150, 0.3912, 0.2267, 0.5516, 0.5507],
  25652. [0.6187, 0.4104, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  25653. [0.6124, 0.4075, 0.7696, 0.4153, 0.3475, 0.3767, 0.5157, 0.5427],
  25654. [0.6145, 0.4008, 0.8750, 0.5383, 0.3975, 0.4650, 0.5563, 0.5533],
  25655. [ nan, nan, 0.8888, 0.3100, 0.5262, 0.2817, 0.7145, 0.6003],
  25656. [0.6113, 0.4088, 0.6859, 0.2208, 0.4363, 0.1700, 0.5188, 0.5533],
  25657. [0.6222, 0.4169, 0.8638, 0.5650, 0.4313, 0.4783, 0.5637, 0.5633],
  25658. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103]],
  25659. device='cuda:0', dtype=torch.float64)
  25660. predictions are: tensor([[ 0.6678, 0.4515, 0.6983, 0.1999, 0.3860, 0.2305, 0.5475, 0.5385],
  25661. [ 0.6567, 0.4500, 0.7093, 0.2060, 0.3911, 0.2487, 0.5784, 0.5428],
  25662. [ 0.6494, 0.4381, 0.7825, 0.4043, 0.3370, 0.3878, 0.5269, 0.5209],
  25663. [ 0.5652, 0.3896, 0.8628, 0.5309, 0.3852, 0.4684, 0.5787, 0.5374],
  25664. [-0.0491, -0.0215, 0.9024, 0.3033, 0.5342, 0.2410, 0.7110, 0.5998],
  25665. [ 0.7045, 0.4607, 0.7032, 0.2196, 0.4323, 0.1755, 0.5227, 0.5387],
  25666. [ 0.6341, 0.4296, 0.8694, 0.5635, 0.4121, 0.4887, 0.5844, 0.5403],
  25667. [ 0.6351, 0.4425, 0.8114, 0.2984, 0.3520, 0.3987, 0.5705, 0.5170]],
  25668. device='cuda:0', grad_fn=<AddmmBackward>)
  25669. landmarks are: tensor([[[0.6072, 0.4029, 0.7038, 0.2150, 0.3913, 0.2267, 0.5516, 0.5507],
  25670. [0.6187, 0.4103, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  25671. [0.6124, 0.4075, 0.7696, 0.4153, 0.3475, 0.3767, 0.5157, 0.5427],
  25672. [0.6145, 0.4008, 0.8750, 0.5383, 0.3975, 0.4650, 0.5562, 0.5533],
  25673. [0.0000, 0.0000, 0.8888, 0.3100, 0.5263, 0.2817, 0.7145, 0.6003],
  25674. [0.6113, 0.4088, 0.6859, 0.2208, 0.4363, 0.1700, 0.5188, 0.5533],
  25675. [0.6222, 0.4169, 0.8637, 0.5650, 0.4313, 0.4783, 0.5638, 0.5633],
  25676. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103]]],
  25677. device='cuda:0')
  25678. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  25679. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  25680. loss_train: 0.0173894782201387
  25681. step: 28
  25682. running loss: 0.0006210527935763821
  25683.  
  25684. Train Steps: 28/90 Loss: 0.0006 torch.Size([8, 600, 800])
  25685. torch.Size([8, 8])
  25686. tensor([[0.6346, 0.4165, 0.9138, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471],
  25687. [0.6240, 0.4217, 0.8150, 0.3133, 0.4425, 0.2650, 0.5650, 0.5817],
  25688. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  25689. [0.6277, 0.4029, 0.8250, 0.2433, 0.4325, 0.2100, 0.6366, 0.5207],
  25690. [ nan, nan, 0.7512, 0.2117, 0.4288, 0.2000, 0.5600, 0.5617],
  25691. [0.6097, 0.3988, 0.8650, 0.5250, 0.4213, 0.5200, 0.5675, 0.5050],
  25692. [0.6149, 0.4054, 0.6713, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  25693. [0.6239, 0.4123, 0.8313, 0.2550, 0.4500, 0.2050, 0.6175, 0.5400]],
  25694. device='cuda:0', dtype=torch.float64)
  25695. predictions are: tensor([[0.5899, 0.4050, 0.9158, 0.4174, 0.3807, 0.4620, 0.7336, 0.5475],
  25696. [0.5687, 0.4059, 0.8109, 0.3149, 0.4253, 0.2654, 0.5626, 0.5792],
  25697. [0.5833, 0.3953, 0.8511, 0.2648, 0.4415, 0.2807, 0.6835, 0.5603],
  25698. [0.6309, 0.4139, 0.8280, 0.2610, 0.4369, 0.2403, 0.6261, 0.5160],
  25699. [0.1151, 0.0865, 0.7205, 0.2217, 0.4265, 0.1935, 0.5202, 0.5690],
  25700. [0.6392, 0.4385, 0.8658, 0.5408, 0.4172, 0.5399, 0.5582, 0.5038],
  25701. [0.6263, 0.4299, 0.6765, 0.2299, 0.3918, 0.2112, 0.4994, 0.5611],
  25702. [0.6386, 0.4270, 0.8344, 0.2350, 0.4449, 0.1997, 0.6108, 0.5236]],
  25703. device='cuda:0', grad_fn=<AddmmBackward>)
  25704. landmarks are: tensor([[[0.6346, 0.4165, 0.9137, 0.3983, 0.3875, 0.4317, 0.7469, 0.5471],
  25705. [0.6240, 0.4217, 0.8150, 0.3133, 0.4425, 0.2650, 0.5650, 0.5817],
  25706. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  25707. [0.6277, 0.4029, 0.8250, 0.2433, 0.4325, 0.2100, 0.6366, 0.5207],
  25708. [0.0000, 0.0000, 0.7513, 0.2117, 0.4288, 0.2000, 0.5600, 0.5617],
  25709. [0.6097, 0.3988, 0.8650, 0.5250, 0.4212, 0.5200, 0.5675, 0.5050],
  25710. [0.6149, 0.4054, 0.6712, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  25711. [0.6239, 0.4123, 0.8313, 0.2550, 0.4500, 0.2050, 0.6175, 0.5400]]],
  25712. device='cuda:0')
  25713. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  25714. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  25715. loss_train: 0.018027903337497264
  25716. step: 29
  25717. running loss: 0.0006216518392240436
  25718. Train Steps: 29/90 Loss: 0.0006 torch.Size([8, 600, 800])
  25719. torch.Size([8, 8])
  25720. tensor([[0.6201, 0.4064, 0.8688, 0.5050, 0.4225, 0.5100, 0.6138, 0.5500],
  25721. [0.6118, 0.4052, 0.8463, 0.3917, 0.3538, 0.3450, 0.5053, 0.5593],
  25722. [ nan, nan, 0.7515, 0.2708, 0.3987, 0.2267, 0.5162, 0.5567],
  25723. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  25724. [0.6179, 0.4008, 0.8600, 0.4015, 0.3932, 0.2515, 0.5711, 0.5438],
  25725. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  25726. [0.6200, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  25727. [0.6135, 0.3994, 0.7913, 0.3050, 0.3625, 0.3050, 0.5837, 0.5050]],
  25728. device='cuda:0', dtype=torch.float64)
  25729. predictions are: tensor([[0.6289, 0.4156, 0.8779, 0.4982, 0.4202, 0.5133, 0.6175, 0.5748],
  25730. [0.6280, 0.4271, 0.8509, 0.3736, 0.3577, 0.3567, 0.4989, 0.5659],
  25731. [0.0764, 0.0679, 0.7582, 0.2679, 0.4156, 0.2245, 0.5097, 0.5756],
  25732. [0.6544, 0.4261, 0.9022, 0.5067, 0.4018, 0.5514, 0.7537, 0.5663],
  25733. [0.6425, 0.4273, 0.8398, 0.3622, 0.4102, 0.2657, 0.5652, 0.5576],
  25734. [0.6033, 0.3992, 0.8812, 0.3680, 0.3698, 0.3246, 0.5951, 0.5382],
  25735. [0.6100, 0.4027, 0.8141, 0.2939, 0.3899, 0.2922, 0.6108, 0.5352],
  25736. [0.6272, 0.4086, 0.7947, 0.2922, 0.3735, 0.3166, 0.5857, 0.5293]],
  25737. device='cuda:0', grad_fn=<AddmmBackward>)
  25738. landmarks are: tensor([[[0.6201, 0.4064, 0.8687, 0.5050, 0.4225, 0.5100, 0.6137, 0.5500],
  25739. [0.6118, 0.4052, 0.8462, 0.3917, 0.3537, 0.3450, 0.5053, 0.5593],
  25740. [0.0000, 0.0000, 0.7515, 0.2708, 0.3988, 0.2267, 0.5163, 0.5567],
  25741. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  25742. [0.6179, 0.4008, 0.8600, 0.4015, 0.3932, 0.2515, 0.5711, 0.5438],
  25743. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  25744. [0.6201, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  25745. [0.6135, 0.3994, 0.7912, 0.3050, 0.3625, 0.3050, 0.5838, 0.5050]]],
  25746. device='cuda:0')
  25747. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  25748. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  25749. loss_train: 0.018376984866335988
  25750. step: 30
  25751. running loss: 0.0006125661622111996
  25752. Train Steps: 30/90 Loss: 0.0006 torch.Size([8, 600, 800])
  25753. torch.Size([8, 8])
  25754. tensor([[0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5837, 0.5500],
  25755. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6188, 0.5283],
  25756. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  25757. [0.6142, 0.3982, 0.8650, 0.4883, 0.3912, 0.4317, 0.5315, 0.5350],
  25758. [0.6112, 0.4029, 0.8638, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  25759. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  25760. [0.6200, 0.3961, 0.8461, 0.5497, 0.4142, 0.4577, 0.5892, 0.5402],
  25761. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810]],
  25762. device='cuda:0', dtype=torch.float64)
  25763. predictions are: tensor([[0.5082, 0.3386, 0.9000, 0.4524, 0.4091, 0.5285, 0.5743, 0.5588],
  25764. [0.6539, 0.4280, 0.9310, 0.3621, 0.3786, 0.2677, 0.6350, 0.5476],
  25765. [0.5480, 0.3625, 0.6927, 0.2695, 0.4049, 0.2346, 0.5865, 0.5690],
  25766. [0.5836, 0.3790, 0.8909, 0.5103, 0.3678, 0.4543, 0.5583, 0.5349],
  25767. [0.5760, 0.3655, 0.8916, 0.4884, 0.4742, 0.5025, 0.5968, 0.5890],
  25768. [0.5642, 0.3593, 0.9249, 0.3504, 0.4558, 0.3376, 0.7363, 0.5549],
  25769. [0.5540, 0.3576, 0.8712, 0.5523, 0.3851, 0.4578, 0.6114, 0.5569],
  25770. [0.6044, 0.3882, 0.8784, 0.3101, 0.3934, 0.2648, 0.6318, 0.4963]],
  25771. device='cuda:0', grad_fn=<AddmmBackward>)
  25772. landmarks are: tensor([[[0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5838, 0.5500],
  25773. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6187, 0.5283],
  25774. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  25775. [0.6143, 0.3982, 0.8650, 0.4883, 0.3913, 0.4317, 0.5315, 0.5350],
  25776. [0.6112, 0.4029, 0.8637, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  25777. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  25778. [0.6200, 0.3961, 0.8461, 0.5497, 0.4142, 0.4577, 0.5892, 0.5402],
  25779. [0.6262, 0.4085, 0.8438, 0.3150, 0.4025, 0.2633, 0.6339, 0.4810]]],
  25780. device='cuda:0')
  25781. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  25782. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  25783. loss_train: 0.019246386946178973
  25784. step: 31
  25785. running loss: 0.0006208511918122249
  25786. Train Steps: 31/90 Loss: 0.0006 torch.Size([8, 600, 800])
  25787. torch.Size([8, 8])
  25788. tensor([[ nan, nan, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  25789. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116],
  25790. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  25791. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  25792. [0.6163, 0.4006, 0.8788, 0.4683, 0.3663, 0.4883, 0.5887, 0.5017],
  25793. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  25794. [0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274],
  25795. [0.6361, 0.4076, 0.8862, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297]],
  25796. device='cuda:0', dtype=torch.float64)
  25797. predictions are: tensor([[-0.0313, -0.0344, 0.6675, 0.2363, 0.4315, 0.1697, 0.5175, 0.5917],
  25798. [ 0.7018, 0.4575, 0.8709, 0.4227, 0.3885, 0.5999, 0.5903, 0.5508],
  25799. [ 0.6001, 0.3855, 0.9231, 0.4592, 0.4337, 0.3570, 0.7339, 0.6016],
  25800. [ 0.5920, 0.3758, 0.8785, 0.3899, 0.3640, 0.3185, 0.6041, 0.5439],
  25801. [ 0.6057, 0.3949, 0.8959, 0.4820, 0.3706, 0.4992, 0.6088, 0.5532],
  25802. [ 0.6054, 0.3812, 0.7647, 0.2485, 0.3684, 0.3098, 0.6237, 0.5450],
  25803. [ 0.6482, 0.4166, 0.8815, 0.3270, 0.4856, 0.1997, 0.6592, 0.5412],
  25804. [ 0.6420, 0.4174, 0.8970, 0.5579, 0.3575, 0.4708, 0.6795, 0.5610]],
  25805. device='cuda:0', grad_fn=<AddmmBackward>)
  25806. landmarks are: tensor([[[0.0000, 0.0000, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  25807. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116],
  25808. [0.6343, 0.4097, 0.9287, 0.4367, 0.4313, 0.3600, 0.7248, 0.5841],
  25809. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  25810. [0.6163, 0.4006, 0.8788, 0.4683, 0.3663, 0.4883, 0.5888, 0.5017],
  25811. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  25812. [0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274],
  25813. [0.6361, 0.4076, 0.8863, 0.5350, 0.3713, 0.4650, 0.6654, 0.5297]]],
  25814. device='cuda:0')
  25815. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  25816. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  25817. loss_train: 0.019706396793480963
  25818. step: 32
  25819. running loss: 0.0006158248997962801
  25820.  
  25821. Train Steps: 32/90 Loss: 0.0006 torch.Size([8, 600, 800])
  25822. torch.Size([8, 8])
  25823. tensor([[0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  25824. [0.6187, 0.4104, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  25825. [0.6097, 0.4000, 0.7325, 0.2667, 0.3450, 0.3517, 0.5284, 0.5045],
  25826. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  25827. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  25828. [0.6245, 0.4100, 0.7762, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  25829. [0.6275, 0.4024, 0.8500, 0.5383, 0.3912, 0.4883, 0.6288, 0.5100],
  25830. [0.6289, 0.4024, 0.9088, 0.4567, 0.3937, 0.5633, 0.7058, 0.5609]],
  25831. device='cuda:0', dtype=torch.float64)
  25832. predictions are: tensor([[0.5915, 0.3648, 0.8859, 0.3691, 0.3865, 0.2511, 0.6040, 0.4740],
  25833. [0.5877, 0.3866, 0.7135, 0.2236, 0.3872, 0.2383, 0.5726, 0.5677],
  25834. [0.5589, 0.3525, 0.7566, 0.2944, 0.3430, 0.3362, 0.5234, 0.5213],
  25835. [0.5980, 0.3662, 0.8820, 0.4993, 0.4230, 0.5237, 0.6867, 0.5484],
  25836. [0.5208, 0.3225, 0.9172, 0.3313, 0.4996, 0.2441, 0.7192, 0.5617],
  25837. [0.5197, 0.3270, 0.7861, 0.2565, 0.4710, 0.1536, 0.5770, 0.5490],
  25838. [0.5806, 0.3592, 0.8683, 0.5452, 0.3901, 0.4801, 0.6250, 0.5073],
  25839. [0.5634, 0.3657, 0.9120, 0.4750, 0.3990, 0.5621, 0.7043, 0.5740]],
  25840. device='cuda:0', grad_fn=<AddmmBackward>)
  25841. landmarks are: tensor([[[0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  25842. [0.6187, 0.4103, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  25843. [0.6097, 0.4000, 0.7325, 0.2667, 0.3450, 0.3517, 0.5284, 0.5045],
  25844. [0.6279, 0.4008, 0.8600, 0.4883, 0.4325, 0.5283, 0.7010, 0.5378],
  25845. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  25846. [0.6245, 0.4100, 0.7763, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  25847. [0.6275, 0.4024, 0.8500, 0.5383, 0.3913, 0.4883, 0.6288, 0.5100],
  25848. [0.6289, 0.4024, 0.9087, 0.4567, 0.3938, 0.5633, 0.7058, 0.5609]]],
  25849. device='cuda:0')
  25850. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  25851. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  25852. loss_train: 0.020707069255877286
  25853. step: 33
  25854. running loss: 0.0006274869471477965
  25855. Train Steps: 33/90 Loss: 0.0006 torch.Size([8, 600, 800])
  25856. torch.Size([8, 8])
  25857. tensor([[0.6127, 0.4115, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  25858. [0.6171, 0.4127, 0.8900, 0.4800, 0.4325, 0.5783, 0.5769, 0.5090],
  25859. [0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  25860. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  25861. [0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  25862. [0.6226, 0.4103, 0.8575, 0.3450, 0.4388, 0.2067, 0.5787, 0.5383],
  25863. [0.6162, 0.4134, 0.6700, 0.2467, 0.3962, 0.2533, 0.5737, 0.5467],
  25864. [0.6064, 0.3953, 0.8738, 0.4417, 0.3663, 0.4683, 0.5511, 0.5416]],
  25865. device='cuda:0', dtype=torch.float64)
  25866. predictions are: tensor([[0.6007, 0.3772, 0.7200, 0.2945, 0.3763, 0.2790, 0.5664, 0.5535],
  25867. [0.5869, 0.3720, 0.9010, 0.4810, 0.4447, 0.5518, 0.5943, 0.4871],
  25868. [0.6122, 0.3806, 0.8564, 0.4092, 0.3590, 0.4167, 0.6089, 0.5095],
  25869. [0.5744, 0.3446, 0.8711, 0.5548, 0.4080, 0.4152, 0.6318, 0.5181],
  25870. [0.5921, 0.3604, 0.7916, 0.1798, 0.4442, 0.2247, 0.6660, 0.4946],
  25871. [0.5655, 0.3537, 0.8685, 0.3517, 0.4483, 0.2133, 0.5934, 0.5236],
  25872. [0.5666, 0.3661, 0.6690, 0.2627, 0.4073, 0.2356, 0.6014, 0.5319],
  25873. [0.5724, 0.3551, 0.8916, 0.4487, 0.3799, 0.4450, 0.5941, 0.5318]],
  25874. device='cuda:0', grad_fn=<AddmmBackward>)
  25875. landmarks are: tensor([[[0.6127, 0.4114, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  25876. [0.6171, 0.4127, 0.8900, 0.4800, 0.4325, 0.5783, 0.5769, 0.5090],
  25877. [0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  25878. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  25879. [0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  25880. [0.6226, 0.4103, 0.8575, 0.3450, 0.4387, 0.2067, 0.5788, 0.5383],
  25881. [0.6162, 0.4134, 0.6700, 0.2467, 0.3963, 0.2533, 0.5738, 0.5467],
  25882. [0.6064, 0.3952, 0.8737, 0.4417, 0.3663, 0.4683, 0.5511, 0.5416]]],
  25883. device='cuda:0')
  25884. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  25885. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  25886. loss_train: 0.021375142212491482
  25887. step: 34
  25888. running loss: 0.000628680653308573
  25889. Train Steps: 34/90 Loss: 0.0006 torch.Size([8, 600, 800])
  25890. torch.Size([8, 8])
  25891. tensor([[0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6038, 0.6167],
  25892. [ nan, nan, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  25893. [0.6111, 0.4033, 0.8300, 0.3267, 0.3588, 0.3333, 0.5444, 0.5637],
  25894. [0.6296, 0.4045, 0.9138, 0.4100, 0.4232, 0.4242, 0.7422, 0.5297],
  25895. [0.6179, 0.4118, 0.7278, 0.4237, 0.3588, 0.3400, 0.5675, 0.5917],
  25896. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791],
  25897. [0.6152, 0.4131, 0.6863, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  25898. [ nan, nan, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621]],
  25899. device='cuda:0', dtype=torch.float64)
  25900. predictions are: tensor([[0.6458, 0.4290, 0.7912, 0.3173, 0.3675, 0.3524, 0.5973, 0.5699],
  25901. [0.0816, 0.0432, 0.8271, 0.2405, 0.5499, 0.2154, 0.6919, 0.5486],
  25902. [0.6858, 0.4386, 0.8185, 0.3300, 0.3570, 0.2973, 0.5412, 0.5117],
  25903. [0.7040, 0.4585, 0.8861, 0.4092, 0.4093, 0.4017, 0.7211, 0.5145],
  25904. [0.6847, 0.4358, 0.7422, 0.3976, 0.3619, 0.3103, 0.5389, 0.5399],
  25905. [0.7012, 0.4481, 0.8807, 0.4806, 0.4154, 0.4745, 0.6137, 0.4292],
  25906. [0.6557, 0.4236, 0.6619, 0.2601, 0.3675, 0.3147, 0.5812, 0.5161],
  25907. [0.0547, 0.0357, 0.8371, 0.2571, 0.5378, 0.1797, 0.6543, 0.5340]],
  25908. device='cuda:0', grad_fn=<AddmmBackward>)
  25909. landmarks are: tensor([[[0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6037, 0.6167],
  25910. [0.0000, 0.0000, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  25911. [0.6111, 0.4033, 0.8300, 0.3267, 0.3587, 0.3333, 0.5444, 0.5637],
  25912. [0.6296, 0.4045, 0.9137, 0.4100, 0.4232, 0.4242, 0.7422, 0.5297],
  25913. [0.6179, 0.4118, 0.7278, 0.4237, 0.3587, 0.3400, 0.5675, 0.5917],
  25914. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791],
  25915. [0.6152, 0.4131, 0.6862, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  25916. [0.0000, 0.0000, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621]]],
  25917. device='cuda:0')
  25918. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  25919. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  25920. loss_train: 0.022489308787044138
  25921. step: 35
  25922. running loss: 0.0006425516796298325
  25923. Train Steps: 35/90 Loss: 0.0006 torch.Size([8, 600, 800])
  25924. torch.Size([8, 8])
  25925. tensor([[ nan, nan, 0.7425, 0.2117, 0.3937, 0.2433, 0.5438, 0.5567],
  25926. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  25927. [0.6299, 0.4303, 0.7963, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  25928. [0.6264, 0.4071, 0.9038, 0.3867, 0.3663, 0.3917, 0.6338, 0.5283],
  25929. [ nan, nan, 0.7512, 0.2117, 0.4288, 0.2000, 0.5600, 0.5617],
  25930. [0.6133, 0.4066, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358],
  25931. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  25932. [0.6210, 0.4164, 0.7202, 0.2930, 0.4025, 0.2483, 0.5687, 0.5567]],
  25933. device='cuda:0', dtype=torch.float64)
  25934. predictions are: tensor([[0.0647, 0.0577, 0.7094, 0.2131, 0.4062, 0.2470, 0.5556, 0.5255],
  25935. [0.7170, 0.4711, 0.8321, 0.5520, 0.4031, 0.4867, 0.6873, 0.5007],
  25936. [0.6999, 0.4748, 0.7984, 0.3779, 0.5006, 0.2218, 0.5600, 0.6054],
  25937. [0.6240, 0.4170, 0.9029, 0.3641, 0.3922, 0.3674, 0.6687, 0.5235],
  25938. [0.0865, 0.0725, 0.7177, 0.2025, 0.4379, 0.2046, 0.5630, 0.5395],
  25939. [0.6507, 0.4434, 0.6844, 0.2552, 0.4029, 0.2298, 0.5541, 0.5148],
  25940. [0.6804, 0.4533, 0.8606, 0.5392, 0.3824, 0.3423, 0.5690, 0.5038],
  25941. [0.5626, 0.3770, 0.7266, 0.2565, 0.4269, 0.2528, 0.5726, 0.5420]],
  25942. device='cuda:0', grad_fn=<AddmmBackward>)
  25943. landmarks are: tensor([[[0.0000, 0.0000, 0.7425, 0.2117, 0.3938, 0.2433, 0.5437, 0.5567],
  25944. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  25945. [0.6299, 0.4303, 0.7962, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  25946. [0.6264, 0.4071, 0.9038, 0.3867, 0.3663, 0.3917, 0.6338, 0.5283],
  25947. [0.0000, 0.0000, 0.7513, 0.2117, 0.4288, 0.2000, 0.5600, 0.5617],
  25948. [0.6133, 0.4065, 0.6787, 0.2617, 0.3800, 0.2433, 0.5147, 0.5358],
  25949. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  25950. [0.6210, 0.4164, 0.7202, 0.2930, 0.4025, 0.2483, 0.5688, 0.5567]]],
  25951. device='cuda:0')
  25952. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  25953. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  25954. loss_train: 0.023610311618540436
  25955. step: 36
  25956. running loss: 0.000655841989403901
  25957.  
  25958. Train Steps: 36/90 Loss: 0.0007 torch.Size([8, 600, 800])
  25959. torch.Size([8, 8])
  25960. tensor([[0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510],
  25961. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284],
  25962. [0.6201, 0.4082, 0.8827, 0.3715, 0.3825, 0.2712, 0.5845, 0.5412],
  25963. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  25964. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  25965. [0.6275, 0.4024, 0.8600, 0.2283, 0.5350, 0.1800, 0.7074, 0.5413],
  25966. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  25967. [0.6135, 0.3994, 0.7913, 0.3050, 0.3625, 0.3050, 0.5837, 0.5050]],
  25968. device='cuda:0', dtype=torch.float64)
  25969. predictions are: tensor([[0.5305, 0.3545, 0.6789, 0.2938, 0.3590, 0.3198, 0.5248, 0.5454],
  25970. [0.6119, 0.3984, 0.7472, 0.2210, 0.4547, 0.1996, 0.5760, 0.5310],
  25971. [0.6218, 0.4105, 0.8767, 0.3654, 0.3838, 0.2859, 0.5847, 0.5196],
  25972. [0.5847, 0.3989, 0.6975, 0.2397, 0.3882, 0.3049, 0.5973, 0.5677],
  25973. [0.5755, 0.3947, 0.8557, 0.5025, 0.4140, 0.5360, 0.5966, 0.5015],
  25974. [0.5259, 0.3512, 0.8314, 0.2048, 0.5442, 0.2115, 0.6936, 0.5579],
  25975. [0.6149, 0.4094, 0.8563, 0.4835, 0.4030, 0.5499, 0.6354, 0.5013],
  25976. [0.6113, 0.4014, 0.7697, 0.3001, 0.3698, 0.3214, 0.5749, 0.5048]],
  25977. device='cuda:0', grad_fn=<AddmmBackward>)
  25978. landmarks are: tensor([[[0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510],
  25979. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284],
  25980. [0.6201, 0.4082, 0.8827, 0.3715, 0.3825, 0.2712, 0.5845, 0.5412],
  25981. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  25982. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  25983. [0.6275, 0.4024, 0.8600, 0.2283, 0.5350, 0.1800, 0.7074, 0.5413],
  25984. [0.6258, 0.4038, 0.8750, 0.4883, 0.3900, 0.5500, 0.6375, 0.5217],
  25985. [0.6135, 0.3994, 0.7912, 0.3050, 0.3625, 0.3050, 0.5838, 0.5050]]],
  25986. device='cuda:0')
  25987. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  25988. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  25989. loss_train: 0.024129573779646307
  25990. step: 37
  25991. running loss: 0.0006521506426931435
  25992. Train Steps: 37/90 Loss: 0.0007 torch.Size([8, 600, 800])
  25993. torch.Size([8, 8])
  25994. tensor([[0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  25995. [0.6286, 0.3977, 0.9038, 0.4733, 0.3900, 0.4150, 0.7074, 0.5320],
  25996. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  25997. [0.6275, 0.4003, 0.9100, 0.3783, 0.4388, 0.3133, 0.7058, 0.5343],
  25998. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  25999. [0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  26000. [0.6109, 0.4036, 0.7188, 0.1750, 0.3850, 0.2550, 0.5863, 0.5567],
  26001. [ nan, nan, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702]],
  26002. device='cuda:0', dtype=torch.float64)
  26003. predictions are: tensor([[0.5915, 0.4067, 0.7443, 0.3600, 0.3274, 0.4274, 0.5061, 0.5529],
  26004. [0.6377, 0.4158, 0.8687, 0.4530, 0.3734, 0.4218, 0.6892, 0.5308],
  26005. [0.5794, 0.3966, 0.8383, 0.4475, 0.4386, 0.5175, 0.5149, 0.4928],
  26006. [0.6338, 0.4248, 0.8548, 0.3710, 0.4222, 0.3116, 0.6730, 0.5358],
  26007. [0.7068, 0.4678, 0.7913, 0.5072, 0.4048, 0.5658, 0.7041, 0.5697],
  26008. [0.6408, 0.4241, 0.8375, 0.3433, 0.3799, 0.2638, 0.5986, 0.4770],
  26009. [0.6320, 0.4270, 0.6807, 0.1809, 0.3609, 0.2769, 0.5538, 0.5540],
  26010. [0.0137, 0.0190, 0.8765, 0.3296, 0.4928, 0.2607, 0.7076, 0.5754]],
  26011. device='cuda:0', grad_fn=<AddmmBackward>)
  26012. landmarks are: tensor([[[0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  26013. [0.6286, 0.3977, 0.9038, 0.4733, 0.3900, 0.4150, 0.7074, 0.5320],
  26014. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  26015. [0.6275, 0.4003, 0.9100, 0.3783, 0.4387, 0.3133, 0.7058, 0.5343],
  26016. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  26017. [0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  26018. [0.6108, 0.4036, 0.7188, 0.1750, 0.3850, 0.2550, 0.5863, 0.5567],
  26019. [0.0000, 0.0000, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702]]],
  26020. device='cuda:0')
  26021. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  26022. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  26023. loss_train: 0.02468388859415427
  26024. step: 38
  26025. running loss: 0.0006495760156356386
  26026. Train Steps: 38/90 Loss: 0.0006 torch.Size([8, 600, 800])
  26027. torch.Size([8, 8])
  26028. tensor([[0.6090, 0.4010, 0.7838, 0.3483, 0.3538, 0.3783, 0.5462, 0.5077],
  26029. [0.6133, 0.4094, 0.8495, 0.4028, 0.3588, 0.3200, 0.5003, 0.5407],
  26030. [0.6262, 0.4052, 0.8888, 0.4700, 0.3675, 0.5117, 0.6350, 0.5233],
  26031. [ nan, nan, 0.7981, 0.3194, 0.3625, 0.3167, 0.5040, 0.5563],
  26032. [0.6274, 0.4003, 0.8638, 0.5967, 0.3688, 0.4900, 0.6108, 0.4661],
  26033. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  26034. [0.6128, 0.4084, 0.8738, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  26035. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927]],
  26036. device='cuda:0', dtype=torch.float64)
  26037. predictions are: tensor([[0.6129, 0.4063, 0.7917, 0.3597, 0.3569, 0.3850, 0.5596, 0.5365],
  26038. [0.6281, 0.4296, 0.8484, 0.4040, 0.3571, 0.3257, 0.5343, 0.5817],
  26039. [0.6373, 0.4325, 0.8869, 0.4529, 0.3767, 0.5279, 0.6611, 0.5714],
  26040. [0.1334, 0.0916, 0.7740, 0.3285, 0.3362, 0.3465, 0.5411, 0.5706],
  26041. [0.6513, 0.4156, 0.8368, 0.5899, 0.3804, 0.4836, 0.6301, 0.5235],
  26042. [0.6552, 0.4389, 0.8928, 0.4076, 0.3683, 0.3872, 0.6558, 0.5557],
  26043. [0.6481, 0.4358, 0.8740, 0.4644, 0.3661, 0.3698, 0.5430, 0.5807],
  26044. [0.6519, 0.4143, 0.8887, 0.4221, 0.3605, 0.4488, 0.6484, 0.5383]],
  26045. device='cuda:0', grad_fn=<AddmmBackward>)
  26046. landmarks are: tensor([[[0.6090, 0.4010, 0.7837, 0.3483, 0.3537, 0.3783, 0.5462, 0.5077],
  26047. [0.6133, 0.4094, 0.8495, 0.4028, 0.3587, 0.3200, 0.5003, 0.5407],
  26048. [0.6262, 0.4052, 0.8888, 0.4700, 0.3675, 0.5117, 0.6350, 0.5233],
  26049. [0.0000, 0.0000, 0.7981, 0.3194, 0.3625, 0.3167, 0.5040, 0.5563],
  26050. [0.6274, 0.4003, 0.8637, 0.5967, 0.3688, 0.4900, 0.6108, 0.4661],
  26051. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  26052. [0.6127, 0.4084, 0.8737, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  26053. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927]]],
  26054. device='cuda:0')
  26055. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  26056. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  26057. loss_train: 0.025601829052902758
  26058. step: 39
  26059. running loss: 0.0006564571552026349
  26060. Train Steps: 39/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26061. torch.Size([8, 8])
  26062. tensor([[0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  26063. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  26064. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  26065. [0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726],
  26066. [0.6124, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  26067. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  26068. [0.6216, 0.4167, 0.8588, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667],
  26069. [0.6185, 0.4080, 0.8625, 0.3483, 0.3788, 0.2650, 0.5320, 0.5272]],
  26070. device='cuda:0', dtype=torch.float64)
  26071. predictions are: tensor([[0.5923, 0.3935, 0.8309, 0.4015, 0.3470, 0.4915, 0.5611, 0.5772],
  26072. [0.5971, 0.3880, 0.9268, 0.4489, 0.4108, 0.3010, 0.6907, 0.5608],
  26073. [0.5750, 0.3796, 0.8649, 0.3878, 0.3541, 0.3999, 0.5692, 0.5792],
  26074. [0.6181, 0.4089, 0.7601, 0.2960, 0.3571, 0.2836, 0.5407, 0.4981],
  26075. [0.5764, 0.3802, 0.8338, 0.5028, 0.3890, 0.4537, 0.5403, 0.5809],
  26076. [0.5548, 0.3543, 0.8884, 0.4880, 0.4015, 0.5276, 0.6515, 0.5027],
  26077. [0.5754, 0.3802, 0.8593, 0.5455, 0.3952, 0.5068, 0.6076, 0.5875],
  26078. [0.6041, 0.3931, 0.8534, 0.3662, 0.3729, 0.2801, 0.5366, 0.5386]],
  26079. device='cuda:0', grad_fn=<AddmmBackward>)
  26080. landmarks are: tensor([[[0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  26081. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  26082. [0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  26083. [0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726],
  26084. [0.6123, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  26085. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  26086. [0.6216, 0.4167, 0.8587, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667],
  26087. [0.6186, 0.4080, 0.8625, 0.3483, 0.3787, 0.2650, 0.5320, 0.5272]]],
  26088. device='cuda:0')
  26089. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  26090. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  26091. loss_train: 0.026024623948615044
  26092. step: 40
  26093. running loss: 0.0006506155987153761
  26094.  
  26095. Train Steps: 40/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26096. torch.Size([8, 8])
  26097. tensor([[0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5637, 0.5633],
  26098. [0.6339, 0.4123, 0.8638, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  26099. [0.6265, 0.4251, 0.7113, 0.3550, 0.4375, 0.2117, 0.5587, 0.6118],
  26100. [0.6275, 0.4048, 0.8488, 0.2883, 0.4463, 0.2033, 0.6321, 0.5155],
  26101. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  26102. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  26103. [0.6262, 0.4163, 0.8850, 0.5183, 0.3763, 0.4150, 0.6025, 0.5500],
  26104. [0.6200, 0.4071, 0.7338, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517]],
  26105. device='cuda:0', dtype=torch.float64)
  26106. predictions are: tensor([[0.5710, 0.3669, 0.8803, 0.4861, 0.3399, 0.3925, 0.5360, 0.5629],
  26107. [0.6074, 0.3811, 0.8732, 0.5031, 0.3837, 0.5618, 0.7136, 0.5447],
  26108. [0.6312, 0.4340, 0.7334, 0.3560, 0.4082, 0.2233, 0.5435, 0.6089],
  26109. [0.5749, 0.3547, 0.8671, 0.2773, 0.4278, 0.2215, 0.6111, 0.4986],
  26110. [0.5160, 0.3339, 0.8919, 0.4469, 0.3652, 0.4611, 0.5324, 0.5568],
  26111. [0.5851, 0.3807, 0.8481, 0.4560, 0.4351, 0.2759, 0.5394, 0.6058],
  26112. [0.5667, 0.3640, 0.8894, 0.5041, 0.3626, 0.4124, 0.5844, 0.5541],
  26113. [0.5533, 0.3655, 0.7546, 0.2004, 0.3957, 0.2607, 0.6060, 0.5405]],
  26114. device='cuda:0', grad_fn=<AddmmBackward>)
  26115. landmarks are: tensor([[[0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5638, 0.5633],
  26116. [0.6339, 0.4123, 0.8637, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  26117. [0.6265, 0.4251, 0.7113, 0.3550, 0.4375, 0.2117, 0.5587, 0.6118],
  26118. [0.6275, 0.4048, 0.8487, 0.2883, 0.4462, 0.2033, 0.6321, 0.5155],
  26119. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  26120. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  26121. [0.6262, 0.4163, 0.8850, 0.5183, 0.3762, 0.4150, 0.6025, 0.5500],
  26122. [0.6200, 0.4071, 0.7337, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517]]],
  26123. device='cuda:0')
  26124. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  26125. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  26126. loss_train: 0.02692328830016777
  26127. step: 41
  26128. running loss: 0.0006566655682967748
  26129. Train Steps: 41/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26130. torch.Size([8, 8])
  26131. tensor([[0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6362, 0.5200],
  26132. [0.6164, 0.4102, 0.8850, 0.4867, 0.4213, 0.5983, 0.5609, 0.5038],
  26133. [0.6175, 0.4093, 0.8800, 0.4433, 0.4075, 0.4367, 0.5128, 0.4901],
  26134. [ nan, nan, 0.7225, 0.2167, 0.3987, 0.2283, 0.5427, 0.5181],
  26135. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  26136. [0.6205, 0.4081, 0.8950, 0.4017, 0.3788, 0.4700, 0.5963, 0.5667],
  26137. [0.6339, 0.4149, 0.8800, 0.5000, 0.3900, 0.5283, 0.7541, 0.5424],
  26138. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220]],
  26139. device='cuda:0', dtype=torch.float64)
  26140. predictions are: tensor([[0.6538, 0.4220, 0.8873, 0.3851, 0.3581, 0.5102, 0.6221, 0.5243],
  26141. [0.6167, 0.4197, 0.9097, 0.5269, 0.4462, 0.5868, 0.5661, 0.5541],
  26142. [0.6237, 0.4020, 0.9165, 0.4854, 0.4226, 0.4382, 0.4926, 0.5166],
  26143. [0.0564, 0.0392, 0.7513, 0.2278, 0.3858, 0.2244, 0.5631, 0.5290],
  26144. [0.6554, 0.4202, 0.9537, 0.4827, 0.4247, 0.2915, 0.6790, 0.5506],
  26145. [0.6010, 0.3892, 0.9020, 0.4234, 0.3771, 0.4627, 0.5740, 0.5795],
  26146. [0.6802, 0.4307, 0.9158, 0.5482, 0.3977, 0.5348, 0.6974, 0.5587],
  26147. [0.6571, 0.4200, 0.7340, 0.2392, 0.4364, 0.1543, 0.5420, 0.5228]],
  26148. device='cuda:0', grad_fn=<AddmmBackward>)
  26149. landmarks are: tensor([[[0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6363, 0.5200],
  26150. [0.6164, 0.4102, 0.8850, 0.4867, 0.4212, 0.5983, 0.5609, 0.5038],
  26151. [0.6175, 0.4093, 0.8800, 0.4433, 0.4075, 0.4367, 0.5128, 0.4901],
  26152. [0.0000, 0.0000, 0.7225, 0.2167, 0.3988, 0.2283, 0.5427, 0.5181],
  26153. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  26154. [0.6205, 0.4081, 0.8950, 0.4017, 0.3787, 0.4700, 0.5962, 0.5667],
  26155. [0.6339, 0.4149, 0.8800, 0.5000, 0.3900, 0.5283, 0.7541, 0.5424],
  26156. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220]]],
  26157. device='cuda:0')
  26158. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  26159. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  26160. loss_train: 0.027541582996491343
  26161. step: 42
  26162. running loss: 0.0006557519761069367
  26163. Train Steps: 42/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26164. torch.Size([8, 8])
  26165. tensor([[0.6222, 0.3957, 0.8838, 0.5017, 0.3937, 0.4600, 0.5900, 0.5017],
  26166. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901],
  26167. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  26168. [0.6140, 0.4070, 0.8700, 0.5000, 0.4612, 0.4900, 0.5260, 0.5852],
  26169. [0.6076, 0.3953, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954],
  26170. [0.6265, 0.4071, 0.8875, 0.3367, 0.3975, 0.3350, 0.6312, 0.5250],
  26171. [0.6229, 0.4198, 0.7662, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783],
  26172. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395]],
  26173. device='cuda:0', dtype=torch.float64)
  26174. predictions are: tensor([[0.6118, 0.3809, 0.9111, 0.5268, 0.4022, 0.4456, 0.5834, 0.5014],
  26175. [0.5783, 0.3738, 0.8171, 0.3106, 0.3702, 0.2482, 0.5296, 0.4924],
  26176. [0.5396, 0.3483, 0.8902, 0.5831, 0.3964, 0.4131, 0.5367, 0.4819],
  26177. [0.5588, 0.3638, 0.9016, 0.5124, 0.4678, 0.4860, 0.5316, 0.5854],
  26178. [0.5850, 0.3821, 0.8373, 0.3984, 0.3518, 0.3892, 0.5544, 0.4998],
  26179. [0.5832, 0.3674, 0.9191, 0.3663, 0.4033, 0.3351, 0.6651, 0.5223],
  26180. [0.5833, 0.3871, 0.7874, 0.2756, 0.4619, 0.2136, 0.5767, 0.5614],
  26181. [0.5861, 0.3766, 0.7723, 0.2376, 0.4008, 0.2322, 0.6142, 0.5345]],
  26182. device='cuda:0', grad_fn=<AddmmBackward>)
  26183. landmarks are: tensor([[[0.6222, 0.3957, 0.8838, 0.5017, 0.3938, 0.4600, 0.5900, 0.5017],
  26184. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901],
  26185. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  26186. [0.6140, 0.4070, 0.8700, 0.5000, 0.4613, 0.4900, 0.5260, 0.5852],
  26187. [0.6076, 0.3952, 0.8125, 0.3917, 0.3525, 0.4083, 0.5346, 0.4954],
  26188. [0.6265, 0.4071, 0.8875, 0.3367, 0.3975, 0.3350, 0.6313, 0.5250],
  26189. [0.6229, 0.4198, 0.7663, 0.2700, 0.4700, 0.2133, 0.5675, 0.5783],
  26190. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395]]],
  26191. device='cuda:0')
  26192. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  26193. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  26194. loss_train: 0.028159943351056427
  26195. step: 43
  26196. running loss: 0.0006548824035129402
  26197. Train Steps: 43/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26198. torch.Size([8, 8])
  26199. tensor([[0.6260, 0.4214, 0.8538, 0.5500, 0.3663, 0.3767, 0.5587, 0.5983],
  26200. [0.6193, 0.4108, 0.7438, 0.2700, 0.3650, 0.3683, 0.6238, 0.5717],
  26201. [0.6092, 0.4001, 0.8638, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  26202. [ nan, nan, 0.8300, 0.3150, 0.3588, 0.3383, 0.5208, 0.5194],
  26203. [0.6250, 0.4110, 0.7238, 0.2067, 0.4263, 0.1883, 0.5625, 0.5633],
  26204. [0.6261, 0.3987, 0.9045, 0.4208, 0.3600, 0.4633, 0.6570, 0.5162],
  26205. [0.6275, 0.4003, 0.9100, 0.3783, 0.4388, 0.3133, 0.7058, 0.5343],
  26206. [0.6142, 0.3982, 0.8650, 0.4883, 0.3912, 0.4317, 0.5315, 0.5350]],
  26207. device='cuda:0', dtype=torch.float64)
  26208. predictions are: tensor([[0.6471, 0.4310, 0.8641, 0.5437, 0.3815, 0.3744, 0.5533, 0.5701],
  26209. [0.6438, 0.4260, 0.7552, 0.2701, 0.3719, 0.3479, 0.6202, 0.5499],
  26210. [0.6414, 0.4279, 0.8818, 0.4889, 0.4370, 0.5421, 0.5285, 0.4820],
  26211. [0.0014, 0.0012, 0.8453, 0.3188, 0.3674, 0.3091, 0.5420, 0.5049],
  26212. [0.6419, 0.4286, 0.7340, 0.2096, 0.4328, 0.1638, 0.5606, 0.5363],
  26213. [0.6953, 0.4560, 0.9059, 0.4319, 0.3676, 0.4678, 0.6474, 0.4943],
  26214. [0.6849, 0.4451, 0.9102, 0.3899, 0.4505, 0.2879, 0.6936, 0.5105],
  26215. [0.6586, 0.4198, 0.8778, 0.5090, 0.4006, 0.4320, 0.5175, 0.4929]],
  26216. device='cuda:0', grad_fn=<AddmmBackward>)
  26217. landmarks are: tensor([[[0.6260, 0.4214, 0.8537, 0.5500, 0.3663, 0.3767, 0.5587, 0.5983],
  26218. [0.6193, 0.4108, 0.7437, 0.2700, 0.3650, 0.3683, 0.6237, 0.5717],
  26219. [0.6092, 0.4001, 0.8637, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  26220. [0.0000, 0.0000, 0.8300, 0.3150, 0.3587, 0.3383, 0.5208, 0.5194],
  26221. [0.6250, 0.4110, 0.7237, 0.2067, 0.4263, 0.1883, 0.5625, 0.5633],
  26222. [0.6261, 0.3987, 0.9045, 0.4208, 0.3600, 0.4633, 0.6570, 0.5162],
  26223. [0.6275, 0.4003, 0.9100, 0.3783, 0.4387, 0.3133, 0.7058, 0.5343],
  26224. [0.6143, 0.3982, 0.8650, 0.4883, 0.3913, 0.4317, 0.5315, 0.5350]]],
  26225. device='cuda:0')
  26226. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  26227. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  26228. loss_train: 0.028655284026172012
  26229. step: 44
  26230. running loss: 0.000651256455140273
  26231.  
  26232. Train Steps: 44/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26233. torch.Size([8, 8])
  26234. tensor([[0.6183, 0.4076, 0.8838, 0.4517, 0.3813, 0.4483, 0.5775, 0.5633],
  26235. [0.6137, 0.4038, 0.8563, 0.4050, 0.3813, 0.2550, 0.5106, 0.4954],
  26236. [0.6228, 0.4004, 0.8750, 0.5250, 0.3825, 0.5233, 0.6362, 0.5000],
  26237. [0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  26238. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  26239. [0.6124, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  26240. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  26241. [0.6221, 0.4107, 0.7788, 0.3033, 0.3950, 0.2817, 0.6075, 0.5517]],
  26242. device='cuda:0', dtype=torch.float64)
  26243. predictions are: tensor([[0.5749, 0.3847, 0.8832, 0.4350, 0.3910, 0.4564, 0.5788, 0.5423],
  26244. [0.5752, 0.3708, 0.8686, 0.4065, 0.4086, 0.2554, 0.5023, 0.4715],
  26245. [0.5742, 0.3650, 0.8813, 0.5068, 0.3964, 0.5120, 0.6576, 0.4739],
  26246. [0.6020, 0.3978, 0.8577, 0.5034, 0.5040, 0.5041, 0.5080, 0.5240],
  26247. [0.6640, 0.4409, 0.7202, 0.2139, 0.4073, 0.2258, 0.5346, 0.5368],
  26248. [0.6279, 0.4088, 0.8483, 0.4965, 0.4003, 0.4486, 0.5340, 0.5275],
  26249. [0.6190, 0.3872, 0.8941, 0.4202, 0.3754, 0.3232, 0.5813, 0.4973],
  26250. [0.6085, 0.4068, 0.8025, 0.2739, 0.4027, 0.2712, 0.6047, 0.5237]],
  26251. device='cuda:0', grad_fn=<AddmmBackward>)
  26252. landmarks are: tensor([[[0.6183, 0.4076, 0.8838, 0.4517, 0.3812, 0.4483, 0.5775, 0.5633],
  26253. [0.6137, 0.4038, 0.8562, 0.4050, 0.3812, 0.2550, 0.5106, 0.4954],
  26254. [0.6228, 0.4004, 0.8750, 0.5250, 0.3825, 0.5233, 0.6363, 0.5000],
  26255. [0.6095, 0.4002, 0.8533, 0.5168, 0.5031, 0.5094, 0.5125, 0.5433],
  26256. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  26257. [0.6123, 0.4069, 0.8314, 0.5001, 0.3738, 0.4650, 0.5167, 0.5402],
  26258. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  26259. [0.6221, 0.4107, 0.7788, 0.3033, 0.3950, 0.2817, 0.6075, 0.5517]]],
  26260. device='cuda:0')
  26261. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  26262. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  26263. loss_train: 0.02907089513610117
  26264. step: 45
  26265. running loss: 0.0006460198919133593
  26266. Train Steps: 45/90 Loss: 0.0006 torch.Size([8, 600, 800])
  26267. torch.Size([8, 8])
  26268. tensor([[0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  26269. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  26270. [0.6075, 0.4000, 0.8513, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  26271. [0.6082, 0.4024, 0.8738, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  26272. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795],
  26273. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  26274. [0.6200, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  26275. [0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153]],
  26276. device='cuda:0', dtype=torch.float64)
  26277. predictions are: tensor([[0.6424, 0.4252, 0.8419, 0.5430, 0.4097, 0.4271, 0.5317, 0.5429],
  26278. [0.6416, 0.4061, 0.8005, 0.2153, 0.4281, 0.2328, 0.6118, 0.5114],
  26279. [0.5964, 0.3905, 0.8354, 0.4923, 0.4562, 0.5312, 0.5099, 0.5059],
  26280. [0.6142, 0.4033, 0.8649, 0.3768, 0.3471, 0.3857, 0.5197, 0.4894],
  26281. [0.6468, 0.4079, 0.8238, 0.5167, 0.3851, 0.5406, 0.6945, 0.5502],
  26282. [0.5346, 0.3638, 0.7716, 0.1978, 0.4523, 0.1814, 0.6084, 0.5259],
  26283. [0.5847, 0.3879, 0.8422, 0.2714, 0.4162, 0.2223, 0.5703, 0.5198],
  26284. [0.5986, 0.3831, 0.8542, 0.4812, 0.4572, 0.5328, 0.5836, 0.4991]],
  26285. device='cuda:0', grad_fn=<AddmmBackward>)
  26286. landmarks are: tensor([[[0.6198, 0.4075, 0.8650, 0.5617, 0.4150, 0.4367, 0.5450, 0.5650],
  26287. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  26288. [0.6075, 0.4000, 0.8512, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  26289. [0.6082, 0.4024, 0.8737, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  26290. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795],
  26291. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  26292. [0.6199, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  26293. [0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153]]],
  26294. device='cuda:0')
  26295. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  26296. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  26297. loss_train: 0.02950749092269689
  26298. step: 46
  26299. running loss: 0.0006414671939716715
  26300. Train Steps: 46/90 Loss: 0.0006 torch.Size([8, 600, 800])
  26301. torch.Size([8, 8])
  26302. tensor([[0.6075, 0.4000, 0.8513, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  26303. [0.6127, 0.4115, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  26304. [0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364],
  26305. [ nan, nan, 0.8938, 0.2850, 0.4662, 0.3117, 0.7406, 0.5528],
  26306. [0.6364, 0.4165, 0.9088, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  26307. [0.6178, 0.4059, 0.7525, 0.2250, 0.4313, 0.1783, 0.5404, 0.5240],
  26308. [0.6072, 0.4029, 0.7037, 0.2150, 0.3912, 0.2267, 0.5516, 0.5507],
  26309. [0.6346, 0.4144, 0.9088, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899]],
  26310. device='cuda:0', dtype=torch.float64)
  26311. predictions are: tensor([[0.6545, 0.4337, 0.8315, 0.5277, 0.4499, 0.5490, 0.5029, 0.5197],
  26312. [0.6328, 0.4295, 0.6849, 0.2833, 0.3487, 0.3037, 0.4949, 0.5662],
  26313. [0.6140, 0.4039, 0.7645, 0.2126, 0.4507, 0.2166, 0.6423, 0.5386],
  26314. [0.1316, 0.0792, 0.9086, 0.2893, 0.4728, 0.2985, 0.7566, 0.5720],
  26315. [0.6985, 0.4589, 0.8860, 0.4478, 0.4048, 0.3236, 0.6257, 0.5292],
  26316. [0.5595, 0.3648, 0.7252, 0.2200, 0.4176, 0.1788, 0.5110, 0.5336],
  26317. [0.6185, 0.4166, 0.6774, 0.2080, 0.3846, 0.2352, 0.5337, 0.5579],
  26318. [0.6581, 0.4324, 0.8878, 0.4624, 0.3838, 0.4349, 0.6888, 0.5846]],
  26319. device='cuda:0', grad_fn=<AddmmBackward>)
  26320. landmarks are: tensor([[[0.6075, 0.4000, 0.8512, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  26321. [0.6127, 0.4114, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  26322. [0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364],
  26323. [0.0000, 0.0000, 0.8938, 0.2850, 0.4663, 0.3117, 0.7406, 0.5528],
  26324. [0.6364, 0.4165, 0.9087, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  26325. [0.6178, 0.4059, 0.7525, 0.2250, 0.4313, 0.1783, 0.5404, 0.5240],
  26326. [0.6072, 0.4029, 0.7038, 0.2150, 0.3913, 0.2267, 0.5516, 0.5507],
  26327. [0.6346, 0.4144, 0.9087, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899]]],
  26328. device='cuda:0')
  26329. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  26330. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  26331. loss_train: 0.030339945456944406
  26332. step: 47
  26333. running loss: 0.0006455307544030725
  26334. Train Steps: 47/90 Loss: 0.0006 torch.Size([8, 600, 800])
  26335. torch.Size([8, 8])
  26336. tensor([[0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  26337. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  26338. [0.6197, 0.4118, 0.8688, 0.5517, 0.4037, 0.5233, 0.5875, 0.5600],
  26339. [0.6289, 0.4032, 0.8419, 0.5446, 0.4075, 0.5017, 0.6312, 0.5117],
  26340. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  26341. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  26342. [0.6353, 0.4128, 0.8488, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  26343. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6063, 0.5617]],
  26344. device='cuda:0', dtype=torch.float64)
  26345. predictions are: tensor([[0.6954, 0.4573, 0.8934, 0.3729, 0.4420, 0.2353, 0.6405, 0.5110],
  26346. [0.6154, 0.4141, 0.7470, 0.3557, 0.3407, 0.4372, 0.5124, 0.5594],
  26347. [0.6399, 0.4215, 0.8431, 0.5455, 0.3859, 0.5617, 0.5871, 0.5451],
  26348. [0.6166, 0.3978, 0.8250, 0.5295, 0.3898, 0.5311, 0.6521, 0.5186],
  26349. [0.6482, 0.4284, 0.8307, 0.3686, 0.3432, 0.4072, 0.6018, 0.5702],
  26350. [0.6427, 0.4318, 0.8174, 0.4371, 0.4419, 0.2847, 0.5541, 0.6283],
  26351. [0.4102, 0.2716, 0.8553, 0.2385, 0.5360, 0.1767, 0.7007, 0.5699],
  26352. [0.5646, 0.3775, 0.8764, 0.4672, 0.3465, 0.4999, 0.6117, 0.5606]],
  26353. device='cuda:0', grad_fn=<AddmmBackward>)
  26354. landmarks are: tensor([[[0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  26355. [0.6048, 0.3987, 0.7620, 0.3861, 0.3475, 0.4167, 0.5137, 0.5466],
  26356. [0.6197, 0.4118, 0.8687, 0.5517, 0.4038, 0.5233, 0.5875, 0.5600],
  26357. [0.6289, 0.4031, 0.8419, 0.5446, 0.4075, 0.5017, 0.6313, 0.5117],
  26358. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  26359. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  26360. [0.6353, 0.4128, 0.8487, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  26361. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6062, 0.5617]]],
  26362. device='cuda:0')
  26363. loss_train_step before backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  26364. loss_train_step after backward: tensor(0.0016, device='cuda:0', grad_fn=<MseLossBackward>)
  26365. loss_train: 0.03189117752481252
  26366. step: 48
  26367. running loss: 0.0006643995317669275
  26368.  
  26369. Train Steps: 48/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26370. torch.Size([8, 8])
  26371. tensor([[0.6043, 0.4022, 0.6887, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136],
  26372. [0.6126, 0.3954, 0.8538, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  26373. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167],
  26374. [0.6156, 0.4125, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  26375. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5837, 0.5583],
  26376. [0.6239, 0.4123, 0.8313, 0.2550, 0.4500, 0.2050, 0.6175, 0.5400],
  26377. [0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367],
  26378. [ nan, nan, 0.7425, 0.2117, 0.3937, 0.2433, 0.5438, 0.5567]],
  26379. device='cuda:0', dtype=torch.float64)
  26380. predictions are: tensor([[ 0.6015, 0.4091, 0.6792, 0.2112, 0.3803, 0.2605, 0.5657, 0.5407],
  26381. [ 0.6865, 0.4520, 0.8424, 0.5097, 0.4201, 0.4669, 0.5732, 0.5527],
  26382. [ 0.6460, 0.4397, 0.7416, 0.2950, 0.3965, 0.2750, 0.6165, 0.6558],
  26383. [ 0.6428, 0.4430, 0.8711, 0.5092, 0.4494, 0.5808, 0.6035, 0.5408],
  26384. [ 0.6505, 0.4395, 0.8684, 0.4644, 0.4637, 0.5359, 0.6041, 0.5740],
  26385. [ 0.6279, 0.4122, 0.8126, 0.2524, 0.4543, 0.1920, 0.6404, 0.5525],
  26386. [ 0.6940, 0.4552, 0.8476, 0.3173, 0.3913, 0.2747, 0.6630, 0.5650],
  26387. [-0.0232, -0.0028, 0.7549, 0.2287, 0.4013, 0.2235, 0.5537, 0.5746]],
  26388. device='cuda:0', grad_fn=<AddmmBackward>)
  26389. landmarks are: tensor([[[0.6043, 0.4022, 0.6888, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136],
  26390. [0.6126, 0.3954, 0.8537, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  26391. [0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167],
  26392. [0.6155, 0.4124, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  26393. [0.6080, 0.4010, 0.8750, 0.4500, 0.4825, 0.5617, 0.5838, 0.5583],
  26394. [0.6239, 0.4123, 0.8313, 0.2550, 0.4500, 0.2050, 0.6175, 0.5400],
  26395. [0.6286, 0.4040, 0.8696, 0.3047, 0.3924, 0.2887, 0.6300, 0.5367],
  26396. [0.0000, 0.0000, 0.7425, 0.2117, 0.3938, 0.2433, 0.5437, 0.5567]]],
  26397. device='cuda:0')
  26398. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  26399. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  26400. loss_train: 0.032499423483386636
  26401. step: 49
  26402. running loss: 0.0006632535404772783
  26403. Train Steps: 49/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26404. torch.Size([8, 8])
  26405. tensor([[0.6263, 0.4233, 0.7924, 0.4626, 0.3788, 0.2883, 0.5573, 0.6047],
  26406. [0.6097, 0.3988, 0.8650, 0.5250, 0.4213, 0.5200, 0.5675, 0.5050],
  26407. [0.6122, 0.3993, 0.8738, 0.4667, 0.4517, 0.4879, 0.5155, 0.4927],
  26408. [0.6163, 0.4006, 0.8788, 0.4683, 0.3663, 0.4883, 0.5887, 0.5017],
  26409. [0.6145, 0.4008, 0.8750, 0.5383, 0.3975, 0.4650, 0.5563, 0.5533],
  26410. [0.6185, 0.4067, 0.8838, 0.4450, 0.4037, 0.4733, 0.5213, 0.5142],
  26411. [0.6346, 0.4144, 0.9088, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  26412. [ nan, nan, 0.8850, 0.3000, 0.5363, 0.2250, 0.7343, 0.5771]],
  26413. device='cuda:0', dtype=torch.float64)
  26414. predictions are: tensor([[0.6716, 0.4577, 0.7748, 0.4710, 0.3777, 0.2858, 0.5700, 0.6375],
  26415. [0.6477, 0.4321, 0.8409, 0.5273, 0.4271, 0.5287, 0.5504, 0.5211],
  26416. [0.6724, 0.4409, 0.8604, 0.4746, 0.4422, 0.5025, 0.5418, 0.5029],
  26417. [0.6832, 0.4455, 0.8589, 0.4731, 0.3588, 0.5001, 0.5931, 0.5311],
  26418. [0.6424, 0.4260, 0.8493, 0.5241, 0.3827, 0.4609, 0.5877, 0.5796],
  26419. [0.6399, 0.4265, 0.8529, 0.4411, 0.3920, 0.4770, 0.5210, 0.5491],
  26420. [0.6415, 0.4153, 0.8961, 0.4563, 0.3989, 0.4180, 0.7343, 0.6127],
  26421. [0.0013, 0.0094, 0.9064, 0.2967, 0.5239, 0.2192, 0.7298, 0.5917]],
  26422. device='cuda:0', grad_fn=<AddmmBackward>)
  26423. landmarks are: tensor([[[0.6263, 0.4232, 0.7924, 0.4626, 0.3787, 0.2883, 0.5573, 0.6047],
  26424. [0.6097, 0.3988, 0.8650, 0.5250, 0.4212, 0.5200, 0.5675, 0.5050],
  26425. [0.6122, 0.3993, 0.8737, 0.4667, 0.4517, 0.4879, 0.5155, 0.4927],
  26426. [0.6163, 0.4006, 0.8788, 0.4683, 0.3663, 0.4883, 0.5888, 0.5017],
  26427. [0.6145, 0.4008, 0.8750, 0.5383, 0.3975, 0.4650, 0.5562, 0.5533],
  26428. [0.6185, 0.4067, 0.8838, 0.4450, 0.4038, 0.4733, 0.5213, 0.5142],
  26429. [0.6346, 0.4144, 0.9087, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  26430. [0.0000, 0.0000, 0.8850, 0.3000, 0.5362, 0.2250, 0.7343, 0.5771]]],
  26431. device='cuda:0')
  26432. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  26433. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  26434. loss_train: 0.03302222432103008
  26435. step: 50
  26436. running loss: 0.0006604444864206017
  26437. Train Steps: 50/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26438. torch.Size([8, 8])
  26439. tensor([[0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5988, 0.5700],
  26440. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  26441. [ nan, nan, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  26442. [0.6236, 0.3966, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  26443. [0.6102, 0.3999, 0.8750, 0.5133, 0.3825, 0.4750, 0.5637, 0.5083],
  26444. [0.6310, 0.4017, 0.8563, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006],
  26445. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  26446. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284]],
  26447. device='cuda:0', dtype=torch.float64)
  26448. predictions are: tensor([[0.6350, 0.4143, 0.7841, 0.2905, 0.3769, 0.3715, 0.5888, 0.5981],
  26449. [0.6583, 0.4429, 0.8623, 0.4535, 0.3923, 0.4489, 0.5495, 0.5690],
  26450. [0.0326, 0.0221, 0.9258, 0.3567, 0.5297, 0.2380, 0.7122, 0.5763],
  26451. [0.6316, 0.4160, 0.8865, 0.4939, 0.3749, 0.4176, 0.6057, 0.5378],
  26452. [0.5989, 0.4009, 0.8659, 0.5195, 0.3898, 0.4810, 0.5427, 0.5191],
  26453. [0.6340, 0.4147, 0.8485, 0.5868, 0.3658, 0.4756, 0.6203, 0.5119],
  26454. [0.6492, 0.4303, 0.9090, 0.4505, 0.4247, 0.3199, 0.6731, 0.5689],
  26455. [0.7051, 0.4617, 0.7650, 0.2274, 0.4712, 0.1993, 0.5996, 0.5446]],
  26456. device='cuda:0', grad_fn=<AddmmBackward>)
  26457. landmarks are: tensor([[[0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5987, 0.5700],
  26458. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  26459. [0.0000, 0.0000, 0.9050, 0.3500, 0.5138, 0.2300, 0.7359, 0.5702],
  26460. [0.6236, 0.3965, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  26461. [0.6102, 0.3999, 0.8750, 0.5133, 0.3825, 0.4750, 0.5638, 0.5083],
  26462. [0.6310, 0.4017, 0.8562, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006],
  26463. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  26464. [0.6196, 0.4068, 0.7645, 0.2234, 0.4575, 0.1737, 0.5926, 0.5284]]],
  26465. device='cuda:0')
  26466. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  26467. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  26468. loss_train: 0.03343431546818465
  26469. step: 51
  26470. running loss: 0.0006555748131016598
  26471. Train Steps: 51/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26472. torch.Size([8, 8])
  26473. tensor([[0.6202, 0.4066, 0.8746, 0.3376, 0.3717, 0.3090, 0.5842, 0.5165],
  26474. [0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  26475. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  26476. [0.6135, 0.3994, 0.7913, 0.3050, 0.3625, 0.3050, 0.5837, 0.5050],
  26477. [0.6163, 0.4114, 0.7650, 0.2017, 0.3763, 0.2867, 0.5631, 0.5071],
  26478. [0.6311, 0.3998, 0.7975, 0.5767, 0.3838, 0.4850, 0.7327, 0.5343],
  26479. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  26480. [0.6109, 0.4036, 0.7188, 0.1750, 0.3850, 0.2550, 0.5863, 0.5567]],
  26481. device='cuda:0', dtype=torch.float64)
  26482. predictions are: tensor([[0.5903, 0.3858, 0.8913, 0.3479, 0.4020, 0.3025, 0.5814, 0.5338],
  26483. [0.6145, 0.3955, 0.9317, 0.4338, 0.3939, 0.4503, 0.6172, 0.5075],
  26484. [0.5838, 0.3638, 0.8573, 0.3198, 0.3734, 0.3658, 0.6123, 0.5476],
  26485. [0.6290, 0.4002, 0.8091, 0.3085, 0.3937, 0.3147, 0.5827, 0.5231],
  26486. [0.6167, 0.4136, 0.7696, 0.2070, 0.4055, 0.2694, 0.5670, 0.5264],
  26487. [0.5940, 0.3831, 0.8313, 0.5702, 0.4039, 0.4745, 0.7099, 0.5393],
  26488. [0.6499, 0.4372, 0.9018, 0.4441, 0.3836, 0.3688, 0.5787, 0.5647],
  26489. [0.6392, 0.4080, 0.7350, 0.1964, 0.4205, 0.2529, 0.5691, 0.5825]],
  26490. device='cuda:0', grad_fn=<AddmmBackward>)
  26491. landmarks are: tensor([[[0.6202, 0.4066, 0.8746, 0.3376, 0.3717, 0.3090, 0.5842, 0.5165],
  26492. [0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  26493. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  26494. [0.6135, 0.3994, 0.7912, 0.3050, 0.3625, 0.3050, 0.5838, 0.5050],
  26495. [0.6163, 0.4114, 0.7650, 0.2017, 0.3762, 0.2867, 0.5631, 0.5071],
  26496. [0.6311, 0.3998, 0.7975, 0.5767, 0.3837, 0.4850, 0.7327, 0.5343],
  26497. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  26498. [0.6108, 0.4036, 0.7188, 0.1750, 0.3850, 0.2550, 0.5863, 0.5567]]],
  26499. device='cuda:0')
  26500. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  26501. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  26502. loss_train: 0.03381663607433438
  26503. step: 52
  26504. running loss: 0.0006503199245064304
  26505.  
  26506. Train Steps: 52/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26507. torch.Size([8, 8])
  26508. tensor([[0.6239, 0.4174, 0.8425, 0.5733, 0.4825, 0.4500, 0.5625, 0.5933],
  26509. [0.6175, 0.4091, 0.7863, 0.2800, 0.3638, 0.3583, 0.6188, 0.5433],
  26510. [0.6147, 0.4026, 0.6600, 0.2467, 0.4088, 0.2150, 0.5489, 0.5773],
  26511. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  26512. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667],
  26513. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  26514. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  26515. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609]],
  26516. device='cuda:0', dtype=torch.float64)
  26517. predictions are: tensor([[0.6216, 0.4100, 0.8862, 0.5867, 0.4946, 0.4252, 0.5894, 0.5776],
  26518. [0.5798, 0.3730, 0.8235, 0.2799, 0.3553, 0.3580, 0.5904, 0.5176],
  26519. [0.5872, 0.3870, 0.6935, 0.2447, 0.4144, 0.2064, 0.5287, 0.5627],
  26520. [0.5558, 0.3519, 0.9112, 0.5062, 0.3737, 0.4437, 0.5028, 0.5373],
  26521. [0.5895, 0.3918, 0.8638, 0.3317, 0.3426, 0.3596, 0.5723, 0.5447],
  26522. [0.5498, 0.3511, 0.7831, 0.2351, 0.3763, 0.3139, 0.6081, 0.5010],
  26523. [0.5845, 0.3780, 0.9155, 0.5267, 0.3823, 0.3203, 0.6370, 0.4975],
  26524. [0.5984, 0.3738, 0.8927, 0.2421, 0.5714, 0.2009, 0.7444, 0.5272]],
  26525. device='cuda:0', grad_fn=<AddmmBackward>)
  26526. landmarks are: tensor([[[0.6239, 0.4174, 0.8425, 0.5733, 0.4825, 0.4500, 0.5625, 0.5933],
  26527. [0.6175, 0.4091, 0.7862, 0.2800, 0.3638, 0.3583, 0.6187, 0.5433],
  26528. [0.6147, 0.4026, 0.6600, 0.2467, 0.4087, 0.2150, 0.5489, 0.5773],
  26529. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  26530. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667],
  26531. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  26532. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  26533. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609]]],
  26534. device='cuda:0')
  26535. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  26536. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  26537. loss_train: 0.03452799213118851
  26538. step: 53
  26539. running loss: 0.0006514715496450663
  26540. Train Steps: 53/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26541. torch.Size([8, 8])
  26542. tensor([[0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  26543. [0.6092, 0.4001, 0.8638, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  26544. [0.6082, 0.4024, 0.8738, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  26545. [0.6154, 0.4112, 0.7037, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  26546. [0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431],
  26547. [0.6218, 0.4098, 0.7238, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  26548. [0.6276, 0.4002, 0.8800, 0.5533, 0.3575, 0.4400, 0.6132, 0.4672],
  26549. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617]],
  26550. device='cuda:0', dtype=torch.float64)
  26551. predictions are: tensor([[0.5949, 0.3693, 0.9166, 0.3965, 0.4438, 0.2190, 0.6329, 0.4888],
  26552. [0.5678, 0.3612, 0.8888, 0.4901, 0.4240, 0.5450, 0.5531, 0.5003],
  26553. [0.5673, 0.3601, 0.8953, 0.4152, 0.3538, 0.3924, 0.5423, 0.4920],
  26554. [0.6909, 0.4426, 0.7072, 0.2401, 0.4204, 0.1919, 0.5422, 0.5530],
  26555. [0.5453, 0.3431, 0.8851, 0.4871, 0.4364, 0.5039, 0.5110, 0.5051],
  26556. [0.6636, 0.4255, 0.7338, 0.2045, 0.4271, 0.2490, 0.6436, 0.5372],
  26557. [0.5560, 0.3329, 0.8886, 0.5402, 0.3658, 0.4380, 0.6252, 0.4615],
  26558. [0.6067, 0.3854, 0.8926, 0.4606, 0.3783, 0.4371, 0.5586, 0.5382]],
  26559. device='cuda:0', grad_fn=<AddmmBackward>)
  26560. landmarks are: tensor([[[0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  26561. [0.6092, 0.4001, 0.8637, 0.4867, 0.4288, 0.5367, 0.5484, 0.5064],
  26562. [0.6082, 0.4024, 0.8737, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136],
  26563. [0.6154, 0.4112, 0.7038, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  26564. [0.6107, 0.4050, 0.8700, 0.4850, 0.4470, 0.4848, 0.5043, 0.5431],
  26565. [0.6218, 0.4098, 0.7237, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  26566. [0.6276, 0.4002, 0.8800, 0.5533, 0.3575, 0.4400, 0.6132, 0.4672],
  26567. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617]]],
  26568. device='cuda:0')
  26569. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  26570. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  26571. loss_train: 0.035225964384153485
  26572. step: 54
  26573. running loss: 0.0006523326737806201
  26574. Train Steps: 54/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26575. torch.Size([8, 8])
  26576. tensor([[0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153],
  26577. [0.6059, 0.4002, 0.7562, 0.2767, 0.3538, 0.3033, 0.5529, 0.5455],
  26578. [0.6173, 0.4114, 0.7325, 0.2500, 0.4213, 0.1917, 0.5338, 0.5700],
  26579. [0.6026, 0.3979, 0.8550, 0.4233, 0.3613, 0.5233, 0.5582, 0.4967],
  26580. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  26581. [0.6058, 0.3986, 0.8324, 0.4626, 0.3838, 0.4983, 0.5147, 0.5466],
  26582. [0.6266, 0.4067, 0.8588, 0.2867, 0.4300, 0.2850, 0.6325, 0.5267],
  26583. [0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166]],
  26584. device='cuda:0', dtype=torch.float64)
  26585. predictions are: tensor([[0.5942, 0.3785, 0.8978, 0.5211, 0.4668, 0.5194, 0.6236, 0.4956],
  26586. [0.5834, 0.3925, 0.7657, 0.2807, 0.3593, 0.2938, 0.5746, 0.5117],
  26587. [0.6970, 0.4616, 0.7445, 0.2467, 0.4266, 0.1788, 0.5442, 0.5388],
  26588. [0.5240, 0.3442, 0.8705, 0.4321, 0.3640, 0.5122, 0.5731, 0.4977],
  26589. [0.5492, 0.3548, 0.8602, 0.3845, 0.3575, 0.3699, 0.6041, 0.5329],
  26590. [0.5581, 0.3591, 0.8418, 0.4721, 0.3978, 0.4798, 0.5361, 0.5162],
  26591. [0.6528, 0.4251, 0.8704, 0.2919, 0.4347, 0.2784, 0.6402, 0.5054],
  26592. [0.6352, 0.3991, 0.9248, 0.4779, 0.3932, 0.5056, 0.6813, 0.4867]],
  26593. device='cuda:0', grad_fn=<AddmmBackward>)
  26594. landmarks are: tensor([[[0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153],
  26595. [0.6059, 0.4002, 0.7563, 0.2767, 0.3537, 0.3033, 0.5529, 0.5455],
  26596. [0.6173, 0.4114, 0.7325, 0.2500, 0.4212, 0.1917, 0.5337, 0.5700],
  26597. [0.6026, 0.3979, 0.8550, 0.4233, 0.3613, 0.5233, 0.5582, 0.4967],
  26598. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  26599. [0.6058, 0.3986, 0.8324, 0.4626, 0.3837, 0.4983, 0.5147, 0.5466],
  26600. [0.6266, 0.4067, 0.8587, 0.2867, 0.4300, 0.2850, 0.6325, 0.5267],
  26601. [0.6236, 0.3977, 0.8985, 0.4806, 0.3835, 0.5216, 0.6613, 0.5166]]],
  26602. device='cuda:0')
  26603. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  26604. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  26605. loss_train: 0.03592075948836282
  26606. step: 55
  26607. running loss: 0.0006531047179702331
  26608. Train Steps: 55/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26609. torch.Size([8, 8])
  26610. tensor([[0.6250, 0.4106, 0.8700, 0.3717, 0.3588, 0.4967, 0.6038, 0.5167],
  26611. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  26612. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5413, 0.5717],
  26613. [0.6147, 0.4026, 0.6600, 0.2467, 0.4088, 0.2150, 0.5489, 0.5773],
  26614. [0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  26615. [0.6189, 0.4049, 0.8888, 0.4417, 0.4213, 0.5200, 0.5988, 0.5633],
  26616. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  26617. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385]],
  26618. device='cuda:0', dtype=torch.float64)
  26619. predictions are: tensor([[0.6009, 0.3886, 0.8640, 0.3660, 0.3473, 0.4980, 0.6004, 0.5025],
  26620. [0.6008, 0.3929, 0.8754, 0.4582, 0.3570, 0.5216, 0.5903, 0.4871],
  26621. [0.6102, 0.4029, 0.8723, 0.4841, 0.4216, 0.4752, 0.5486, 0.5360],
  26622. [0.6062, 0.4069, 0.6612, 0.2374, 0.3979, 0.2247, 0.5234, 0.5572],
  26623. [0.5861, 0.3793, 0.8570, 0.3509, 0.3809, 0.2627, 0.6151, 0.4455],
  26624. [0.6013, 0.3983, 0.8778, 0.4348, 0.4101, 0.5217, 0.5876, 0.5433],
  26625. [0.6018, 0.3773, 0.8865, 0.4585, 0.3473, 0.3775, 0.6076, 0.4694],
  26626. [0.6657, 0.4308, 0.8050, 0.2541, 0.4369, 0.2434, 0.6707, 0.5185]],
  26627. device='cuda:0', grad_fn=<AddmmBackward>)
  26628. landmarks are: tensor([[[0.6250, 0.4105, 0.8700, 0.3717, 0.3587, 0.4967, 0.6037, 0.5167],
  26629. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  26630. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5412, 0.5717],
  26631. [0.6147, 0.4026, 0.6600, 0.2467, 0.4087, 0.2150, 0.5489, 0.5773],
  26632. [0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  26633. [0.6189, 0.4049, 0.8888, 0.4417, 0.4212, 0.5200, 0.5987, 0.5633],
  26634. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  26635. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385]]],
  26636. device='cuda:0')
  26637. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  26638. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  26639. loss_train: 0.03618784763966687
  26640. step: 56
  26641. running loss: 0.0006462115649940513
  26642.  
  26643. Train Steps: 56/90 Loss: 0.0006 torch.Size([8, 600, 800])
  26644. torch.Size([8, 8])
  26645. tensor([[ nan, nan, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  26646. [0.6175, 0.3957, 0.8700, 0.4817, 0.4662, 0.5133, 0.5800, 0.5517],
  26647. [0.6275, 0.4157, 0.8337, 0.5800, 0.3763, 0.4200, 0.5547, 0.6125],
  26648. [0.6277, 0.4029, 0.8250, 0.2433, 0.4325, 0.2100, 0.6366, 0.5207],
  26649. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5413, 0.5717],
  26650. [0.6200, 0.3999, 0.8653, 0.5207, 0.4100, 0.5125, 0.5975, 0.5103],
  26651. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  26652. [0.6214, 0.4112, 0.7838, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083]],
  26653. device='cuda:0', dtype=torch.float64)
  26654. predictions are: tensor([[0.1162, 0.0801, 0.6802, 0.1951, 0.4182, 0.2635, 0.5322, 0.5675],
  26655. [0.6565, 0.4317, 0.8646, 0.4593, 0.4474, 0.4991, 0.5942, 0.5406],
  26656. [0.6423, 0.4291, 0.8379, 0.5595, 0.3659, 0.4292, 0.5599, 0.6017],
  26657. [0.6656, 0.4248, 0.8112, 0.2486, 0.4119, 0.2329, 0.6516, 0.5067],
  26658. [0.6346, 0.4177, 0.8700, 0.4794, 0.4136, 0.4643, 0.5676, 0.5569],
  26659. [0.6783, 0.4386, 0.8775, 0.5302, 0.3981, 0.5214, 0.5827, 0.4985],
  26660. [0.6109, 0.3950, 0.8578, 0.5303, 0.3491, 0.3746, 0.5703, 0.5402],
  26661. [0.6370, 0.4200, 0.7620, 0.1908, 0.3480, 0.3000, 0.5853, 0.4956]],
  26662. device='cuda:0', grad_fn=<AddmmBackward>)
  26663. landmarks are: tensor([[[0.0000, 0.0000, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  26664. [0.6175, 0.3957, 0.8700, 0.4817, 0.4663, 0.5133, 0.5800, 0.5517],
  26665. [0.6275, 0.4157, 0.8338, 0.5800, 0.3762, 0.4200, 0.5547, 0.6125],
  26666. [0.6277, 0.4029, 0.8250, 0.2433, 0.4325, 0.2100, 0.6366, 0.5207],
  26667. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5412, 0.5717],
  26668. [0.6200, 0.3999, 0.8653, 0.5207, 0.4100, 0.5125, 0.5975, 0.5103],
  26669. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  26670. [0.6214, 0.4112, 0.7837, 0.2117, 0.3650, 0.3133, 0.5675, 0.5083]]],
  26671. device='cuda:0')
  26672. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  26673. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  26674. loss_train: 0.036844901071162894
  26675. step: 57
  26676. running loss: 0.0006464017731782964
  26677. Train Steps: 57/90 Loss: 0.0006 torch.Size([8, 600, 800])
  26678. torch.Size([8, 8])
  26679. tensor([[0.6201, 0.4064, 0.8688, 0.5050, 0.4225, 0.5100, 0.6138, 0.5500],
  26680. [ nan, nan, 0.8850, 0.2817, 0.5112, 0.2183, 0.7184, 0.5436],
  26681. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  26682. [0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  26683. [0.6114, 0.4018, 0.7213, 0.1967, 0.3763, 0.2700, 0.5875, 0.5533],
  26684. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  26685. [0.6132, 0.4118, 0.8200, 0.3633, 0.3563, 0.5400, 0.5787, 0.5136],
  26686. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343]],
  26687. device='cuda:0', dtype=torch.float64)
  26688. predictions are: tensor([[0.6270, 0.4258, 0.8462, 0.4879, 0.3983, 0.5099, 0.5488, 0.5659],
  26689. [0.1573, 0.1195, 0.8400, 0.2562, 0.4836, 0.2424, 0.6582, 0.5480],
  26690. [0.6321, 0.4317, 0.7994, 0.2551, 0.4191, 0.2894, 0.6537, 0.5727],
  26691. [0.6228, 0.4126, 0.8361, 0.6025, 0.3768, 0.4785, 0.5576, 0.4819],
  26692. [0.5831, 0.3975, 0.6814, 0.1729, 0.3564, 0.2846, 0.5472, 0.5592],
  26693. [0.6615, 0.4358, 0.8663, 0.3155, 0.4706, 0.2533, 0.6759, 0.5524],
  26694. [0.5915, 0.4250, 0.7890, 0.3555, 0.3250, 0.5123, 0.5517, 0.5376],
  26695. [0.6395, 0.4154, 0.8463, 0.2810, 0.4706, 0.2473, 0.6771, 0.5393]],
  26696. device='cuda:0', grad_fn=<AddmmBackward>)
  26697. landmarks are: tensor([[[0.6201, 0.4064, 0.8687, 0.5050, 0.4225, 0.5100, 0.6137, 0.5500],
  26698. [0.0000, 0.0000, 0.8850, 0.2817, 0.5113, 0.2183, 0.7184, 0.5436],
  26699. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  26700. [0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  26701. [0.6114, 0.4018, 0.7212, 0.1967, 0.3762, 0.2700, 0.5875, 0.5533],
  26702. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  26703. [0.6132, 0.4118, 0.8200, 0.3633, 0.3562, 0.5400, 0.5787, 0.5136],
  26704. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343]]],
  26705. device='cuda:0')
  26706. loss_train_step before backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  26707. loss_train_step after backward: tensor(0.0013, device='cuda:0', grad_fn=<MseLossBackward>)
  26708. loss_train: 0.038141772645758465
  26709. step: 58
  26710. running loss: 0.0006576167697544563
  26711. Train Steps: 58/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26712. torch.Size([8, 8])
  26713. tensor([[0.6111, 0.4033, 0.8300, 0.3267, 0.3588, 0.3333, 0.5444, 0.5637],
  26714. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5637, 0.5633],
  26715. [0.6200, 0.4049, 0.8638, 0.5617, 0.4125, 0.5100, 0.6013, 0.5317],
  26716. [0.6277, 0.4057, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  26717. [0.6201, 0.4102, 0.7288, 0.2417, 0.4150, 0.2383, 0.6100, 0.5500],
  26718. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  26719. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395],
  26720. [0.6087, 0.3976, 0.8337, 0.3867, 0.3713, 0.3117, 0.5938, 0.5300]],
  26721. device='cuda:0', dtype=torch.float64)
  26722. predictions are: tensor([[0.5434, 0.3604, 0.8125, 0.3093, 0.3506, 0.3417, 0.5186, 0.5918],
  26723. [0.6032, 0.4228, 0.8550, 0.4658, 0.3610, 0.3783, 0.5531, 0.5995],
  26724. [0.6350, 0.4280, 0.8495, 0.5566, 0.4133, 0.5111, 0.6015, 0.5680],
  26725. [0.6236, 0.4199, 0.8116, 0.2509, 0.4350, 0.2018, 0.6264, 0.5068],
  26726. [0.5966, 0.3983, 0.7309, 0.2129, 0.3987, 0.2585, 0.5930, 0.5883],
  26727. [0.6126, 0.3992, 0.8214, 0.5925, 0.3858, 0.5042, 0.5894, 0.5219],
  26728. [0.5975, 0.3883, 0.7192, 0.2058, 0.3947, 0.2623, 0.5913, 0.5678],
  26729. [0.6132, 0.4144, 0.8182, 0.3650, 0.3688, 0.3395, 0.5849, 0.5806]],
  26730. device='cuda:0', grad_fn=<AddmmBackward>)
  26731. landmarks are: tensor([[[0.6111, 0.4033, 0.8300, 0.3267, 0.3587, 0.3333, 0.5444, 0.5637],
  26732. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5638, 0.5633],
  26733. [0.6199, 0.4049, 0.8637, 0.5617, 0.4125, 0.5100, 0.6012, 0.5317],
  26734. [0.6277, 0.4056, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  26735. [0.6201, 0.4102, 0.7287, 0.2417, 0.4150, 0.2383, 0.6100, 0.5500],
  26736. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  26737. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395],
  26738. [0.6087, 0.3976, 0.8338, 0.3867, 0.3713, 0.3117, 0.5938, 0.5300]]],
  26739. device='cuda:0')
  26740. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  26741. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  26742. loss_train: 0.0386340361146722
  26743. step: 59
  26744. running loss: 0.000654814171435122
  26745. Train Steps: 59/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26746. torch.Size([8, 8])
  26747. tensor([[0.6245, 0.4100, 0.7762, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  26748. [0.6236, 0.4084, 0.7738, 0.2133, 0.3663, 0.3233, 0.5813, 0.5567],
  26749. [0.6197, 0.4050, 0.7527, 0.2000, 0.4042, 0.2249, 0.5895, 0.4995],
  26750. [0.6200, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  26751. [0.6293, 0.4024, 0.8750, 0.5000, 0.4012, 0.5733, 0.7121, 0.5633],
  26752. [0.6142, 0.4127, 0.7575, 0.3067, 0.3438, 0.4383, 0.5778, 0.5207],
  26753. [0.6131, 0.4037, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  26754. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356]],
  26755. device='cuda:0', dtype=torch.float64)
  26756. predictions are: tensor([[0.6260, 0.4173, 0.7693, 0.2477, 0.4896, 0.1354, 0.5837, 0.5601],
  26757. [0.5821, 0.3951, 0.7618, 0.2472, 0.3904, 0.3180, 0.5785, 0.5780],
  26758. [0.5168, 0.3570, 0.7499, 0.2007, 0.4158, 0.2174, 0.6111, 0.5105],
  26759. [0.6055, 0.4041, 0.8060, 0.3109, 0.3993, 0.2829, 0.6012, 0.5408],
  26760. [0.6270, 0.4130, 0.8786, 0.5441, 0.4146, 0.5860, 0.7014, 0.5775],
  26761. [0.5542, 0.3892, 0.7627, 0.2894, 0.3791, 0.4210, 0.5780, 0.5523],
  26762. [0.5910, 0.3917, 0.6935, 0.2902, 0.3789, 0.2751, 0.5223, 0.5864],
  26763. [0.6000, 0.3858, 0.8019, 0.2483, 0.4373, 0.2467, 0.6337, 0.5463]],
  26764. device='cuda:0', grad_fn=<AddmmBackward>)
  26765. landmarks are: tensor([[[0.6245, 0.4100, 0.7763, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  26766. [0.6236, 0.4084, 0.7738, 0.2133, 0.3663, 0.3233, 0.5813, 0.5567],
  26767. [0.6197, 0.4050, 0.7527, 0.2000, 0.4042, 0.2249, 0.5895, 0.4995],
  26768. [0.6201, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  26769. [0.6293, 0.4024, 0.8750, 0.5000, 0.4013, 0.5733, 0.7121, 0.5633],
  26770. [0.6142, 0.4127, 0.7575, 0.3067, 0.3438, 0.4383, 0.5778, 0.5207],
  26771. [0.6131, 0.4036, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  26772. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356]]],
  26773. device='cuda:0')
  26774. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  26775. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  26776. loss_train: 0.03914449064177461
  26777. step: 60
  26778. running loss: 0.0006524081773629102
  26779.  
  26780. Train Steps: 60/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26781. torch.Size([8, 8])
  26782. tensor([[0.6213, 0.4001, 0.7712, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183],
  26783. [0.6109, 0.4041, 0.6975, 0.3167, 0.3513, 0.3383, 0.5153, 0.5319],
  26784. [0.6126, 0.4039, 0.8237, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138],
  26785. [0.6332, 0.4165, 0.9100, 0.3350, 0.4188, 0.3683, 0.7438, 0.5528],
  26786. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683],
  26787. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  26788. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  26789. [0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862]],
  26790. device='cuda:0', dtype=torch.float64)
  26791. predictions are: tensor([[0.6002, 0.3741, 0.7528, 0.2012, 0.4403, 0.1921, 0.5857, 0.5105],
  26792. [0.5763, 0.3811, 0.7150, 0.3156, 0.3673, 0.3497, 0.5242, 0.5349],
  26793. [0.6692, 0.4403, 0.8189, 0.4075, 0.3779, 0.3896, 0.6108, 0.6077],
  26794. [0.5852, 0.3847, 0.8953, 0.3522, 0.4302, 0.3928, 0.7393, 0.5533],
  26795. [0.5822, 0.3798, 0.8676, 0.5529, 0.3981, 0.3941, 0.5389, 0.5724],
  26796. [0.5916, 0.3839, 0.7662, 0.2230, 0.4605, 0.1916, 0.6009, 0.5561],
  26797. [0.5540, 0.3630, 0.8852, 0.3655, 0.4472, 0.2216, 0.6218, 0.5047],
  26798. [0.6438, 0.4159, 0.8962, 0.4108, 0.3776, 0.4650, 0.6316, 0.4925]],
  26799. device='cuda:0', grad_fn=<AddmmBackward>)
  26800. landmarks are: tensor([[[0.6213, 0.4001, 0.7713, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183],
  26801. [0.6109, 0.4041, 0.6975, 0.3167, 0.3512, 0.3383, 0.5153, 0.5319],
  26802. [0.6126, 0.4038, 0.8238, 0.3967, 0.3625, 0.3600, 0.5894, 0.6138],
  26803. [0.6332, 0.4165, 0.9100, 0.3350, 0.4187, 0.3683, 0.7438, 0.5528],
  26804. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683],
  26805. [0.6201, 0.4050, 0.7757, 0.2234, 0.4459, 0.1798, 0.5975, 0.5426],
  26806. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986],
  26807. [0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862]]],
  26808. device='cuda:0')
  26809. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  26810. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  26811. loss_train: 0.039610637846635655
  26812. step: 61
  26813. running loss: 0.0006493547187973058
  26814. Train Steps: 61/90 Loss: 0.0006 torch.Size([8, 600, 800])
  26815. torch.Size([8, 8])
  26816. tensor([[0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274],
  26817. [0.6102, 0.4005, 0.8688, 0.5100, 0.4813, 0.5400, 0.5404, 0.5064],
  26818. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  26819. [0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6262, 0.5367],
  26820. [0.6332, 0.4118, 0.9238, 0.4267, 0.4012, 0.4733, 0.7525, 0.5436],
  26821. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5837, 0.5500],
  26822. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  26823. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6188, 0.5283]],
  26824. device='cuda:0', dtype=torch.float64)
  26825. predictions are: tensor([[0.6490, 0.4308, 0.8600, 0.3135, 0.4776, 0.1786, 0.6367, 0.5065],
  26826. [0.5685, 0.3786, 0.8593, 0.4983, 0.4870, 0.5088, 0.5418, 0.4965],
  26827. [0.5677, 0.3706, 0.8188, 0.2409, 0.4533, 0.2450, 0.7117, 0.5441],
  26828. [0.6425, 0.4133, 0.8852, 0.4869, 0.4178, 0.5740, 0.6153, 0.5150],
  26829. [0.6038, 0.3772, 0.9128, 0.4450, 0.4081, 0.4684, 0.7507, 0.5351],
  26830. [0.6241, 0.3928, 0.8778, 0.4593, 0.4418, 0.5083, 0.5830, 0.5158],
  26831. [0.6153, 0.3958, 0.8522, 0.3992, 0.3652, 0.3736, 0.5988, 0.5468],
  26832. [0.5823, 0.3788, 0.8807, 0.3593, 0.3960, 0.2584, 0.6143, 0.5197]],
  26833. device='cuda:0', grad_fn=<AddmmBackward>)
  26834. landmarks are: tensor([[[0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274],
  26835. [0.6102, 0.4005, 0.8687, 0.5100, 0.4812, 0.5400, 0.5404, 0.5064],
  26836. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  26837. [0.6161, 0.4076, 0.8900, 0.4667, 0.4125, 0.5917, 0.6263, 0.5367],
  26838. [0.6332, 0.4118, 0.9237, 0.4267, 0.4013, 0.4733, 0.7525, 0.5436],
  26839. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5838, 0.5500],
  26840. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  26841. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6187, 0.5283]]],
  26842. device='cuda:0')
  26843. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  26844. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  26845. loss_train: 0.039949122903635725
  26846. step: 62
  26847. running loss: 0.0006443406919941246
  26848. Train Steps: 62/90 Loss: 0.0006 torch.Size([8, 600, 800])
  26849. torch.Size([8, 8])
  26850. tensor([[0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  26851. [0.6250, 0.4110, 0.7238, 0.2067, 0.4263, 0.1883, 0.5625, 0.5633],
  26852. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  26853. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5413, 0.5717],
  26854. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  26855. [0.6273, 0.4110, 0.8900, 0.3817, 0.4188, 0.2167, 0.5858, 0.4835],
  26856. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  26857. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795]],
  26858. device='cuda:0', dtype=torch.float64)
  26859. predictions are: tensor([[0.6489, 0.4172, 0.8604, 0.5817, 0.3890, 0.5051, 0.6722, 0.5223],
  26860. [0.5758, 0.3802, 0.7537, 0.2132, 0.4380, 0.1834, 0.5846, 0.5585],
  26861. [0.6042, 0.4036, 0.8379, 0.2920, 0.4279, 0.2024, 0.5808, 0.5277],
  26862. [0.6568, 0.4249, 0.9017, 0.5003, 0.4414, 0.4782, 0.5668, 0.5598],
  26863. [0.3651, 0.2423, 0.7194, 0.2475, 0.4385, 0.2289, 0.5711, 0.5761],
  26864. [0.6485, 0.4133, 0.9225, 0.3724, 0.4294, 0.2090, 0.6064, 0.4847],
  26865. [0.5770, 0.3819, 0.7592, 0.1862, 0.4164, 0.2548, 0.6278, 0.5392],
  26866. [0.6558, 0.4108, 0.8697, 0.5434, 0.4043, 0.5355, 0.7380, 0.5651]],
  26867. device='cuda:0', grad_fn=<AddmmBackward>)
  26868. landmarks are: tensor([[[0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  26869. [0.6250, 0.4110, 0.7237, 0.2067, 0.4263, 0.1883, 0.5625, 0.5633],
  26870. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  26871. [0.6125, 0.3983, 0.8750, 0.4867, 0.4275, 0.4783, 0.5412, 0.5717],
  26872. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  26873. [0.6273, 0.4110, 0.8900, 0.3817, 0.4187, 0.2167, 0.5858, 0.4835],
  26874. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  26875. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795]]],
  26876. device='cuda:0')
  26877. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  26878. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  26879. loss_train: 0.04169243582873605
  26880. step: 63
  26881. running loss: 0.000661784695694223
  26882. Train Steps: 63/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26883. torch.Size([8, 8])
  26884. tensor([[0.6317, 0.4038, 0.8287, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  26885. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  26886. [0.6226, 0.4098, 0.8912, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  26887. [0.6250, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6088, 0.5183],
  26888. [0.6240, 0.4217, 0.8150, 0.3133, 0.4425, 0.2650, 0.5650, 0.5817],
  26889. [0.6132, 0.4118, 0.8200, 0.3633, 0.3563, 0.5400, 0.5787, 0.5136],
  26890. [0.6307, 0.4029, 0.8988, 0.4817, 0.3937, 0.3500, 0.7311, 0.5378],
  26891. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116]],
  26892. device='cuda:0', dtype=torch.float64)
  26893. predictions are: tensor([[0.6751, 0.4236, 0.8554, 0.5983, 0.3808, 0.4436, 0.6582, 0.5019],
  26894. [0.6391, 0.4155, 0.8110, 0.3076, 0.3625, 0.3694, 0.5915, 0.5181],
  26895. [0.6277, 0.4008, 0.9097, 0.3904, 0.4020, 0.2135, 0.5999, 0.5297],
  26896. [0.6436, 0.3957, 0.8956, 0.4783, 0.4681, 0.5160, 0.6527, 0.5159],
  26897. [0.5659, 0.3881, 0.8140, 0.3132, 0.4450, 0.2331, 0.6016, 0.5673],
  26898. [0.6419, 0.4246, 0.8367, 0.3678, 0.3640, 0.4855, 0.6123, 0.5136],
  26899. [0.6364, 0.4119, 0.9123, 0.4822, 0.4031, 0.3257, 0.7460, 0.5331],
  26900. [0.6155, 0.3943, 0.8747, 0.4187, 0.4091, 0.5571, 0.5951, 0.5081]],
  26901. device='cuda:0', grad_fn=<AddmmBackward>)
  26902. landmarks are: tensor([[[0.6317, 0.4038, 0.8288, 0.5900, 0.3800, 0.4717, 0.6295, 0.4986],
  26903. [0.6223, 0.4130, 0.8100, 0.2983, 0.3525, 0.3900, 0.5694, 0.5103],
  26904. [0.6226, 0.4098, 0.8913, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  26905. [0.6251, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6087, 0.5183],
  26906. [0.6240, 0.4217, 0.8150, 0.3133, 0.4425, 0.2650, 0.5650, 0.5817],
  26907. [0.6132, 0.4118, 0.8200, 0.3633, 0.3562, 0.5400, 0.5787, 0.5136],
  26908. [0.6307, 0.4029, 0.8988, 0.4817, 0.3938, 0.3500, 0.7311, 0.5378],
  26909. [0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116]]],
  26910. device='cuda:0')
  26911. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  26912. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  26913. loss_train: 0.04215108629432507
  26914. step: 64
  26915. running loss: 0.0006586107233488292
  26916.  
  26917. Train Steps: 64/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26918. torch.Size([8, 8])
  26919. tensor([[0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  26920. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  26921. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6038, 0.4833],
  26922. [0.6273, 0.4105, 0.8988, 0.4517, 0.3912, 0.2550, 0.5894, 0.4811],
  26923. [ nan, nan, 0.7525, 0.2291, 0.3838, 0.3017, 0.6050, 0.5667],
  26924. [0.6199, 0.4060, 0.8888, 0.4667, 0.3800, 0.5050, 0.6188, 0.5433],
  26925. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  26926. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758]],
  26927. device='cuda:0', dtype=torch.float64)
  26928. predictions are: tensor([[0.6841, 0.4391, 0.8785, 0.4640, 0.3834, 0.4200, 0.5930, 0.5421],
  26929. [0.6607, 0.4358, 0.8822, 0.5192, 0.4290, 0.5394, 0.6074, 0.5179],
  26930. [0.6769, 0.4259, 0.8845, 0.5061, 0.3831, 0.4829, 0.6011, 0.4728],
  26931. [0.6597, 0.4213, 0.9095, 0.4465, 0.4110, 0.2541, 0.5965, 0.4808],
  26932. [0.0435, 0.0366, 0.7480, 0.2466, 0.3787, 0.2960, 0.6180, 0.5568],
  26933. [0.7205, 0.4638, 0.9050, 0.4852, 0.3870, 0.5248, 0.6364, 0.5229],
  26934. [0.7102, 0.4620, 0.9185, 0.4474, 0.4126, 0.3109, 0.6806, 0.5438],
  26935. [0.7577, 0.4880, 0.9010, 0.5120, 0.4162, 0.5258, 0.6563, 0.4690]],
  26936. device='cuda:0', grad_fn=<AddmmBackward>)
  26937. landmarks are: tensor([[[0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  26938. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  26939. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6037, 0.4833],
  26940. [0.6273, 0.4105, 0.8988, 0.4517, 0.3913, 0.2550, 0.5894, 0.4811],
  26941. [0.0000, 0.0000, 0.7525, 0.2291, 0.3837, 0.3017, 0.6050, 0.5667],
  26942. [0.6199, 0.4060, 0.8888, 0.4667, 0.3800, 0.5050, 0.6187, 0.5433],
  26943. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  26944. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758]]],
  26945. device='cuda:0')
  26946. loss_train_step before backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  26947. loss_train_step after backward: tensor(0.0011, device='cuda:0', grad_fn=<MseLossBackward>)
  26948. loss_train: 0.04324785657809116
  26949. step: 65
  26950. running loss: 0.0006653516396629409
  26951. Train Steps: 65/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26952. torch.Size([8, 8])
  26953. tensor([[0.6286, 0.4060, 0.9188, 0.4333, 0.3675, 0.4167, 0.7034, 0.5528],
  26954. [0.6339, 0.4102, 0.9088, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  26955. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  26956. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  26957. [0.6197, 0.4118, 0.8688, 0.5517, 0.4037, 0.5233, 0.5875, 0.5600],
  26958. [0.6261, 0.3987, 0.8688, 0.4917, 0.4300, 0.5333, 0.7010, 0.5309],
  26959. [0.6250, 0.4131, 0.8688, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  26960. [0.6229, 0.4066, 0.7612, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350]],
  26961. device='cuda:0', dtype=torch.float64)
  26962. predictions are: tensor([[0.6728, 0.4450, 0.9098, 0.4187, 0.3732, 0.4188, 0.6945, 0.5405],
  26963. [0.6582, 0.4256, 0.9019, 0.4799, 0.3953, 0.5400, 0.7138, 0.5375],
  26964. [0.6464, 0.4291, 0.7980, 0.2633, 0.4291, 0.1830, 0.5571, 0.5232],
  26965. [0.6475, 0.4214, 0.8394, 0.5691, 0.3625, 0.5164, 0.6335, 0.5250],
  26966. [0.6215, 0.4021, 0.8771, 0.5360, 0.3976, 0.5311, 0.5583, 0.5489],
  26967. [0.6519, 0.4227, 0.8778, 0.4887, 0.4175, 0.5327, 0.6750, 0.5320],
  26968. [0.6482, 0.4330, 0.8629, 0.2958, 0.4095, 0.2383, 0.6013, 0.5273],
  26969. [0.6709, 0.4361, 0.7736, 0.2770, 0.4107, 0.2393, 0.5732, 0.5426]],
  26970. device='cuda:0', grad_fn=<AddmmBackward>)
  26971. landmarks are: tensor([[[0.6286, 0.4060, 0.9187, 0.4333, 0.3675, 0.4167, 0.7034, 0.5528],
  26972. [0.6339, 0.4102, 0.9087, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  26973. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  26974. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  26975. [0.6197, 0.4118, 0.8687, 0.5517, 0.4038, 0.5233, 0.5875, 0.5600],
  26976. [0.6261, 0.3987, 0.8687, 0.4917, 0.4300, 0.5333, 0.7010, 0.5309],
  26977. [0.6250, 0.4131, 0.8687, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  26978. [0.6229, 0.4066, 0.7613, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350]]],
  26979. device='cuda:0')
  26980. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  26981. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  26982. loss_train: 0.043584351253230125
  26983. step: 66
  26984. running loss: 0.0006603689583822746
  26985. Train Steps: 66/90 Loss: 0.0007 torch.Size([8, 600, 800])
  26986. torch.Size([8, 8])
  26987. tensor([[0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  26988. [0.6205, 0.4081, 0.8950, 0.4017, 0.3788, 0.4700, 0.5963, 0.5667],
  26989. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  26990. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5787, 0.5550],
  26991. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  26992. [0.6109, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  26993. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  26994. [0.6228, 0.4119, 0.7938, 0.2233, 0.4674, 0.1773, 0.6188, 0.5433]],
  26995. device='cuda:0', dtype=torch.float64)
  26996. predictions are: tensor([[0.6590, 0.4183, 0.7758, 0.2653, 0.3696, 0.3135, 0.6233, 0.5215],
  26997. [0.6834, 0.4374, 0.8891, 0.4067, 0.3751, 0.4616, 0.6058, 0.5631],
  26998. [0.6540, 0.4188, 0.8939, 0.4728, 0.4269, 0.5779, 0.6053, 0.5424],
  26999. [0.6287, 0.4122, 0.8230, 0.2927, 0.3532, 0.3614, 0.5817, 0.5574],
  27000. [0.6505, 0.4088, 0.8768, 0.5240, 0.4169, 0.5441, 0.6076, 0.5048],
  27001. [0.6724, 0.4390, 0.8796, 0.4794, 0.3679, 0.4075, 0.5526, 0.5108],
  27002. [0.6069, 0.3898, 0.8456, 0.5323, 0.4040, 0.5576, 0.7046, 0.5654],
  27003. [0.7121, 0.4588, 0.8121, 0.2477, 0.4546, 0.1666, 0.6290, 0.5217]],
  27004. device='cuda:0', grad_fn=<AddmmBackward>)
  27005. landmarks are: tensor([[[0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  27006. [0.6205, 0.4081, 0.8950, 0.4017, 0.3787, 0.4700, 0.5962, 0.5667],
  27007. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  27008. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5788, 0.5550],
  27009. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  27010. [0.6108, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  27011. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  27012. [0.6228, 0.4119, 0.7937, 0.2233, 0.4674, 0.1773, 0.6187, 0.5433]]],
  27013. device='cuda:0')
  27014. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  27015. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  27016. loss_train: 0.04411605268251151
  27017. step: 67
  27018. running loss: 0.0006584485475001718
  27019. Train Steps: 67/90 Loss: 0.0007 torch.Size([8, 600, 800])
  27020. torch.Size([8, 8])
  27021. tensor([[0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  27022. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  27023. [0.6199, 0.4065, 0.7598, 0.2385, 0.4317, 0.1981, 0.5933, 0.5221],
  27024. [0.6197, 0.4050, 0.7527, 0.2000, 0.4042, 0.2249, 0.5895, 0.4995],
  27025. [0.6145, 0.4007, 0.8775, 0.4533, 0.4562, 0.5533, 0.6088, 0.5533],
  27026. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  27027. [0.6179, 0.3998, 0.8396, 0.3505, 0.3552, 0.3768, 0.6064, 0.5329],
  27028. [0.6075, 0.4007, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280]],
  27029. device='cuda:0', dtype=torch.float64)
  27030. predictions are: tensor([[0.6283, 0.4003, 0.8561, 0.3963, 0.3301, 0.3903, 0.6070, 0.5052],
  27031. [0.6088, 0.3770, 0.9110, 0.3335, 0.4759, 0.2409, 0.7457, 0.5498],
  27032. [0.6425, 0.4072, 0.7553, 0.2335, 0.4060, 0.2231, 0.5940, 0.5234],
  27033. [0.6522, 0.4164, 0.7659, 0.1988, 0.3875, 0.2213, 0.6210, 0.5014],
  27034. [0.5815, 0.3730, 0.8650, 0.4474, 0.4370, 0.5787, 0.6060, 0.5581],
  27035. [0.6791, 0.4359, 0.8828, 0.4949, 0.3584, 0.4545, 0.5194, 0.5556],
  27036. [0.6427, 0.4100, 0.8443, 0.3498, 0.3293, 0.3790, 0.6097, 0.5322],
  27037. [0.6070, 0.3958, 0.8130, 0.4942, 0.4050, 0.5089, 0.5191, 0.5420]],
  27038. device='cuda:0', grad_fn=<AddmmBackward>)
  27039. landmarks are: tensor([[[0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131],
  27040. [0.6271, 0.4040, 0.9000, 0.3250, 0.4938, 0.2300, 0.7192, 0.5413],
  27041. [0.6199, 0.4065, 0.7598, 0.2385, 0.4317, 0.1981, 0.5933, 0.5221],
  27042. [0.6197, 0.4050, 0.7527, 0.2000, 0.4042, 0.2249, 0.5895, 0.4995],
  27043. [0.6145, 0.4007, 0.8775, 0.4533, 0.4563, 0.5533, 0.6087, 0.5533],
  27044. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  27045. [0.6179, 0.3998, 0.8396, 0.3505, 0.3552, 0.3768, 0.6064, 0.5329],
  27046. [0.6075, 0.4006, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280]]],
  27047. device='cuda:0')
  27048. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27049. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27050. loss_train: 0.044399865000741556
  27051. step: 68
  27052. running loss: 0.0006529391911873758
  27053.  
  27054. Train Steps: 68/90 Loss: 0.0007 torch.Size([8, 600, 800])
  27055. torch.Size([8, 8])
  27056. tensor([[0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  27057. [0.6205, 0.4081, 0.8950, 0.4017, 0.3788, 0.4700, 0.5963, 0.5667],
  27058. [ nan, nan, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  27059. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  27060. [0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717],
  27061. [0.6085, 0.4005, 0.8400, 0.4317, 0.3763, 0.4750, 0.5476, 0.5058],
  27062. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  27063. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400]],
  27064. device='cuda:0', dtype=torch.float64)
  27065. predictions are: tensor([[0.6684, 0.4103, 0.8983, 0.4467, 0.3564, 0.4676, 0.6509, 0.4979],
  27066. [0.6697, 0.4285, 0.8877, 0.3976, 0.3835, 0.4953, 0.6155, 0.5776],
  27067. [0.1189, 0.0600, 0.6698, 0.2783, 0.3735, 0.2384, 0.5544, 0.5748],
  27068. [0.6611, 0.4208, 0.7464, 0.2118, 0.4308, 0.2101, 0.6222, 0.5398],
  27069. [0.6449, 0.4069, 0.7567, 0.2398, 0.3928, 0.2952, 0.5959, 0.5886],
  27070. [0.6412, 0.4086, 0.8399, 0.4462, 0.3675, 0.4937, 0.5392, 0.5161],
  27071. [0.6842, 0.4399, 0.8795, 0.5051, 0.3818, 0.4660, 0.5229, 0.5636],
  27072. [0.6966, 0.4450, 0.7876, 0.2467, 0.4578, 0.2063, 0.6285, 0.5483]],
  27073. device='cuda:0', grad_fn=<AddmmBackward>)
  27074. landmarks are: tensor([[[0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  27075. [0.6205, 0.4081, 0.8950, 0.4017, 0.3787, 0.4700, 0.5962, 0.5667],
  27076. [0.0000, 0.0000, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  27077. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  27078. [0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717],
  27079. [0.6084, 0.4005, 0.8400, 0.4317, 0.3762, 0.4750, 0.5476, 0.5058],
  27080. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  27081. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400]]],
  27082. device='cuda:0')
  27083. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  27084. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  27085. loss_train: 0.04514229335472919
  27086. step: 69
  27087. running loss: 0.0006542361355757854
  27088. Train Steps: 69/90 Loss: 0.0007 torch.Size([8, 600, 800])
  27089. torch.Size([8, 8])
  27090. tensor([[0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  27091. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6362, 0.5200],
  27092. [0.6289, 0.4019, 0.8113, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332],
  27093. [0.6216, 0.4167, 0.8588, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667],
  27094. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  27095. [0.6268, 0.4061, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  27096. [0.6161, 0.4024, 0.8662, 0.4683, 0.4935, 0.5364, 0.6063, 0.5567],
  27097. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283]],
  27098. device='cuda:0', dtype=torch.float64)
  27099. predictions are: tensor([[0.5932, 0.3854, 0.8746, 0.4924, 0.4156, 0.5271, 0.5691, 0.5149],
  27100. [0.6065, 0.3985, 0.8727, 0.3528, 0.3805, 0.4974, 0.6261, 0.5165],
  27101. [0.5919, 0.3854, 0.8122, 0.5360, 0.3914, 0.4945, 0.6972, 0.5451],
  27102. [0.5852, 0.3936, 0.8503, 0.5523, 0.4109, 0.5134, 0.5696, 0.5734],
  27103. [0.5774, 0.3899, 0.8421, 0.3239, 0.3606, 0.4490, 0.6091, 0.5548],
  27104. [0.6649, 0.4226, 0.8385, 0.2488, 0.4642, 0.1997, 0.6237, 0.5406],
  27105. [0.5808, 0.3709, 0.8631, 0.4307, 0.4711, 0.4975, 0.5966, 0.5669],
  27106. [0.5930, 0.3836, 0.8309, 0.5664, 0.4003, 0.4188, 0.5687, 0.5385]],
  27107. device='cuda:0', grad_fn=<AddmmBackward>)
  27108. landmarks are: tensor([[[0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  27109. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6363, 0.5200],
  27110. [0.6289, 0.4019, 0.8112, 0.5467, 0.3875, 0.5017, 0.7367, 0.5332],
  27111. [0.6216, 0.4167, 0.8587, 0.5583, 0.3975, 0.5167, 0.5775, 0.5667],
  27112. [0.6168, 0.4081, 0.8438, 0.3367, 0.3500, 0.4667, 0.6212, 0.5400],
  27113. [0.6268, 0.4060, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  27114. [0.6161, 0.4024, 0.8662, 0.4683, 0.4935, 0.5364, 0.6062, 0.5567],
  27115. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283]]],
  27116. device='cuda:0')
  27117. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27118. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27119. loss_train: 0.045491363474866375
  27120. step: 70
  27121. running loss: 0.0006498766210695197
  27122. Train Steps: 70/90 Loss: 0.0006 torch.Size([8, 600, 800])
  27123. torch.Size([8, 8])
  27124. tensor([[0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6063, 0.5617],
  27125. [0.6200, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  27126. [0.6311, 0.4008, 0.7935, 0.5746, 0.3900, 0.5033, 0.6955, 0.5366],
  27127. [0.6098, 0.3991, 0.8638, 0.4717, 0.4263, 0.4967, 0.5212, 0.5650],
  27128. [0.6198, 0.4130, 0.8762, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  27129. [0.6371, 0.4092, 0.8337, 0.5850, 0.3950, 0.5117, 0.6559, 0.5262],
  27130. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  27131. [0.6224, 0.4097, 0.7438, 0.2267, 0.3850, 0.2850, 0.5988, 0.5250]],
  27132. device='cuda:0', dtype=torch.float64)
  27133. predictions are: tensor([[0.6027, 0.3895, 0.9023, 0.4631, 0.3711, 0.4540, 0.5978, 0.5551],
  27134. [0.5901, 0.3676, 0.8123, 0.2965, 0.3895, 0.2630, 0.5769, 0.5181],
  27135. [0.5757, 0.3723, 0.7930, 0.5494, 0.3922, 0.4865, 0.6760, 0.5425],
  27136. [0.5815, 0.3743, 0.8654, 0.4666, 0.4310, 0.4858, 0.5209, 0.5583],
  27137. [0.6159, 0.4109, 0.8653, 0.4103, 0.3742, 0.4880, 0.5513, 0.5205],
  27138. [0.6212, 0.4156, 0.8455, 0.5615, 0.4003, 0.5054, 0.6516, 0.5298],
  27139. [0.6269, 0.3967, 0.7233, 0.2441, 0.3980, 0.2832, 0.6001, 0.5731],
  27140. [0.5960, 0.3751, 0.7550, 0.2161, 0.4011, 0.2795, 0.5930, 0.5292]],
  27141. device='cuda:0', grad_fn=<AddmmBackward>)
  27142. landmarks are: tensor([[[0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6062, 0.5617],
  27143. [0.6201, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  27144. [0.6311, 0.4008, 0.7935, 0.5746, 0.3900, 0.5033, 0.6955, 0.5366],
  27145. [0.6098, 0.3991, 0.8637, 0.4717, 0.4263, 0.4967, 0.5213, 0.5650],
  27146. [0.6198, 0.4130, 0.8763, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  27147. [0.6371, 0.4092, 0.8338, 0.5850, 0.3950, 0.5117, 0.6559, 0.5262],
  27148. [0.6206, 0.4123, 0.7175, 0.2400, 0.3887, 0.2933, 0.6225, 0.5750],
  27149. [0.6224, 0.4097, 0.7437, 0.2267, 0.3850, 0.2850, 0.5987, 0.5250]]],
  27150. device='cuda:0')
  27151. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27152. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27153. loss_train: 0.04574897524435073
  27154. step: 71
  27155. running loss: 0.0006443517640049399
  27156. Train Steps: 71/90 Loss: 0.0006 torch.Size([8, 600, 800])
  27157. torch.Size([8, 8])
  27158. tensor([[0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461],
  27159. [0.6038, 0.3946, 0.8413, 0.4883, 0.3563, 0.4550, 0.5266, 0.4693],
  27160. [0.6175, 0.3957, 0.8700, 0.4817, 0.4662, 0.5133, 0.5800, 0.5517],
  27161. [0.6243, 0.4128, 0.7762, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417],
  27162. [0.6102, 0.3999, 0.8750, 0.5133, 0.3825, 0.4750, 0.5637, 0.5083],
  27163. [0.6196, 0.4094, 0.7562, 0.2817, 0.3937, 0.3183, 0.6013, 0.6183],
  27164. [0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  27165. [0.6261, 0.4066, 0.8325, 0.2150, 0.4763, 0.2667, 0.7002, 0.5633]],
  27166. device='cuda:0', dtype=torch.float64)
  27167. predictions are: tensor([[0.5633, 0.3836, 0.8420, 0.5337, 0.4440, 0.5154, 0.5364, 0.5381],
  27168. [0.5384, 0.3566, 0.8358, 0.4855, 0.3616, 0.4613, 0.5191, 0.4777],
  27169. [0.5452, 0.3580, 0.8498, 0.4636, 0.4535, 0.5147, 0.5744, 0.5431],
  27170. [0.5719, 0.3858, 0.7744, 0.2861, 0.3919, 0.3032, 0.6182, 0.5462],
  27171. [0.5891, 0.3935, 0.8548, 0.5052, 0.3834, 0.4553, 0.5580, 0.5031],
  27172. [0.5469, 0.3664, 0.7412, 0.2916, 0.4010, 0.3196, 0.6024, 0.6164],
  27173. [0.5542, 0.3592, 0.8591, 0.4615, 0.4228, 0.4564, 0.5400, 0.5590],
  27174. [0.5844, 0.3841, 0.8278, 0.2302, 0.4739, 0.2515, 0.7062, 0.5385]],
  27175. device='cuda:0', grad_fn=<AddmmBackward>)
  27176. landmarks are: tensor([[[0.6127, 0.4119, 0.8650, 0.5300, 0.4512, 0.5250, 0.5260, 0.5461],
  27177. [0.6038, 0.3946, 0.8413, 0.4883, 0.3562, 0.4550, 0.5266, 0.4693],
  27178. [0.6175, 0.3957, 0.8700, 0.4817, 0.4663, 0.5133, 0.5800, 0.5517],
  27179. [0.6243, 0.4128, 0.7763, 0.2717, 0.3825, 0.3133, 0.6212, 0.5417],
  27180. [0.6102, 0.3999, 0.8750, 0.5133, 0.3825, 0.4750, 0.5638, 0.5083],
  27181. [0.6196, 0.4094, 0.7563, 0.2817, 0.3938, 0.3183, 0.6012, 0.6183],
  27182. [0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  27183. [0.6261, 0.4066, 0.8325, 0.2150, 0.4762, 0.2667, 0.7002, 0.5633]]],
  27184. device='cuda:0')
  27185. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  27186. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  27187. loss_train: 0.04638030182104558
  27188. step: 72
  27189. running loss: 0.0006441708586256331
  27190.  
  27191. Train Steps: 72/90 Loss: 0.0006 torch.Size([8, 600, 800])
  27192. torch.Size([8, 8])
  27193. tensor([[0.6136, 0.4117, 0.8700, 0.5167, 0.4188, 0.5083, 0.5147, 0.5495],
  27194. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6138, 0.5333],
  27195. [0.6219, 0.4097, 0.8738, 0.3400, 0.3563, 0.4117, 0.5975, 0.5683],
  27196. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  27197. [0.6353, 0.4128, 0.8488, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  27198. [0.6222, 0.3937, 0.8350, 0.5617, 0.4138, 0.4600, 0.5800, 0.5233],
  27199. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  27200. [0.6196, 0.4094, 0.7562, 0.2817, 0.3937, 0.3183, 0.6013, 0.6183]],
  27201. device='cuda:0', dtype=torch.float64)
  27202. predictions are: tensor([[0.5476, 0.3818, 0.8540, 0.5142, 0.4185, 0.5273, 0.5357, 0.5329],
  27203. [0.5659, 0.3972, 0.8821, 0.4454, 0.3786, 0.5081, 0.6200, 0.5237],
  27204. [0.5426, 0.3779, 0.8634, 0.3470, 0.3503, 0.4327, 0.5906, 0.5590],
  27205. [0.5297, 0.3636, 0.7151, 0.2196, 0.4366, 0.1792, 0.5518, 0.5310],
  27206. [0.6560, 0.4389, 0.8403, 0.2511, 0.5428, 0.1872, 0.6646, 0.5584],
  27207. [0.5384, 0.3503, 0.8167, 0.5637, 0.4109, 0.4789, 0.5900, 0.5254],
  27208. [0.5101, 0.3595, 0.7190, 0.2721, 0.3746, 0.2856, 0.5514, 0.5176],
  27209. [0.5184, 0.3575, 0.7436, 0.2908, 0.4046, 0.3418, 0.6030, 0.6151]],
  27210. device='cuda:0', grad_fn=<AddmmBackward>)
  27211. landmarks are: tensor([[[0.6136, 0.4117, 0.8700, 0.5167, 0.4187, 0.5083, 0.5147, 0.5495],
  27212. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6137, 0.5333],
  27213. [0.6219, 0.4097, 0.8737, 0.3400, 0.3562, 0.4117, 0.5975, 0.5683],
  27214. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  27215. [0.6353, 0.4128, 0.8487, 0.2600, 0.5525, 0.1616, 0.6694, 0.5540],
  27216. [0.6222, 0.3937, 0.8350, 0.5617, 0.4137, 0.4600, 0.5800, 0.5233],
  27217. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  27218. [0.6196, 0.4094, 0.7563, 0.2817, 0.3938, 0.3183, 0.6012, 0.6183]]],
  27219. device='cuda:0')
  27220. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  27221. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  27222. loss_train: 0.04742685717064887
  27223. step: 73
  27224. running loss: 0.0006496829749403955
  27225. Train Steps: 73/90 Loss: 0.0006 torch.Size([8, 600, 800])
  27226. torch.Size([8, 8])
  27227. tensor([[0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  27228. [0.6114, 0.4018, 0.7213, 0.1967, 0.3763, 0.2700, 0.5875, 0.5533],
  27229. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  27230. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  27231. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200],
  27232. [0.6293, 0.4024, 0.8750, 0.5000, 0.4012, 0.5733, 0.7121, 0.5633],
  27233. [0.6339, 0.4118, 0.7988, 0.5800, 0.3912, 0.4583, 0.7343, 0.5760],
  27234. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6138, 0.5333]],
  27235. device='cuda:0', dtype=torch.float64)
  27236. predictions are: tensor([[0.5360, 0.3843, 0.8365, 0.4999, 0.4472, 0.5498, 0.5199, 0.5358],
  27237. [0.5817, 0.4067, 0.7153, 0.1968, 0.3909, 0.2734, 0.5824, 0.5477],
  27238. [0.6057, 0.4180, 0.7590, 0.2297, 0.4045, 0.2917, 0.6042, 0.5432],
  27239. [0.5751, 0.4026, 0.7735, 0.2757, 0.4548, 0.1920, 0.5601, 0.5316],
  27240. [0.5739, 0.4009, 0.8505, 0.3059, 0.4595, 0.2439, 0.5841, 0.5173],
  27241. [0.5428, 0.3725, 0.8629, 0.4832, 0.4021, 0.5831, 0.6896, 0.5574],
  27242. [0.5653, 0.3907, 0.7998, 0.5568, 0.3800, 0.4650, 0.6964, 0.5689],
  27243. [0.5964, 0.4303, 0.8868, 0.4445, 0.3831, 0.4982, 0.6010, 0.5317]],
  27244. device='cuda:0', grad_fn=<AddmmBackward>)
  27245. landmarks are: tensor([[[0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  27246. [0.6114, 0.4018, 0.7212, 0.1967, 0.3762, 0.2700, 0.5875, 0.5533],
  27247. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  27248. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  27249. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200],
  27250. [0.6293, 0.4024, 0.8750, 0.5000, 0.4013, 0.5733, 0.7121, 0.5633],
  27251. [0.6339, 0.4118, 0.7987, 0.5800, 0.3913, 0.4583, 0.7343, 0.5760],
  27252. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6137, 0.5333]]],
  27253. device='cuda:0')
  27254. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  27255. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  27256. loss_train: 0.047969518578611314
  27257. step: 74
  27258. running loss: 0.0006482367375488015
  27259. Train Steps: 74/90 Loss: 0.0006 torch.Size([8, 600, 800])
  27260. torch.Size([8, 8])
  27261. tensor([[0.6126, 0.4067, 0.8638, 0.5383, 0.4188, 0.4850, 0.5016, 0.5392],
  27262. [0.6229, 0.4107, 0.8137, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  27263. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  27264. [0.6219, 0.4114, 0.8175, 0.2817, 0.3925, 0.2783, 0.5900, 0.5350],
  27265. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  27266. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  27267. [0.6268, 0.4061, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  27268. [0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038]],
  27269. device='cuda:0', dtype=torch.float64)
  27270. predictions are: tensor([[0.5686, 0.4101, 0.8524, 0.5625, 0.4094, 0.4891, 0.5374, 0.5442],
  27271. [0.5875, 0.4116, 0.7930, 0.2905, 0.4642, 0.1922, 0.5773, 0.5451],
  27272. [0.5920, 0.3973, 0.8621, 0.3971, 0.3795, 0.4870, 0.6047, 0.5376],
  27273. [0.5645, 0.3967, 0.7930, 0.2855, 0.4020, 0.2773, 0.5988, 0.5453],
  27274. [0.5812, 0.3969, 0.8747, 0.3446, 0.3547, 0.4323, 0.6985, 0.5203],
  27275. [0.5914, 0.3975, 0.8695, 0.4539, 0.4608, 0.5580, 0.6136, 0.5605],
  27276. [0.5825, 0.3929, 0.8245, 0.2393, 0.4464, 0.2194, 0.6340, 0.5284],
  27277. [0.5679, 0.3847, 0.8464, 0.4783, 0.4435, 0.5161, 0.5504, 0.5069]],
  27278. device='cuda:0', grad_fn=<AddmmBackward>)
  27279. landmarks are: tensor([[[0.6126, 0.4067, 0.8637, 0.5383, 0.4187, 0.4850, 0.5016, 0.5392],
  27280. [0.6229, 0.4107, 0.8138, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  27281. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  27282. [0.6219, 0.4114, 0.8175, 0.2817, 0.3925, 0.2783, 0.5900, 0.5350],
  27283. [0.6271, 0.4024, 0.9000, 0.3517, 0.3700, 0.4517, 0.6931, 0.5285],
  27284. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  27285. [0.6268, 0.4060, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  27286. [0.6086, 0.3981, 0.8700, 0.4750, 0.4512, 0.5283, 0.5324, 0.5038]]],
  27287. device='cuda:0')
  27288. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  27289. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  27290. loss_train: 0.04832018641172908
  27291. step: 75
  27292. running loss: 0.0006442691521563878
  27293. Train Steps: 75/90 Loss: 0.0006 torch.Size([8, 600, 800])
  27294. torch.Size([8, 8])
  27295. tensor([[0.6286, 0.4060, 0.9188, 0.4333, 0.3675, 0.4167, 0.7034, 0.5528],
  27296. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  27297. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  27298. [ nan, nan, 0.8300, 0.3150, 0.3588, 0.3383, 0.5208, 0.5194],
  27299. [0.6206, 0.4001, 0.8900, 0.3933, 0.3588, 0.3567, 0.5837, 0.5083],
  27300. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896],
  27301. [ nan, nan, 0.8525, 0.2217, 0.5413, 0.2367, 0.7367, 0.5482],
  27302. [0.6218, 0.4137, 0.7263, 0.2233, 0.4075, 0.2650, 0.6212, 0.5783]],
  27303. device='cuda:0', dtype=torch.float64)
  27304. predictions are: tensor([[0.6626, 0.4414, 0.9109, 0.4295, 0.3842, 0.4193, 0.7023, 0.5545],
  27305. [0.6793, 0.4529, 0.8732, 0.5148, 0.4264, 0.5669, 0.6052, 0.5096],
  27306. [0.6709, 0.4413, 0.8750, 0.5038, 0.4249, 0.5001, 0.6196, 0.5342],
  27307. [0.2023, 0.1378, 0.8027, 0.3239, 0.3842, 0.3492, 0.5233, 0.5351],
  27308. [0.6833, 0.4441, 0.8951, 0.4070, 0.3554, 0.3622, 0.5769, 0.4957],
  27309. [0.6971, 0.4723, 0.8174, 0.3886, 0.3557, 0.3141, 0.5261, 0.5929],
  27310. [0.1040, 0.0812, 0.8375, 0.2226, 0.5503, 0.2352, 0.7235, 0.5569],
  27311. [0.6915, 0.4451, 0.7427, 0.2376, 0.4066, 0.2849, 0.5986, 0.5743]],
  27312. device='cuda:0', grad_fn=<AddmmBackward>)
  27313. landmarks are: tensor([[[0.6286, 0.4060, 0.9187, 0.4333, 0.3675, 0.4167, 0.7034, 0.5528],
  27314. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  27315. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  27316. [0.0000, 0.0000, 0.8300, 0.3150, 0.3587, 0.3383, 0.5208, 0.5194],
  27317. [0.6206, 0.4001, 0.8900, 0.3933, 0.3587, 0.3567, 0.5838, 0.5083],
  27318. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896],
  27319. [0.0000, 0.0000, 0.8525, 0.2217, 0.5412, 0.2367, 0.7367, 0.5482],
  27320. [0.6218, 0.4137, 0.7262, 0.2233, 0.4075, 0.2650, 0.6212, 0.5783]]],
  27321. device='cuda:0')
  27322. loss_train_step before backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  27323. loss_train_step after backward: tensor(0.0019, device='cuda:0', grad_fn=<MseLossBackward>)
  27324. loss_train: 0.05019753027590923
  27325. step: 76
  27326. running loss: 0.0006604938194198583
  27327.  
  27328. Train Steps: 76/90 Loss: 0.0007 torch.Size([8, 600, 800])
  27329. torch.Size([8, 8])
  27330. tensor([[0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  27331. [ nan, nan, 0.7525, 0.2291, 0.3838, 0.3017, 0.6050, 0.5667],
  27332. [0.6100, 0.4016, 0.8600, 0.5067, 0.4612, 0.5233, 0.5086, 0.5519],
  27333. [0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5988, 0.5700],
  27334. [0.6200, 0.4071, 0.7338, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  27335. [0.6272, 0.4071, 0.8738, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  27336. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  27337. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374]],
  27338. device='cuda:0', dtype=torch.float64)
  27339. predictions are: tensor([[0.6749, 0.4230, 0.8694, 0.3523, 0.3925, 0.2650, 0.6397, 0.4864],
  27340. [0.0600, 0.0301, 0.7583, 0.2349, 0.3915, 0.3052, 0.6079, 0.5664],
  27341. [0.6410, 0.4068, 0.8578, 0.5020, 0.4575, 0.5220, 0.5322, 0.5538],
  27342. [0.6767, 0.4299, 0.8070, 0.2836, 0.3702, 0.3566, 0.6059, 0.5776],
  27343. [0.6658, 0.4451, 0.7451, 0.1958, 0.4262, 0.2458, 0.6349, 0.5657],
  27344. [0.6659, 0.4249, 0.8828, 0.5640, 0.3632, 0.3914, 0.6147, 0.4700],
  27345. [0.6506, 0.4232, 0.8870, 0.4793, 0.3805, 0.4489, 0.5319, 0.5572],
  27346. [0.6846, 0.4381, 0.8686, 0.3619, 0.3694, 0.3212, 0.6116, 0.5322]],
  27347. device='cuda:0', grad_fn=<AddmmBackward>)
  27348. landmarks are: tensor([[[0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742],
  27349. [0.0000, 0.0000, 0.7525, 0.2291, 0.3837, 0.3017, 0.6050, 0.5667],
  27350. [0.6100, 0.4016, 0.8600, 0.5067, 0.4613, 0.5233, 0.5086, 0.5519],
  27351. [0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5987, 0.5700],
  27352. [0.6200, 0.4071, 0.7337, 0.1917, 0.4200, 0.2450, 0.6150, 0.5517],
  27353. [0.6272, 0.4071, 0.8737, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  27354. [0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  27355. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374]]],
  27356. device='cuda:0')
  27357. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  27358. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  27359. loss_train: 0.05063753016293049
  27360. step: 77
  27361. running loss: 0.0006576302618562401
  27362. Train Steps: 77/90 Loss: 0.0007 torch.Size([8, 600, 800])
  27363. torch.Size([8, 8])
  27364. tensor([[0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  27365. [0.6109, 0.4003, 0.8650, 0.4883, 0.4775, 0.4867, 0.5175, 0.5683],
  27366. [0.6230, 0.4152, 0.7588, 0.2283, 0.4012, 0.2883, 0.6200, 0.5767],
  27367. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  27368. [0.6236, 0.3966, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  27369. [0.6254, 0.3993, 0.8988, 0.4767, 0.3987, 0.5517, 0.6955, 0.5285],
  27370. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783],
  27371. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118]],
  27372. device='cuda:0', dtype=torch.float64)
  27373. predictions are: tensor([[0.6171, 0.3931, 0.7081, 0.2966, 0.3679, 0.2957, 0.5455, 0.5772],
  27374. [0.6125, 0.3821, 0.8967, 0.4685, 0.4851, 0.4934, 0.5463, 0.5613],
  27375. [0.6223, 0.3868, 0.8138, 0.2293, 0.4097, 0.2930, 0.6357, 0.5563],
  27376. [0.6632, 0.4119, 0.7575, 0.2619, 0.4331, 0.2198, 0.5652, 0.5772],
  27377. [0.6501, 0.3907, 0.9321, 0.4798, 0.3699, 0.3962, 0.6038, 0.5200],
  27378. [0.5769, 0.3612, 0.9468, 0.4648, 0.3979, 0.5583, 0.7056, 0.5172],
  27379. [0.6279, 0.3926, 0.8856, 0.3573, 0.3740, 0.2902, 0.5336, 0.5701],
  27380. [0.5986, 0.3918, 0.8499, 0.5497, 0.4113, 0.4810, 0.5777, 0.6208]],
  27381. device='cuda:0', grad_fn=<AddmmBackward>)
  27382. landmarks are: tensor([[[0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  27383. [0.6109, 0.4003, 0.8650, 0.4883, 0.4775, 0.4867, 0.5175, 0.5683],
  27384. [0.6230, 0.4152, 0.7588, 0.2283, 0.4013, 0.2883, 0.6200, 0.5767],
  27385. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  27386. [0.6236, 0.3965, 0.8850, 0.4967, 0.3638, 0.4017, 0.5850, 0.5183],
  27387. [0.6254, 0.3993, 0.8988, 0.4767, 0.3988, 0.5517, 0.6955, 0.5285],
  27388. [0.6184, 0.4079, 0.8350, 0.3700, 0.3675, 0.2883, 0.5312, 0.5783],
  27389. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118]]],
  27390. device='cuda:0')
  27391. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  27392. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  27393. loss_train: 0.051187454839237034
  27394. step: 78
  27395. running loss: 0.0006562494210158595
  27396. Train Steps: 78/90 Loss: 0.0007 torch.Size([8, 600, 800])
  27397. torch.Size([8, 8])
  27398. tensor([[0.6111, 0.4019, 0.8350, 0.3867, 0.3500, 0.4283, 0.5480, 0.5719],
  27399. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  27400. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  27401. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  27402. [0.6142, 0.3982, 0.8650, 0.4883, 0.3912, 0.4317, 0.5315, 0.5350],
  27403. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  27404. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  27405. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517]],
  27406. device='cuda:0', dtype=torch.float64)
  27407. predictions are: tensor([[0.6750, 0.4192, 0.8474, 0.3967, 0.3528, 0.4086, 0.5324, 0.5612],
  27408. [0.6266, 0.3706, 0.8872, 0.4848, 0.4445, 0.4700, 0.5373, 0.5792],
  27409. [0.6138, 0.3691, 0.8315, 0.2530, 0.4299, 0.2373, 0.6471, 0.5371],
  27410. [0.6447, 0.3941, 0.7982, 0.3344, 0.3555, 0.3328, 0.5823, 0.5228],
  27411. [0.6479, 0.4008, 0.8742, 0.4860, 0.3814, 0.4307, 0.5192, 0.5254],
  27412. [0.6559, 0.4105, 0.8597, 0.4973, 0.4129, 0.5586, 0.7078, 0.5791],
  27413. [0.6423, 0.3960, 0.9124, 0.5206, 0.3707, 0.3737, 0.6229, 0.4838],
  27414. [0.6428, 0.3967, 0.9083, 0.4797, 0.4606, 0.5630, 0.6257, 0.5711]],
  27415. device='cuda:0', grad_fn=<AddmmBackward>)
  27416. landmarks are: tensor([[[0.6111, 0.4019, 0.8350, 0.3867, 0.3500, 0.4283, 0.5480, 0.5719],
  27417. [0.6145, 0.3983, 0.8700, 0.5017, 0.4400, 0.4800, 0.5375, 0.5750],
  27418. [0.6286, 0.4034, 0.8191, 0.2414, 0.4262, 0.2393, 0.6365, 0.5356],
  27419. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  27420. [0.6143, 0.3982, 0.8650, 0.4883, 0.3913, 0.4317, 0.5315, 0.5350],
  27421. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  27422. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  27423. [0.6226, 0.4125, 0.8800, 0.4900, 0.4512, 0.5600, 0.6275, 0.5517]]],
  27424. device='cuda:0')
  27425. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27426. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27427. loss_train: 0.051443345699226484
  27428. step: 79
  27429. running loss: 0.0006511815911294491
  27430. Train Steps: 79/90 Loss: 0.0007 torch.Size([8, 600, 800])
  27431. torch.Size([8, 8])
  27432. tensor([[0.6175, 0.4093, 0.8800, 0.4433, 0.4075, 0.4367, 0.5128, 0.4901],
  27433. [0.6229, 0.4066, 0.8513, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  27434. [0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533],
  27435. [0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  27436. [0.6197, 0.4091, 0.8800, 0.4783, 0.3538, 0.4767, 0.5950, 0.5550],
  27437. [0.6307, 0.4045, 0.8025, 0.5833, 0.3775, 0.4867, 0.6892, 0.5459],
  27438. [0.6311, 0.3998, 0.7975, 0.5767, 0.3838, 0.4850, 0.7327, 0.5343],
  27439. [0.6268, 0.4061, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300]],
  27440. device='cuda:0', dtype=torch.float64)
  27441. predictions are: tensor([[0.6421, 0.4114, 0.9054, 0.4475, 0.4030, 0.4379, 0.4815, 0.5277],
  27442. [0.6290, 0.3797, 0.8638, 0.5687, 0.4455, 0.4976, 0.5647, 0.5426],
  27443. [0.6579, 0.4121, 0.7245, 0.2258, 0.3609, 0.3025, 0.5843, 0.5721],
  27444. [0.6677, 0.4169, 0.8331, 0.2470, 0.4378, 0.1644, 0.6066, 0.5146],
  27445. [0.6490, 0.4068, 0.9010, 0.4690, 0.3478, 0.4634, 0.5632, 0.5723],
  27446. [0.6050, 0.3680, 0.8381, 0.5526, 0.3784, 0.4725, 0.6788, 0.5530],
  27447. [0.6591, 0.4004, 0.8229, 0.5532, 0.3631, 0.4770, 0.6958, 0.5461],
  27448. [0.5582, 0.3386, 0.8518, 0.2422, 0.4435, 0.2046, 0.6203, 0.5380]],
  27449. device='cuda:0', grad_fn=<AddmmBackward>)
  27450. landmarks are: tensor([[[0.6175, 0.4093, 0.8800, 0.4433, 0.4075, 0.4367, 0.5128, 0.4901],
  27451. [0.6229, 0.4066, 0.8512, 0.5717, 0.4500, 0.5050, 0.6025, 0.5350],
  27452. [0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533],
  27453. [0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  27454. [0.6197, 0.4091, 0.8800, 0.4783, 0.3537, 0.4767, 0.5950, 0.5550],
  27455. [0.6307, 0.4045, 0.8025, 0.5833, 0.3775, 0.4867, 0.6892, 0.5459],
  27456. [0.6311, 0.3998, 0.7975, 0.5767, 0.3837, 0.4850, 0.7327, 0.5343],
  27457. [0.6268, 0.4060, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300]]],
  27458. device='cuda:0')
  27459. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  27460. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  27461. loss_train: 0.052004264638526365
  27462. step: 80
  27463. running loss: 0.0006500533079815795
  27464.  
  27465. Train Steps: 80/90 Loss: 0.0007 torch.Size([8, 600, 800])
  27466. torch.Size([8, 8])
  27467. tensor([[0.6198, 0.4076, 0.7945, 0.2453, 0.4475, 0.1871, 0.5897, 0.5350],
  27468. [0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  27469. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  27470. [0.6185, 0.4098, 0.8838, 0.4900, 0.4537, 0.5800, 0.6288, 0.5400],
  27471. [0.6239, 0.4206, 0.8750, 0.5400, 0.3688, 0.4850, 0.5737, 0.5700],
  27472. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  27473. [0.6200, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  27474. [0.6199, 0.4093, 0.7913, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700]],
  27475. device='cuda:0', dtype=torch.float64)
  27476. predictions are: tensor([[0.5703, 0.3669, 0.7581, 0.2540, 0.4432, 0.1998, 0.5945, 0.5471],
  27477. [0.6378, 0.4003, 0.7889, 0.2524, 0.4271, 0.1770, 0.6109, 0.5023],
  27478. [0.5915, 0.3625, 0.7947, 0.5689, 0.4100, 0.5077, 0.5169, 0.5048],
  27479. [0.6458, 0.4033, 0.8496, 0.4939, 0.4427, 0.5757, 0.5995, 0.5437],
  27480. [0.6419, 0.4080, 0.8402, 0.5368, 0.3544, 0.4809, 0.5785, 0.5598],
  27481. [0.6207, 0.3765, 0.8509, 0.2847, 0.4755, 0.2091, 0.7005, 0.5387],
  27482. [0.6051, 0.3781, 0.7718, 0.2943, 0.3607, 0.2736, 0.5701, 0.5217],
  27483. [0.5617, 0.3639, 0.7565, 0.2403, 0.4141, 0.2534, 0.6069, 0.5797]],
  27484. device='cuda:0', grad_fn=<AddmmBackward>)
  27485. landmarks are: tensor([[[0.6198, 0.4076, 0.7945, 0.2453, 0.4475, 0.1871, 0.5897, 0.5350],
  27486. [0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  27487. [0.6073, 0.3932, 0.8363, 0.5817, 0.4425, 0.5117, 0.5204, 0.4817],
  27488. [0.6185, 0.4098, 0.8838, 0.4900, 0.4538, 0.5800, 0.6288, 0.5400],
  27489. [0.6239, 0.4206, 0.8750, 0.5400, 0.3688, 0.4850, 0.5738, 0.5700],
  27490. [0.6275, 0.4013, 0.8850, 0.2833, 0.4975, 0.2233, 0.7058, 0.5343],
  27491. [0.6201, 0.4055, 0.8011, 0.2988, 0.3842, 0.2798, 0.5949, 0.5183],
  27492. [0.6198, 0.4093, 0.7912, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700]]],
  27493. device='cuda:0')
  27494. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  27495. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  27496. loss_train: 0.052478661498753354
  27497. step: 81
  27498. running loss: 0.0006478847098611525
  27499. Train Steps: 81/90 Loss: 0.0006 torch.Size([8, 600, 800])
  27500. torch.Size([8, 8])
  27501. tensor([[0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  27502. [0.6164, 0.4119, 0.7913, 0.2650, 0.3538, 0.3500, 0.5614, 0.5038],
  27503. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  27504. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  27505. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5988, 0.5667],
  27506. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  27507. [0.6175, 0.4091, 0.7863, 0.2800, 0.3638, 0.3583, 0.6188, 0.5433],
  27508. [0.6175, 0.4093, 0.8800, 0.4433, 0.4075, 0.4367, 0.5128, 0.4901]],
  27509. device='cuda:0', dtype=torch.float64)
  27510. predictions are: tensor([[0.5892, 0.3835, 0.8204, 0.6024, 0.3746, 0.4807, 0.6545, 0.5273],
  27511. [0.6529, 0.4333, 0.7754, 0.2854, 0.3490, 0.3485, 0.5996, 0.5021],
  27512. [0.5644, 0.3742, 0.7539, 0.2913, 0.3867, 0.2633, 0.5590, 0.5586],
  27513. [0.6384, 0.4279, 0.8462, 0.5287, 0.4235, 0.5200, 0.5983, 0.5212],
  27514. [0.6365, 0.4300, 0.8416, 0.3824, 0.3755, 0.3318, 0.6057, 0.5722],
  27515. [0.6088, 0.3886, 0.8468, 0.4973, 0.4447, 0.4895, 0.6455, 0.5237],
  27516. [0.6316, 0.4297, 0.7667, 0.2996, 0.3584, 0.3493, 0.6122, 0.5377],
  27517. [0.6340, 0.4251, 0.8641, 0.4704, 0.4122, 0.4273, 0.5046, 0.5069]],
  27518. device='cuda:0', grad_fn=<AddmmBackward>)
  27519. landmarks are: tensor([[[0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297],
  27520. [0.6164, 0.4119, 0.7912, 0.2650, 0.3537, 0.3500, 0.5614, 0.5038],
  27521. [0.6101, 0.4042, 0.7775, 0.2617, 0.3713, 0.2817, 0.5440, 0.5650],
  27522. [0.6164, 0.3956, 0.8757, 0.5088, 0.4300, 0.5320, 0.6022, 0.5202],
  27523. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5987, 0.5667],
  27524. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  27525. [0.6175, 0.4091, 0.7862, 0.2800, 0.3638, 0.3583, 0.6187, 0.5433],
  27526. [0.6175, 0.4093, 0.8800, 0.4433, 0.4075, 0.4367, 0.5128, 0.4901]]],
  27527. device='cuda:0')
  27528. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  27529. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  27530. loss_train: 0.05284168719663285
  27531. step: 82
  27532. running loss: 0.0006444108194711323
  27533. Train Steps: 82/90 Loss: 0.0006 torch.Size([8, 600, 800])
  27534. torch.Size([8, 8])
  27535. tensor([[0.6175, 0.3957, 0.8700, 0.4817, 0.4662, 0.5133, 0.5800, 0.5517],
  27536. [0.6162, 0.4014, 0.8800, 0.5333, 0.3750, 0.4817, 0.5988, 0.5283],
  27537. [0.6149, 0.4054, 0.6713, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  27538. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795],
  27539. [0.6250, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6088, 0.5183],
  27540. [0.6203, 0.4073, 0.8189, 0.2398, 0.4400, 0.2054, 0.5929, 0.5501],
  27541. [0.6185, 0.4098, 0.8838, 0.4900, 0.4537, 0.5800, 0.6288, 0.5400],
  27542. [0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364]],
  27543. device='cuda:0', dtype=torch.float64)
  27544. predictions are: tensor([[0.6312, 0.4297, 0.8587, 0.4907, 0.4612, 0.5042, 0.5684, 0.5371],
  27545. [0.5988, 0.4089, 0.8584, 0.5445, 0.3812, 0.4816, 0.5975, 0.5158],
  27546. [0.5220, 0.3592, 0.6690, 0.2524, 0.3919, 0.1902, 0.5083, 0.5494],
  27547. [0.5786, 0.3829, 0.8285, 0.5459, 0.3944, 0.5311, 0.7094, 0.5628],
  27548. [0.5903, 0.3994, 0.8665, 0.5041, 0.4634, 0.5658, 0.6238, 0.5060],
  27549. [0.6346, 0.4473, 0.7965, 0.2709, 0.4465, 0.2053, 0.5946, 0.5359],
  27550. [0.6350, 0.4314, 0.8643, 0.5173, 0.4638, 0.5775, 0.6126, 0.5268],
  27551. [0.6126, 0.4190, 0.7661, 0.2302, 0.4490, 0.2217, 0.6696, 0.5299]],
  27552. device='cuda:0', grad_fn=<AddmmBackward>)
  27553. landmarks are: tensor([[[0.6175, 0.3957, 0.8700, 0.4817, 0.4663, 0.5133, 0.5800, 0.5517],
  27554. [0.6162, 0.4014, 0.8800, 0.5333, 0.3750, 0.4817, 0.5987, 0.5283],
  27555. [0.6149, 0.4054, 0.6712, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  27556. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795],
  27557. [0.6251, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6087, 0.5183],
  27558. [0.6203, 0.4073, 0.8189, 0.2398, 0.4400, 0.2054, 0.5929, 0.5501],
  27559. [0.6185, 0.4098, 0.8838, 0.4900, 0.4538, 0.5800, 0.6288, 0.5400],
  27560. [0.6282, 0.4092, 0.8000, 0.2183, 0.4500, 0.2383, 0.6787, 0.5364]]],
  27561. device='cuda:0')
  27562. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  27563. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  27564. loss_train: 0.05330092628719285
  27565. step: 83
  27566. running loss: 0.000642179834785456
  27567. Train Steps: 83/90 Loss: 0.0006 torch.Size([8, 600, 800])
  27568. torch.Size([8, 8])
  27569. tensor([[0.6218, 0.4137, 0.7263, 0.2233, 0.4075, 0.2650, 0.6212, 0.5783],
  27570. [0.6135, 0.3994, 0.7913, 0.3050, 0.3625, 0.3050, 0.5837, 0.5050],
  27571. [0.6339, 0.4123, 0.8638, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  27572. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5988, 0.5667],
  27573. [0.6057, 0.4011, 0.8750, 0.4267, 0.4400, 0.5800, 0.5845, 0.5585],
  27574. [0.6199, 0.4093, 0.7913, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700],
  27575. [0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6138, 0.5400],
  27576. [0.6310, 0.4017, 0.8563, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006]],
  27577. device='cuda:0', dtype=torch.float64)
  27578. predictions are: tensor([[0.5973, 0.3945, 0.7225, 0.2356, 0.4205, 0.2714, 0.6090, 0.5509],
  27579. [0.6181, 0.4025, 0.7789, 0.3133, 0.3741, 0.3125, 0.5720, 0.4842],
  27580. [0.6007, 0.4061, 0.8564, 0.5378, 0.4134, 0.5479, 0.7513, 0.5355],
  27581. [0.6377, 0.4411, 0.8541, 0.3740, 0.3877, 0.3519, 0.6080, 0.5572],
  27582. [0.6030, 0.4121, 0.8575, 0.4374, 0.4640, 0.5614, 0.5938, 0.5109],
  27583. [0.5986, 0.4167, 0.7761, 0.2509, 0.4511, 0.2576, 0.6155, 0.5611],
  27584. [0.6257, 0.4321, 0.8652, 0.4166, 0.3818, 0.4734, 0.6011, 0.5235],
  27585. [0.6366, 0.4268, 0.8353, 0.5934, 0.3804, 0.4744, 0.6389, 0.4864]],
  27586. device='cuda:0', grad_fn=<AddmmBackward>)
  27587. landmarks are: tensor([[[0.6218, 0.4137, 0.7262, 0.2233, 0.4075, 0.2650, 0.6212, 0.5783],
  27588. [0.6135, 0.3994, 0.7912, 0.3050, 0.3625, 0.3050, 0.5838, 0.5050],
  27589. [0.6339, 0.4123, 0.8637, 0.5367, 0.4075, 0.5467, 0.7517, 0.5436],
  27590. [0.6201, 0.4098, 0.8575, 0.3617, 0.3700, 0.3550, 0.5987, 0.5667],
  27591. [0.6057, 0.4011, 0.8750, 0.4267, 0.4400, 0.5800, 0.5845, 0.5585],
  27592. [0.6198, 0.4093, 0.7912, 0.2533, 0.4288, 0.2467, 0.5975, 0.5700],
  27593. [0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6137, 0.5400],
  27594. [0.6310, 0.4017, 0.8562, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006]]],
  27595. device='cuda:0')
  27596. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27597. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27598. loss_train: 0.05356978881172836
  27599. step: 84
  27600. running loss: 0.0006377355810920042
  27601.  
  27602. Train Steps: 84/90 Loss: 0.0006 torch.Size([8, 600, 800])
  27603. torch.Size([8, 8])
  27604. tensor([[0.6185, 0.4080, 0.8625, 0.3483, 0.3788, 0.2650, 0.5320, 0.5272],
  27605. [0.6178, 0.4059, 0.7525, 0.2250, 0.4313, 0.1783, 0.5404, 0.5240],
  27606. [0.6178, 0.4012, 0.8900, 0.4495, 0.3891, 0.5617, 0.5972, 0.4879],
  27607. [0.6263, 0.4039, 0.9000, 0.4400, 0.3625, 0.4667, 0.6424, 0.4804],
  27608. [0.6311, 0.3998, 0.7975, 0.5767, 0.3838, 0.4850, 0.7327, 0.5343],
  27609. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  27610. [0.6271, 0.4040, 0.9138, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  27611. [0.6277, 0.4036, 0.8688, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742]],
  27612. device='cuda:0', dtype=torch.float64)
  27613. predictions are: tensor([[0.5875, 0.4149, 0.8459, 0.3509, 0.3870, 0.2874, 0.5411, 0.5442],
  27614. [0.5924, 0.4037, 0.7505, 0.2112, 0.4286, 0.1813, 0.5500, 0.5437],
  27615. [0.5634, 0.4030, 0.8751, 0.4425, 0.4029, 0.5792, 0.6001, 0.5064],
  27616. [0.5981, 0.4082, 0.8929, 0.4612, 0.3577, 0.4834, 0.6200, 0.4828],
  27617. [0.6246, 0.4228, 0.7950, 0.5650, 0.3801, 0.5068, 0.7198, 0.5421],
  27618. [0.5945, 0.4179, 0.8718, 0.4739, 0.4445, 0.5150, 0.5740, 0.5743],
  27619. [0.6081, 0.4048, 0.9095, 0.3811, 0.4746, 0.2754, 0.7238, 0.5498],
  27620. [0.5983, 0.4079, 0.8501, 0.3613, 0.3965, 0.2732, 0.6135, 0.4916]],
  27621. device='cuda:0', grad_fn=<AddmmBackward>)
  27622. landmarks are: tensor([[[0.6186, 0.4080, 0.8625, 0.3483, 0.3787, 0.2650, 0.5320, 0.5272],
  27623. [0.6178, 0.4059, 0.7525, 0.2250, 0.4313, 0.1783, 0.5404, 0.5240],
  27624. [0.6178, 0.4012, 0.8900, 0.4495, 0.3891, 0.5617, 0.5972, 0.4879],
  27625. [0.6263, 0.4038, 0.9000, 0.4400, 0.3625, 0.4667, 0.6424, 0.4804],
  27626. [0.6311, 0.3998, 0.7975, 0.5767, 0.3837, 0.4850, 0.7327, 0.5343],
  27627. [0.6169, 0.4052, 0.8742, 0.4779, 0.4288, 0.4917, 0.5763, 0.5617],
  27628. [0.6271, 0.4040, 0.9137, 0.3750, 0.4625, 0.2617, 0.7232, 0.5413],
  27629. [0.6277, 0.4036, 0.8687, 0.3617, 0.3925, 0.2600, 0.6132, 0.4742]]],
  27630. device='cuda:0')
  27631. loss_train_step before backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  27632. loss_train_step after backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  27633. loss_train: 0.05381043939269148
  27634. step: 85
  27635. running loss: 0.0006330639928551939
  27636. Train Steps: 85/90 Loss: 0.0006 torch.Size([8, 600, 800])
  27637. torch.Size([8, 8])
  27638. tensor([[0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  27639. [ nan, nan, 0.7335, 0.2569, 0.3788, 0.2667, 0.5066, 0.5578],
  27640. [0.6134, 0.4090, 0.6926, 0.2819, 0.3538, 0.3233, 0.5563, 0.5667],
  27641. [0.6135, 0.4115, 0.8838, 0.4667, 0.4288, 0.6050, 0.5778, 0.5097],
  27642. [0.6241, 0.4143, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  27643. [ nan, nan, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  27644. [0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  27645. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100]],
  27646. device='cuda:0', dtype=torch.float64)
  27647. predictions are: tensor([[0.6852, 0.4618, 0.8708, 0.3944, 0.3783, 0.4155, 0.6161, 0.5445],
  27648. [0.0702, 0.0619, 0.7505, 0.2546, 0.3969, 0.2535, 0.5469, 0.5487],
  27649. [0.6904, 0.4733, 0.6993, 0.2870, 0.3562, 0.3299, 0.5718, 0.5590],
  27650. [0.7018, 0.4831, 0.8874, 0.4605, 0.4533, 0.6048, 0.5969, 0.4998],
  27651. [0.6675, 0.4485, 0.9028, 0.4753, 0.4376, 0.5313, 0.6724, 0.5408],
  27652. [0.0645, 0.0471, 0.7440, 0.2061, 0.4404, 0.2277, 0.5841, 0.5455],
  27653. [0.6645, 0.4337, 0.8776, 0.4873, 0.4567, 0.4745, 0.5866, 0.5565],
  27654. [0.6834, 0.4593, 0.8890, 0.4265, 0.3717, 0.3716, 0.6853, 0.5162]],
  27655. device='cuda:0', grad_fn=<AddmmBackward>)
  27656. landmarks are: tensor([[[0.6195, 0.4101, 0.8650, 0.3917, 0.3575, 0.4033, 0.5800, 0.5650],
  27657. [0.0000, 0.0000, 0.7335, 0.2569, 0.3787, 0.2667, 0.5066, 0.5578],
  27658. [0.6134, 0.4090, 0.6926, 0.2819, 0.3537, 0.3233, 0.5562, 0.5667],
  27659. [0.6135, 0.4115, 0.8838, 0.4667, 0.4288, 0.6050, 0.5778, 0.5097],
  27660. [0.6241, 0.4142, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  27661. [0.0000, 0.0000, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  27662. [0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  27663. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100]]],
  27664. device='cuda:0')
  27665. loss_train_step before backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  27666. loss_train_step after backward: tensor(0.0012, device='cuda:0', grad_fn=<MseLossBackward>)
  27667. loss_train: 0.055031666153809056
  27668. step: 86
  27669. running loss: 0.0006399030948117332
  27670. Train Steps: 86/90 Loss: 0.0006 torch.Size([8, 600, 800])
  27671. torch.Size([8, 8])
  27672. tensor([[0.6128, 0.4084, 0.8738, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  27673. [0.6346, 0.4092, 0.7712, 0.5917, 0.4037, 0.4767, 0.7343, 0.5725],
  27674. [0.6218, 0.4098, 0.7238, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  27675. [0.6157, 0.4102, 0.8513, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  27676. [0.6204, 0.4007, 0.7838, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  27677. [0.6248, 0.4185, 0.8500, 0.5767, 0.4463, 0.4550, 0.5613, 0.5917],
  27678. [0.6264, 0.4049, 0.8988, 0.4633, 0.3813, 0.4983, 0.6326, 0.4843],
  27679. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426]],
  27680. device='cuda:0', dtype=torch.float64)
  27681. predictions are: tensor([[0.6089, 0.4066, 0.8930, 0.4500, 0.3558, 0.3774, 0.5126, 0.5316],
  27682. [0.6178, 0.4133, 0.8086, 0.5496, 0.3828, 0.4906, 0.7146, 0.5772],
  27683. [0.5650, 0.3830, 0.7388, 0.1678, 0.4170, 0.2644, 0.6357, 0.5449],
  27684. [0.5711, 0.3814, 0.8569, 0.3553, 0.3582, 0.3664, 0.5219, 0.5682],
  27685. [0.6151, 0.3987, 0.7894, 0.2035, 0.4397, 0.1833, 0.6035, 0.5115],
  27686. [0.6286, 0.4139, 0.8668, 0.5642, 0.4487, 0.4700, 0.5509, 0.5796],
  27687. [0.5903, 0.3813, 0.9292, 0.4573, 0.3781, 0.5267, 0.6417, 0.4760],
  27688. [0.6194, 0.4071, 0.9080, 0.4125, 0.3576, 0.3366, 0.5864, 0.5319]],
  27689. device='cuda:0', grad_fn=<AddmmBackward>)
  27690. landmarks are: tensor([[[0.6127, 0.4084, 0.8737, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  27691. [0.6346, 0.4092, 0.7713, 0.5917, 0.4038, 0.4767, 0.7343, 0.5725],
  27692. [0.6218, 0.4098, 0.7237, 0.1850, 0.4238, 0.2517, 0.6288, 0.5350],
  27693. [0.6157, 0.4102, 0.8512, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  27694. [0.6204, 0.4007, 0.7837, 0.2100, 0.4475, 0.1733, 0.5825, 0.5167],
  27695. [0.6248, 0.4185, 0.8500, 0.5767, 0.4462, 0.4550, 0.5612, 0.5917],
  27696. [0.6264, 0.4049, 0.8988, 0.4633, 0.3812, 0.4983, 0.6326, 0.4843],
  27697. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426]]],
  27698. device='cuda:0')
  27699. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27700. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27701. loss_train: 0.05537027501850389
  27702. step: 87
  27703. running loss: 0.000636439942741424
  27704. Train Steps: 87/90 Loss: 0.0006 torch.Size([8, 600, 800])
  27705. torch.Size([8, 8])
  27706. tensor([[0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  27707. [0.6154, 0.4112, 0.7037, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  27708. [0.6289, 0.4024, 0.9088, 0.4567, 0.3937, 0.5633, 0.7058, 0.5609],
  27709. [0.6261, 0.4066, 0.8325, 0.2150, 0.4763, 0.2667, 0.7002, 0.5633],
  27710. [ nan, nan, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  27711. [0.6274, 0.4270, 0.8938, 0.4967, 0.3550, 0.4283, 0.5700, 0.5733],
  27712. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683],
  27713. [0.6272, 0.4045, 0.8538, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649]],
  27714. device='cuda:0', dtype=torch.float64)
  27715. predictions are: tensor([[ 0.6273, 0.3890, 0.8337, 0.2448, 0.4812, 0.1961, 0.6621, 0.5329],
  27716. [ 0.5398, 0.3412, 0.7003, 0.2040, 0.4109, 0.1969, 0.5295, 0.5867],
  27717. [ 0.6335, 0.3931, 0.9081, 0.4608, 0.3957, 0.5802, 0.7179, 0.5608],
  27718. [ 0.6215, 0.3847, 0.8332, 0.2021, 0.4697, 0.2656, 0.6955, 0.5724],
  27719. [ 0.0028, -0.0193, 0.7317, 0.1922, 0.4107, 0.2410, 0.5504, 0.5703],
  27720. [ 0.6439, 0.4225, 0.8855, 0.4874, 0.3613, 0.4436, 0.5802, 0.5788],
  27721. [ 0.6505, 0.4059, 0.8815, 0.5170, 0.3708, 0.3837, 0.5383, 0.5823],
  27722. [ 0.6387, 0.4065, 0.8629, 0.5829, 0.3632, 0.4534, 0.5999, 0.4838]],
  27723. device='cuda:0', grad_fn=<AddmmBackward>)
  27724. landmarks are: tensor([[[0.6350, 0.4118, 0.8363, 0.2717, 0.4850, 0.1700, 0.6441, 0.5170],
  27725. [0.6154, 0.4112, 0.7038, 0.2317, 0.4238, 0.1833, 0.5350, 0.5600],
  27726. [0.6289, 0.4024, 0.9087, 0.4567, 0.3938, 0.5633, 0.7058, 0.5609],
  27727. [0.6261, 0.4066, 0.8325, 0.2150, 0.4762, 0.2667, 0.7002, 0.5633],
  27728. [0.0000, 0.0000, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  27729. [0.6274, 0.4270, 0.8938, 0.4967, 0.3550, 0.4283, 0.5700, 0.5733],
  27730. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683],
  27731. [0.6271, 0.4045, 0.8537, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649]]],
  27732. device='cuda:0')
  27733. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27734. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27735. loss_train: 0.05571220855927095
  27736. step: 88
  27737. running loss: 0.0006330932790826245
  27738.  
  27739. Train Steps: 88/90 Loss: 0.0006 torch.Size([8, 600, 800])
  27740. torch.Size([8, 8])
  27741. tensor([[0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  27742. [0.6219, 0.4097, 0.8738, 0.3400, 0.3563, 0.4117, 0.5975, 0.5683],
  27743. [0.6339, 0.4102, 0.9088, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  27744. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  27745. [0.6111, 0.4033, 0.8300, 0.3267, 0.3588, 0.3333, 0.5444, 0.5637],
  27746. [0.6133, 0.4094, 0.8495, 0.4028, 0.3588, 0.3200, 0.5003, 0.5407],
  27747. [0.6125, 0.4076, 0.8488, 0.3883, 0.3700, 0.3683, 0.5026, 0.5505],
  27748. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250]],
  27749. device='cuda:0', dtype=torch.float64)
  27750. predictions are: tensor([[0.6336, 0.3862, 0.7997, 0.4040, 0.3665, 0.4314, 0.5084, 0.5298],
  27751. [0.6450, 0.4073, 0.8902, 0.3394, 0.3426, 0.4122, 0.5901, 0.5705],
  27752. [0.6138, 0.3758, 0.9187, 0.4917, 0.3973, 0.5376, 0.7533, 0.5515],
  27753. [0.6018, 0.3784, 0.7607, 0.2297, 0.4623, 0.1590, 0.5976, 0.5293],
  27754. [0.6231, 0.3946, 0.8374, 0.3192, 0.3491, 0.3294, 0.5278, 0.5643],
  27755. [0.5974, 0.3757, 0.8539, 0.3863, 0.3613, 0.3103, 0.4865, 0.5473],
  27756. [0.5614, 0.3451, 0.8446, 0.3562, 0.3489, 0.3618, 0.4804, 0.5477],
  27757. [0.6163, 0.3803, 0.9144, 0.4768, 0.4139, 0.5283, 0.6034, 0.5241]],
  27758. device='cuda:0', grad_fn=<AddmmBackward>)
  27759. landmarks are: tensor([[[0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  27760. [0.6219, 0.4097, 0.8737, 0.3400, 0.3562, 0.4117, 0.5975, 0.5683],
  27761. [0.6339, 0.4102, 0.9087, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  27762. [0.6229, 0.4086, 0.7538, 0.2600, 0.4775, 0.1617, 0.5900, 0.5383],
  27763. [0.6111, 0.4033, 0.8300, 0.3267, 0.3587, 0.3333, 0.5444, 0.5637],
  27764. [0.6133, 0.4094, 0.8495, 0.4028, 0.3587, 0.3200, 0.5003, 0.5407],
  27765. [0.6125, 0.4076, 0.8487, 0.3883, 0.3700, 0.3683, 0.5026, 0.5505],
  27766. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250]]],
  27767. device='cuda:0')
  27768. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27769. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27770. loss_train: 0.056046031415462494
  27771. step: 89
  27772. running loss: 0.0006297306900613763
  27773. Train Steps: 89/90 Loss: 0.0006 torch.Size([8, 600, 800])
  27774. torch.Size([8, 8])
  27775. tensor([[ nan, nan, 0.8850, 0.3000, 0.5363, 0.2250, 0.7343, 0.5771],
  27776. [ nan, nan, 0.7192, 0.2346, 0.4037, 0.2050, 0.5138, 0.5650],
  27777. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  27778. [0.6226, 0.4098, 0.8912, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  27779. [0.6239, 0.4061, 0.8850, 0.4600, 0.4225, 0.5200, 0.6138, 0.5450],
  27780. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  27781. [0.6263, 0.4030, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  27782. [0.6134, 0.4090, 0.6926, 0.2819, 0.3538, 0.3233, 0.5563, 0.5667]],
  27783. device='cuda:0', dtype=torch.float64)
  27784. predictions are: tensor([[0.0543, 0.0123, 0.8911, 0.3049, 0.5193, 0.2527, 0.7156, 0.5693],
  27785. [0.0740, 0.0303, 0.7232, 0.2378, 0.4109, 0.2229, 0.4969, 0.5576],
  27786. [0.6706, 0.4200, 0.8034, 0.2314, 0.4377, 0.2683, 0.6421, 0.5472],
  27787. [0.6664, 0.4229, 0.8999, 0.4158, 0.4024, 0.2327, 0.5504, 0.5444],
  27788. [0.6982, 0.4263, 0.9126, 0.4696, 0.4135, 0.5241, 0.6159, 0.5539],
  27789. [0.6612, 0.4072, 0.6893, 0.3068, 0.3739, 0.2782, 0.5454, 0.5953],
  27790. [0.6631, 0.4017, 0.9254, 0.4943, 0.3901, 0.5292, 0.6213, 0.4722],
  27791. [0.6907, 0.4428, 0.7078, 0.2828, 0.3454, 0.3258, 0.5317, 0.5711]],
  27792. device='cuda:0', grad_fn=<AddmmBackward>)
  27793. landmarks are: tensor([[[0.0000, 0.0000, 0.8850, 0.3000, 0.5362, 0.2250, 0.7343, 0.5771],
  27794. [0.0000, 0.0000, 0.7192, 0.2346, 0.4038, 0.2050, 0.5138, 0.5650],
  27795. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  27796. [0.6226, 0.4098, 0.8913, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  27797. [0.6239, 0.4061, 0.8850, 0.4600, 0.4225, 0.5200, 0.6137, 0.5450],
  27798. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  27799. [0.6263, 0.4029, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  27800. [0.6134, 0.4090, 0.6926, 0.2819, 0.3537, 0.3233, 0.5562, 0.5667]]],
  27801. device='cuda:0')
  27802. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  27803. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  27804. loss_train: 0.05668294767383486
  27805. step: 90
  27806. running loss: 0.0006298105297092762
  27807. Valid Steps: 10/10 Loss: nan 06
  27808. --------------------------------------------------
  27809. Epoch: 9 Train Loss: 0.0006 Valid Loss: nan
  27810. --------------------------------------------------
  27811. size of train loader is: 90
  27812. torch.Size([8, 600, 800])
  27813. torch.Size([8, 8])
  27814. tensor([[0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  27815. [0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  27816. [0.6200, 0.3978, 0.8900, 0.4550, 0.3775, 0.5200, 0.6150, 0.5367],
  27817. [0.6268, 0.4052, 0.8175, 0.2250, 0.4688, 0.1917, 0.6375, 0.5267],
  27818. [0.6085, 0.4005, 0.8400, 0.4317, 0.3763, 0.4750, 0.5476, 0.5058],
  27819. [0.6263, 0.4039, 0.9000, 0.4400, 0.3625, 0.4667, 0.6424, 0.4804],
  27820. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  27821. [0.6203, 0.4021, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405]],
  27822. device='cuda:0', dtype=torch.float64)
  27823. predictions are: tensor([[0.6319, 0.4033, 0.8382, 0.5868, 0.4032, 0.4758, 0.5705, 0.6225],
  27824. [0.6193, 0.4029, 0.8496, 0.5472, 0.3828, 0.3973, 0.6909, 0.5676],
  27825. [0.6050, 0.3735, 0.8999, 0.4544, 0.3672, 0.5176, 0.5923, 0.5468],
  27826. [0.5715, 0.3618, 0.8084, 0.2102, 0.4502, 0.1593, 0.6312, 0.5402],
  27827. [0.6272, 0.3849, 0.8458, 0.4138, 0.3584, 0.4622, 0.5293, 0.5167],
  27828. [0.6106, 0.3786, 0.9130, 0.4429, 0.3443, 0.4480, 0.6162, 0.4848],
  27829. [0.6164, 0.3759, 0.8053, 0.2452, 0.4003, 0.2702, 0.5980, 0.5262],
  27830. [0.6215, 0.3952, 0.8878, 0.5006, 0.3623, 0.3742, 0.5542, 0.5449]],
  27831. device='cuda:0', grad_fn=<AddmmBackward>)
  27832. landmarks are: tensor([[[0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117],
  27833. [0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  27834. [0.6199, 0.3978, 0.8900, 0.4550, 0.3775, 0.5200, 0.6150, 0.5367],
  27835. [0.6268, 0.4052, 0.8175, 0.2250, 0.4688, 0.1917, 0.6375, 0.5267],
  27836. [0.6084, 0.4005, 0.8400, 0.4317, 0.3762, 0.4750, 0.5476, 0.5058],
  27837. [0.6263, 0.4038, 0.9000, 0.4400, 0.3625, 0.4667, 0.6424, 0.4804],
  27838. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  27839. [0.6203, 0.4020, 0.8780, 0.5031, 0.3667, 0.3882, 0.5842, 0.5405]]],
  27840. device='cuda:0')
  27841. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27842. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  27843. loss_train: 0.00033010990591719747
  27844. step: 1
  27845. running loss: 0.00033010990591719747
  27846. Train Steps: 1/90 Loss: 0.0003 torch.Size([8, 600, 800])
  27847. torch.Size([8, 8])
  27848. tensor([[ nan, nan, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  27849. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  27850. [0.6038, 0.3946, 0.8413, 0.4883, 0.3563, 0.4550, 0.5266, 0.4693],
  27851. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  27852. [0.6264, 0.4049, 0.8988, 0.4633, 0.3813, 0.4983, 0.6326, 0.4843],
  27853. [0.6112, 0.4029, 0.8638, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  27854. [0.6278, 0.4253, 0.8875, 0.5017, 0.4113, 0.2750, 0.5413, 0.6196],
  27855. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646]],
  27856. device='cuda:0', dtype=torch.float64)
  27857. predictions are: tensor([[-0.0342, -0.0272, 0.8486, 0.2634, 0.4950, 0.2271, 0.7346, 0.5626],
  27858. [ 0.6159, 0.4097, 0.8308, 0.5721, 0.3762, 0.4762, 0.6822, 0.5432],
  27859. [ 0.5889, 0.3787, 0.8285, 0.4872, 0.3594, 0.4427, 0.5375, 0.4913],
  27860. [ 0.6269, 0.4067, 0.8038, 0.3009, 0.3414, 0.3860, 0.6229, 0.5287],
  27861. [ 0.6134, 0.3922, 0.8882, 0.4636, 0.3768, 0.4819, 0.6375, 0.4844],
  27862. [ 0.6016, 0.3828, 0.8553, 0.4822, 0.4793, 0.4714, 0.5879, 0.5612],
  27863. [ 0.6456, 0.4502, 0.8314, 0.4961, 0.4227, 0.2749, 0.5583, 0.5983],
  27864. [ 0.5594, 0.3836, 0.7677, 0.3099, 0.3410, 0.2879, 0.4798, 0.5510]],
  27865. device='cuda:0', grad_fn=<AddmmBackward>)
  27866. landmarks are: tensor([[[0.0000, 0.0000, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  27867. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  27868. [0.6038, 0.3946, 0.8413, 0.4883, 0.3562, 0.4550, 0.5266, 0.4693],
  27869. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  27870. [0.6264, 0.4049, 0.8988, 0.4633, 0.3812, 0.4983, 0.6326, 0.4843],
  27871. [0.6112, 0.4029, 0.8637, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  27872. [0.6278, 0.4253, 0.8875, 0.5017, 0.4112, 0.2750, 0.5413, 0.6196],
  27873. [0.6136, 0.4060, 0.8025, 0.3217, 0.3650, 0.3000, 0.5060, 0.5646]]],
  27874. device='cuda:0')
  27875. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  27876. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  27877. loss_train: 0.0007069317507557571
  27878. step: 2
  27879. running loss: 0.00035346587537787855
  27880.  
  27881. Train Steps: 2/90 Loss: 0.0004 torch.Size([8, 600, 800])
  27882. torch.Size([8, 8])
  27883. tensor([[0.6108, 0.4011, 0.8037, 0.3400, 0.3700, 0.2933, 0.5658, 0.5617],
  27884. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  27885. [0.6128, 0.4022, 0.8738, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  27886. [0.6164, 0.4076, 0.8838, 0.4117, 0.3713, 0.5550, 0.6238, 0.5350],
  27887. [0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  27888. [0.6060, 0.3924, 0.8450, 0.5717, 0.4200, 0.5217, 0.5253, 0.4752],
  27889. [0.6107, 0.4013, 0.8700, 0.4650, 0.5049, 0.5176, 0.5850, 0.5550],
  27890. [0.6104, 0.4029, 0.8738, 0.4900, 0.4088, 0.4533, 0.5070, 0.5510]],
  27891. device='cuda:0', dtype=torch.float64)
  27892. predictions are: tensor([[0.5574, 0.3672, 0.7911, 0.3417, 0.3739, 0.2880, 0.5773, 0.5484],
  27893. [0.5857, 0.4212, 0.7881, 0.5672, 0.3776, 0.4325, 0.5911, 0.6166],
  27894. [0.5877, 0.4114, 0.8576, 0.5004, 0.4741, 0.4730, 0.5572, 0.5137],
  27895. [0.5858, 0.3907, 0.8654, 0.4103, 0.3588, 0.5215, 0.6651, 0.5400],
  27896. [0.6192, 0.4234, 0.7952, 0.2539, 0.4438, 0.1704, 0.6434, 0.4868],
  27897. [0.5735, 0.3856, 0.8259, 0.5744, 0.4119, 0.4878, 0.5524, 0.4930],
  27898. [0.6191, 0.4261, 0.8729, 0.4594, 0.4652, 0.4792, 0.6136, 0.5484],
  27899. [0.6088, 0.4250, 0.8484, 0.4860, 0.4031, 0.4150, 0.5401, 0.5447]],
  27900. device='cuda:0', grad_fn=<AddmmBackward>)
  27901. landmarks are: tensor([[[0.6108, 0.4011, 0.8037, 0.3400, 0.3700, 0.2933, 0.5658, 0.5617],
  27902. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  27903. [0.6128, 0.4022, 0.8737, 0.5067, 0.4983, 0.5231, 0.5364, 0.5064],
  27904. [0.6164, 0.4076, 0.8838, 0.4117, 0.3713, 0.5550, 0.6237, 0.5350],
  27905. [0.6260, 0.4106, 0.8025, 0.2583, 0.4550, 0.1867, 0.6281, 0.4869],
  27906. [0.6060, 0.3924, 0.8450, 0.5717, 0.4200, 0.5217, 0.5253, 0.4752],
  27907. [0.6107, 0.4013, 0.8700, 0.4650, 0.5049, 0.5176, 0.5850, 0.5550],
  27908. [0.6104, 0.4029, 0.8737, 0.4900, 0.4087, 0.4533, 0.5070, 0.5510]]],
  27909. device='cuda:0')
  27910. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  27911. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  27912. loss_train: 0.0011995541281066835
  27913. step: 3
  27914. running loss: 0.00039985137603556115
  27915. Train Steps: 3/90 Loss: 0.0004 torch.Size([8, 600, 800])
  27916. torch.Size([8, 8])
  27917. tensor([[0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116],
  27918. [0.6193, 0.4079, 0.7288, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  27919. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  27920. [0.6264, 0.3972, 0.8853, 0.4771, 0.3853, 0.4511, 0.6293, 0.5334],
  27921. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395],
  27922. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  27923. [0.6314, 0.4107, 0.8750, 0.5100, 0.3788, 0.4900, 0.7121, 0.5864],
  27924. [0.6083, 0.3957, 0.8638, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980]],
  27925. device='cuda:0', dtype=torch.float64)
  27926. predictions are: tensor([[0.5633, 0.3993, 0.8716, 0.4412, 0.3910, 0.5889, 0.5730, 0.5090],
  27927. [0.6544, 0.4508, 0.7246, 0.2789, 0.4351, 0.2671, 0.5843, 0.6092],
  27928. [0.5153, 0.3684, 0.8457, 0.2617, 0.5448, 0.2269, 0.7136, 0.5446],
  27929. [0.5525, 0.3710, 0.9015, 0.4977, 0.3737, 0.4679, 0.6343, 0.5163],
  27930. [0.6168, 0.4171, 0.7332, 0.2345, 0.4105, 0.2477, 0.5846, 0.5238],
  27931. [0.6235, 0.4353, 0.6781, 0.2241, 0.4157, 0.1958, 0.5455, 0.5421],
  27932. [0.5919, 0.4037, 0.8823, 0.5418, 0.3836, 0.4859, 0.7154, 0.5731],
  27933. [0.5329, 0.3702, 0.8681, 0.5204, 0.4421, 0.5144, 0.5354, 0.4807]],
  27934. device='cuda:0', grad_fn=<AddmmBackward>)
  27935. landmarks are: tensor([[[0.6113, 0.4104, 0.8650, 0.4250, 0.3925, 0.5967, 0.5787, 0.5116],
  27936. [0.6193, 0.4078, 0.7287, 0.2500, 0.4250, 0.2550, 0.5989, 0.6266],
  27937. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  27938. [0.6264, 0.3972, 0.8853, 0.4771, 0.3853, 0.4511, 0.6293, 0.5334],
  27939. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395],
  27940. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  27941. [0.6314, 0.4107, 0.8750, 0.5100, 0.3787, 0.4900, 0.7121, 0.5864],
  27942. [0.6083, 0.3957, 0.8637, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980]]],
  27943. device='cuda:0')
  27944. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  27945. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  27946. loss_train: 0.001964781025890261
  27947. step: 4
  27948. running loss: 0.0004911952564725652
  27949. Train Steps: 4/90 Loss: 0.0005 torch.Size([8, 600, 800])
  27950. torch.Size([8, 8])
  27951. tensor([[0.6203, 0.4096, 0.8862, 0.4267, 0.3538, 0.4117, 0.6025, 0.5650],
  27952. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  27953. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167],
  27954. [0.6213, 0.4001, 0.7712, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183],
  27955. [0.6141, 0.4038, 0.8650, 0.4833, 0.4839, 0.5176, 0.5787, 0.5600],
  27956. [0.6262, 0.4163, 0.8850, 0.5183, 0.3763, 0.4150, 0.6025, 0.5500],
  27957. [0.6102, 0.4001, 0.7738, 0.3583, 0.3463, 0.3800, 0.5524, 0.5689],
  27958. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334]],
  27959. device='cuda:0', dtype=torch.float64)
  27960. predictions are: tensor([[0.6021, 0.4175, 0.8881, 0.4239, 0.3671, 0.4225, 0.6052, 0.5651],
  27961. [0.6143, 0.4123, 0.8668, 0.5317, 0.4316, 0.5661, 0.6044, 0.5064],
  27962. [0.5993, 0.4155, 0.8203, 0.3317, 0.3824, 0.3246, 0.5612, 0.5325],
  27963. [0.6159, 0.4190, 0.7533, 0.2214, 0.4518, 0.1878, 0.5800, 0.5094],
  27964. [0.6217, 0.4305, 0.8700, 0.4981, 0.4964, 0.5245, 0.5738, 0.5462],
  27965. [0.6181, 0.4288, 0.8747, 0.5019, 0.3862, 0.4134, 0.5927, 0.5407],
  27966. [0.5921, 0.3948, 0.7740, 0.3583, 0.3586, 0.3819, 0.5415, 0.5482],
  27967. [0.6055, 0.4110, 0.7385, 0.2399, 0.3841, 0.3300, 0.6114, 0.5342]],
  27968. device='cuda:0', grad_fn=<AddmmBackward>)
  27969. landmarks are: tensor([[[0.6203, 0.4096, 0.8863, 0.4267, 0.3537, 0.4117, 0.6025, 0.5650],
  27970. [0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  27971. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167],
  27972. [0.6213, 0.4001, 0.7713, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183],
  27973. [0.6141, 0.4038, 0.8650, 0.4833, 0.4839, 0.5176, 0.5788, 0.5600],
  27974. [0.6262, 0.4163, 0.8850, 0.5183, 0.3762, 0.4150, 0.6025, 0.5500],
  27975. [0.6102, 0.4001, 0.7738, 0.3583, 0.3462, 0.3800, 0.5524, 0.5689],
  27976. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334]]],
  27977. device='cuda:0')
  27978. loss_train_step before backward: tensor(0.0001, device='cuda:0', grad_fn=<MseLossBackward>)
  27979. loss_train_step after backward: tensor(0.0001, device='cuda:0', grad_fn=<MseLossBackward>)
  27980. loss_train: 0.0021081157465232536
  27981. step: 5
  27982. running loss: 0.0004216231493046507
  27983. Train Steps: 5/90 Loss: 0.0004 torch.Size([8, 600, 800])
  27984. torch.Size([8, 8])
  27985. tensor([[0.6206, 0.4001, 0.8900, 0.3933, 0.3588, 0.3567, 0.5837, 0.5083],
  27986. [ nan, nan, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  27987. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  27988. [0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  27989. [0.6152, 0.4131, 0.6863, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  27990. [0.6124, 0.4075, 0.7696, 0.4153, 0.3475, 0.3767, 0.5157, 0.5427],
  27991. [0.6161, 0.4099, 0.8738, 0.4383, 0.3788, 0.5483, 0.5605, 0.5019],
  27992. [0.6332, 0.4165, 0.9100, 0.3350, 0.4188, 0.3683, 0.7438, 0.5528]],
  27993. device='cuda:0', dtype=torch.float64)
  27994. predictions are: tensor([[0.6132, 0.4043, 0.9107, 0.4013, 0.3892, 0.3707, 0.5962, 0.5043],
  27995. [0.0139, 0.0169, 0.7506, 0.2564, 0.4062, 0.2713, 0.5084, 0.5668],
  27996. [0.6754, 0.4624, 0.7043, 0.2228, 0.4741, 0.1872, 0.5517, 0.5211],
  27997. [0.6439, 0.4191, 0.8781, 0.5365, 0.4522, 0.5445, 0.6816, 0.5261],
  27998. [0.6293, 0.4285, 0.6871, 0.2469, 0.3909, 0.3598, 0.5914, 0.5490],
  27999. [0.6005, 0.4066, 0.7894, 0.4102, 0.3767, 0.3983, 0.5329, 0.5549],
  28000. [0.6179, 0.4132, 0.9037, 0.4317, 0.3969, 0.5640, 0.5869, 0.5120],
  28001. [0.6048, 0.4038, 0.9222, 0.3448, 0.4445, 0.3862, 0.7535, 0.5470]],
  28002. device='cuda:0', grad_fn=<AddmmBackward>)
  28003. landmarks are: tensor([[[0.6206, 0.4001, 0.8900, 0.3933, 0.3587, 0.3567, 0.5838, 0.5083],
  28004. [0.0000, 0.0000, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  28005. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  28006. [0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  28007. [0.6152, 0.4131, 0.6862, 0.2567, 0.3625, 0.3300, 0.5765, 0.5305],
  28008. [0.6124, 0.4075, 0.7696, 0.4153, 0.3475, 0.3767, 0.5157, 0.5427],
  28009. [0.6161, 0.4099, 0.8737, 0.4383, 0.3787, 0.5483, 0.5605, 0.5019],
  28010. [0.6332, 0.4165, 0.9100, 0.3350, 0.4187, 0.3683, 0.7438, 0.5528]]],
  28011. device='cuda:0')
  28012. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28013. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28014. loss_train: 0.002486242665327154
  28015. step: 6
  28016. running loss: 0.0004143737775545257
  28017.  
  28018. Train Steps: 6/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28019. torch.Size([8, 8])
  28020. tensor([[0.6095, 0.3970, 0.8688, 0.4767, 0.4860, 0.4879, 0.5191, 0.4940],
  28021. [0.6128, 0.4084, 0.8738, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  28022. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5363, 0.5550],
  28023. [0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  28024. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  28025. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  28026. [0.6268, 0.4052, 0.8175, 0.2250, 0.4688, 0.1917, 0.6375, 0.5267],
  28027. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131]],
  28028. device='cuda:0', dtype=torch.float64)
  28029. predictions are: tensor([[0.6171, 0.4000, 0.8989, 0.4624, 0.4872, 0.5078, 0.5293, 0.5201],
  28030. [0.6218, 0.4176, 0.8943, 0.4584, 0.3685, 0.3988, 0.5197, 0.5399],
  28031. [0.6205, 0.4179, 0.7013, 0.2469, 0.4153, 0.2526, 0.5713, 0.5631],
  28032. [0.6313, 0.3969, 0.8684, 0.5323, 0.4041, 0.4980, 0.6304, 0.5513],
  28033. [0.6323, 0.4256, 0.6733, 0.2832, 0.3674, 0.3227, 0.5471, 0.5779],
  28034. [0.6289, 0.3980, 0.8613, 0.5977, 0.3881, 0.5231, 0.6237, 0.4799],
  28035. [0.6523, 0.4320, 0.8302, 0.2181, 0.4760, 0.2077, 0.6532, 0.5335],
  28036. [0.6019, 0.3899, 0.8802, 0.3725, 0.3655, 0.4080, 0.6073, 0.5136]],
  28037. device='cuda:0', grad_fn=<AddmmBackward>)
  28038. landmarks are: tensor([[[0.6095, 0.3970, 0.8687, 0.4767, 0.4860, 0.4879, 0.5191, 0.4940],
  28039. [0.6127, 0.4084, 0.8737, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  28040. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5362, 0.5550],
  28041. [0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  28042. [0.6168, 0.4111, 0.6517, 0.2875, 0.3688, 0.2817, 0.5228, 0.5837],
  28043. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  28044. [0.6268, 0.4052, 0.8175, 0.2250, 0.4688, 0.1917, 0.6375, 0.5267],
  28045. [0.6201, 0.4036, 0.8596, 0.3850, 0.3492, 0.3785, 0.5978, 0.5131]]],
  28046. device='cuda:0')
  28047. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  28048. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  28049. loss_train: 0.0027903580776182935
  28050. step: 7
  28051. running loss: 0.0003986225825168991
  28052. Train Steps: 7/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28053. torch.Size([8, 8])
  28054. tensor([[0.6192, 0.4128, 0.8513, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  28055. [0.6110, 0.4047, 0.8700, 0.4483, 0.3713, 0.3967, 0.5088, 0.5517],
  28056. [0.6098, 0.3991, 0.8638, 0.4717, 0.4263, 0.4967, 0.5212, 0.5650],
  28057. [0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297],
  28058. [0.6226, 0.4103, 0.8575, 0.3450, 0.4388, 0.2067, 0.5787, 0.5383],
  28059. [0.6260, 0.4214, 0.8538, 0.5500, 0.3663, 0.3767, 0.5587, 0.5983],
  28060. [0.6259, 0.4133, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  28061. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373]],
  28062. device='cuda:0', dtype=torch.float64)
  28063. predictions are: tensor([[0.6655, 0.4275, 0.8717, 0.5391, 0.4057, 0.5384, 0.5896, 0.5467],
  28064. [0.5990, 0.3900, 0.8523, 0.4329, 0.3639, 0.4221, 0.4961, 0.5334],
  28065. [0.6059, 0.3806, 0.8612, 0.4520, 0.4181, 0.5171, 0.5301, 0.5463],
  28066. [0.6200, 0.3915, 0.8680, 0.5255, 0.3656, 0.4592, 0.6387, 0.5337],
  28067. [0.6142, 0.4028, 0.8463, 0.3289, 0.4391, 0.2495, 0.5600, 0.5419],
  28068. [0.6230, 0.4021, 0.8543, 0.5149, 0.3631, 0.4069, 0.5589, 0.5789],
  28069. [0.6494, 0.4203, 0.8158, 0.2036, 0.4982, 0.1942, 0.6319, 0.4887],
  28070. [0.6183, 0.3849, 0.8135, 0.5679, 0.3800, 0.4807, 0.5747, 0.5301]],
  28071. device='cuda:0', grad_fn=<AddmmBackward>)
  28072. landmarks are: tensor([[[0.6192, 0.4128, 0.8512, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  28073. [0.6110, 0.4047, 0.8700, 0.4483, 0.3713, 0.3967, 0.5088, 0.5517],
  28074. [0.6098, 0.3991, 0.8637, 0.4717, 0.4263, 0.4967, 0.5213, 0.5650],
  28075. [0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297],
  28076. [0.6226, 0.4103, 0.8575, 0.3450, 0.4387, 0.2067, 0.5788, 0.5383],
  28077. [0.6260, 0.4214, 0.8537, 0.5500, 0.3663, 0.3767, 0.5587, 0.5983],
  28078. [0.6259, 0.4132, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  28079. [0.6179, 0.3961, 0.8347, 0.6020, 0.3887, 0.4624, 0.5714, 0.5373]]],
  28080. device='cuda:0')
  28081. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  28082. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  28083. loss_train: 0.003120287772617303
  28084. step: 8
  28085. running loss: 0.00039003597157716285
  28086. Train Steps: 8/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28087. torch.Size([8, 8])
  28088. tensor([[0.6250, 0.4103, 0.8950, 0.4400, 0.3912, 0.5650, 0.6050, 0.5133],
  28089. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  28090. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  28091. [0.6083, 0.3957, 0.8638, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  28092. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  28093. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  28094. [0.6229, 0.4107, 0.8137, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  28095. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450]],
  28096. device='cuda:0', dtype=torch.float64)
  28097. predictions are: tensor([[0.6907, 0.4464, 0.8934, 0.4423, 0.3799, 0.5879, 0.5853, 0.5208],
  28098. [0.6341, 0.4092, 0.8970, 0.4907, 0.3774, 0.4600, 0.5645, 0.5343],
  28099. [0.6480, 0.3964, 0.8914, 0.4762, 0.3615, 0.3755, 0.6009, 0.5030],
  28100. [0.6196, 0.3916, 0.8647, 0.4987, 0.4383, 0.5247, 0.5082, 0.4973],
  28101. [0.6607, 0.4309, 0.7277, 0.1843, 0.4020, 0.2626, 0.5911, 0.5482],
  28102. [0.6550, 0.4048, 0.8476, 0.5480, 0.3676, 0.4864, 0.7078, 0.5719],
  28103. [0.6345, 0.4082, 0.8107, 0.2881, 0.4759, 0.1987, 0.5459, 0.5407],
  28104. [0.6141, 0.3883, 0.8778, 0.4642, 0.3770, 0.4266, 0.5691, 0.5419]],
  28105. device='cuda:0', grad_fn=<AddmmBackward>)
  28106. landmarks are: tensor([[[0.6250, 0.4103, 0.8950, 0.4400, 0.3913, 0.5650, 0.6050, 0.5133],
  28107. [0.6219, 0.4089, 0.8938, 0.4800, 0.3825, 0.4450, 0.5850, 0.5200],
  28108. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  28109. [0.6083, 0.3957, 0.8637, 0.4950, 0.4363, 0.5083, 0.5346, 0.4980],
  28110. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  28111. [0.6339, 0.4081, 0.8425, 0.5417, 0.3850, 0.4833, 0.7335, 0.5760],
  28112. [0.6229, 0.4107, 0.8138, 0.2883, 0.4750, 0.1717, 0.5813, 0.5400],
  28113. [0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450]]],
  28114. device='cuda:0')
  28115. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  28116. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  28117. loss_train: 0.0034082017227774486
  28118. step: 9
  28119. running loss: 0.0003786890803086054
  28120. Train Steps: 9/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28121. torch.Size([8, 8])
  28122. tensor([[ nan, nan, 0.8300, 0.3150, 0.3588, 0.3383, 0.5208, 0.5194],
  28123. [0.6111, 0.4019, 0.8350, 0.3867, 0.3500, 0.4283, 0.5480, 0.5719],
  28124. [0.6125, 0.4035, 0.7825, 0.3100, 0.3463, 0.4900, 0.5832, 0.5637],
  28125. [0.6263, 0.4039, 0.9000, 0.4400, 0.3625, 0.4667, 0.6424, 0.4804],
  28126. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  28127. [0.6173, 0.4114, 0.7325, 0.2500, 0.4213, 0.1917, 0.5338, 0.5700],
  28128. [0.6329, 0.4055, 0.9050, 0.4783, 0.3613, 0.3917, 0.6464, 0.5019],
  28129. [0.6081, 0.3950, 0.8538, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954]],
  28130. device='cuda:0', dtype=torch.float64)
  28131. predictions are: tensor([[ 0.0015, -0.0031, 0.7993, 0.3112, 0.3631, 0.3143, 0.5292, 0.5382],
  28132. [ 0.6202, 0.3914, 0.8323, 0.4142, 0.3660, 0.4016, 0.5376, 0.5640],
  28133. [ 0.6413, 0.4078, 0.7725, 0.3255, 0.3465, 0.4617, 0.6007, 0.5419],
  28134. [ 0.6826, 0.4191, 0.8940, 0.4635, 0.3621, 0.4411, 0.6330, 0.4861],
  28135. [ 0.6830, 0.4369, 0.7711, 0.2781, 0.4563, 0.1491, 0.5781, 0.5318],
  28136. [ 0.6227, 0.3945, 0.7264, 0.2345, 0.4269, 0.1709, 0.5418, 0.5762],
  28137. [ 0.6921, 0.4178, 0.8874, 0.4984, 0.3737, 0.3784, 0.6353, 0.5010],
  28138. [ 0.6811, 0.4307, 0.8601, 0.4722, 0.3757, 0.4662, 0.5190, 0.4957]],
  28139. device='cuda:0', grad_fn=<AddmmBackward>)
  28140. landmarks are: tensor([[[0.0000, 0.0000, 0.8300, 0.3150, 0.3587, 0.3383, 0.5208, 0.5194],
  28141. [0.6111, 0.4019, 0.8350, 0.3867, 0.3500, 0.4283, 0.5480, 0.5719],
  28142. [0.6125, 0.4035, 0.7825, 0.3100, 0.3462, 0.4900, 0.5832, 0.5637],
  28143. [0.6263, 0.4038, 0.9000, 0.4400, 0.3625, 0.4667, 0.6424, 0.4804],
  28144. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  28145. [0.6173, 0.4114, 0.7325, 0.2500, 0.4212, 0.1917, 0.5337, 0.5700],
  28146. [0.6329, 0.4055, 0.9050, 0.4783, 0.3613, 0.3917, 0.6464, 0.5019],
  28147. [0.6081, 0.3950, 0.8537, 0.4667, 0.3850, 0.4917, 0.5342, 0.4954]]],
  28148. device='cuda:0')
  28149. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  28150. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  28151. loss_train: 0.0039045417652232572
  28152. step: 10
  28153. running loss: 0.0003904541765223257
  28154.  
  28155. Train Steps: 10/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28156. torch.Size([8, 8])
  28157. tensor([[0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  28158. [0.6115, 0.4005, 0.8838, 0.3867, 0.3763, 0.4700, 0.5800, 0.5550],
  28159. [0.6296, 0.4076, 0.8400, 0.5583, 0.3700, 0.4367, 0.6876, 0.5494],
  28160. [0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921],
  28161. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  28162. [0.6261, 0.4066, 0.8325, 0.2150, 0.4763, 0.2667, 0.7002, 0.5633],
  28163. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  28164. [0.6274, 0.4099, 0.8625, 0.3233, 0.4400, 0.1983, 0.5876, 0.4869]],
  28165. device='cuda:0', dtype=torch.float64)
  28166. predictions are: tensor([[0.6111, 0.3954, 0.8494, 0.5584, 0.3472, 0.3372, 0.5197, 0.5469],
  28167. [0.5739, 0.3684, 0.8690, 0.3896, 0.3697, 0.4466, 0.5364, 0.5605],
  28168. [0.6279, 0.3980, 0.8315, 0.5804, 0.3685, 0.3994, 0.6520, 0.5557],
  28169. [0.6394, 0.4043, 0.8134, 0.2342, 0.3990, 0.2662, 0.6132, 0.5075],
  28170. [0.5666, 0.3551, 0.8625, 0.4179, 0.3445, 0.4102, 0.5713, 0.5328],
  28171. [0.6321, 0.3944, 0.8275, 0.2332, 0.4824, 0.2432, 0.6758, 0.5559],
  28172. [0.6157, 0.3995, 0.7224, 0.1973, 0.4033, 0.2224, 0.5934, 0.5523],
  28173. [0.6081, 0.3950, 0.8581, 0.3306, 0.4382, 0.1855, 0.5523, 0.5056]],
  28174. device='cuda:0', grad_fn=<AddmmBackward>)
  28175. landmarks are: tensor([[[0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  28176. [0.6115, 0.4005, 0.8838, 0.3867, 0.3762, 0.4700, 0.5800, 0.5550],
  28177. [0.6296, 0.4076, 0.8400, 0.5583, 0.3700, 0.4367, 0.6876, 0.5494],
  28178. [0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921],
  28179. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  28180. [0.6261, 0.4066, 0.8325, 0.2150, 0.4762, 0.2667, 0.7002, 0.5633],
  28181. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  28182. [0.6274, 0.4099, 0.8625, 0.3233, 0.4400, 0.1983, 0.5876, 0.4869]]],
  28183. device='cuda:0')
  28184. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28185. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28186. loss_train: 0.0043209217401454225
  28187. step: 11
  28188. running loss: 0.0003928110672859475
  28189. Train Steps: 11/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28190. torch.Size([8, 8])
  28191. tensor([[0.6250, 0.4013, 0.8525, 0.5417, 0.4037, 0.5117, 0.6325, 0.5017],
  28192. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  28193. [0.6197, 0.4091, 0.8800, 0.4783, 0.3538, 0.4767, 0.5950, 0.5550],
  28194. [0.6254, 0.4076, 0.8700, 0.3267, 0.4150, 0.3083, 0.7050, 0.5609],
  28195. [0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533],
  28196. [0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  28197. [0.6100, 0.4016, 0.8600, 0.5067, 0.4612, 0.5233, 0.5086, 0.5519],
  28198. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385]],
  28199. device='cuda:0', dtype=torch.float64)
  28200. predictions are: tensor([[0.6466, 0.4067, 0.8561, 0.5692, 0.3907, 0.4857, 0.6159, 0.4922],
  28201. [0.5812, 0.3606, 0.8624, 0.5676, 0.3698, 0.4190, 0.5720, 0.5218],
  28202. [0.6026, 0.3910, 0.8887, 0.4900, 0.3473, 0.4561, 0.5905, 0.5322],
  28203. [0.6874, 0.4356, 0.8931, 0.3268, 0.3932, 0.2698, 0.6987, 0.5500],
  28204. [0.6209, 0.3994, 0.7293, 0.2280, 0.3702, 0.2973, 0.6075, 0.5366],
  28205. [0.6020, 0.3824, 0.9123, 0.4212, 0.3538, 0.4350, 0.6238, 0.4767],
  28206. [0.6339, 0.4077, 0.8634, 0.5291, 0.4464, 0.4912, 0.4958, 0.5456],
  28207. [0.6225, 0.3932, 0.8367, 0.2665, 0.4396, 0.2139, 0.6829, 0.5355]],
  28208. device='cuda:0', grad_fn=<AddmmBackward>)
  28209. landmarks are: tensor([[[0.6250, 0.4013, 0.8525, 0.5417, 0.4038, 0.5117, 0.6325, 0.5017],
  28210. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  28211. [0.6197, 0.4091, 0.8800, 0.4783, 0.3537, 0.4767, 0.5950, 0.5550],
  28212. [0.6254, 0.4076, 0.8700, 0.3267, 0.4150, 0.3083, 0.7050, 0.5609],
  28213. [0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533],
  28214. [0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  28215. [0.6100, 0.4016, 0.8600, 0.5067, 0.4613, 0.5233, 0.5086, 0.5519],
  28216. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385]]],
  28217. device='cuda:0')
  28218. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28219. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28220. loss_train: 0.004716772076790221
  28221. step: 12
  28222. running loss: 0.0003930643397325184
  28223. Train Steps: 12/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28224. torch.Size([8, 8])
  28225. tensor([[0.6175, 0.3957, 0.8700, 0.4817, 0.4662, 0.5133, 0.5800, 0.5517],
  28226. [0.6185, 0.4080, 0.8625, 0.3483, 0.3788, 0.2650, 0.5320, 0.5272],
  28227. [0.6273, 0.4110, 0.8900, 0.3817, 0.4188, 0.2167, 0.5858, 0.4835],
  28228. [0.6048, 0.3928, 0.8538, 0.5433, 0.3875, 0.5117, 0.5266, 0.4719],
  28229. [0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  28230. [0.6205, 0.4081, 0.8950, 0.4017, 0.3788, 0.4700, 0.5963, 0.5667],
  28231. [ nan, nan, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  28232. [0.6163, 0.4114, 0.7650, 0.2017, 0.3763, 0.2867, 0.5631, 0.5071]],
  28233. device='cuda:0', dtype=torch.float64)
  28234. predictions are: tensor([[0.6150, 0.3924, 0.8681, 0.4778, 0.4595, 0.5068, 0.6044, 0.5595],
  28235. [0.6532, 0.4270, 0.8590, 0.3559, 0.3754, 0.2603, 0.5552, 0.5468],
  28236. [0.6754, 0.4331, 0.9025, 0.3900, 0.4097, 0.2071, 0.6262, 0.5093],
  28237. [0.6148, 0.4029, 0.8485, 0.5439, 0.3948, 0.4812, 0.5710, 0.5134],
  28238. [0.6420, 0.4042, 0.8782, 0.5306, 0.3821, 0.4582, 0.6024, 0.5224],
  28239. [0.6530, 0.4177, 0.8900, 0.3976, 0.3744, 0.4545, 0.6184, 0.5700],
  28240. [0.0528, 0.0365, 0.6978, 0.2690, 0.3843, 0.2045, 0.5725, 0.5857],
  28241. [0.6206, 0.4146, 0.7602, 0.2088, 0.3685, 0.2788, 0.6094, 0.5118]],
  28242. device='cuda:0', grad_fn=<AddmmBackward>)
  28243. landmarks are: tensor([[[0.6175, 0.3957, 0.8700, 0.4817, 0.4663, 0.5133, 0.5800, 0.5517],
  28244. [0.6186, 0.4080, 0.8625, 0.3483, 0.3787, 0.2650, 0.5320, 0.5272],
  28245. [0.6273, 0.4110, 0.8900, 0.3817, 0.4187, 0.2167, 0.5858, 0.4835],
  28246. [0.6048, 0.3928, 0.8537, 0.5433, 0.3875, 0.5117, 0.5266, 0.4719],
  28247. [0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  28248. [0.6205, 0.4081, 0.8950, 0.4017, 0.3787, 0.4700, 0.5962, 0.5667],
  28249. [0.0000, 0.0000, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819],
  28250. [0.6163, 0.4114, 0.7650, 0.2017, 0.3762, 0.2867, 0.5631, 0.5071]]],
  28251. device='cuda:0')
  28252. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28253. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28254. loss_train: 0.00511384145647753
  28255. step: 13
  28256. running loss: 0.0003933724197290408
  28257. Train Steps: 13/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28258. torch.Size([8, 8])
  28259. tensor([[0.6286, 0.4078, 0.8063, 0.2267, 0.4788, 0.1533, 0.5953, 0.4913],
  28260. [0.6255, 0.4017, 0.8688, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  28261. [0.6043, 0.4022, 0.6887, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136],
  28262. [0.6199, 0.4102, 0.8950, 0.4417, 0.4012, 0.5367, 0.6112, 0.5967],
  28263. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200],
  28264. [ nan, nan, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  28265. [0.6193, 0.3930, 0.8949, 0.4437, 0.3852, 0.5435, 0.6263, 0.5263],
  28266. [0.6109, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117]],
  28267. device='cuda:0', dtype=torch.float64)
  28268. predictions are: tensor([[ 0.6139, 0.3975, 0.7991, 0.2557, 0.4762, 0.1688, 0.6188, 0.5154],
  28269. [ 0.6325, 0.4076, 0.8653, 0.3249, 0.3729, 0.3577, 0.6646, 0.5236],
  28270. [ 0.5999, 0.3938, 0.6965, 0.2217, 0.3897, 0.2614, 0.5847, 0.5240],
  28271. [ 0.6546, 0.4266, 0.9047, 0.4479, 0.4026, 0.5428, 0.6200, 0.5899],
  28272. [ 0.6073, 0.4046, 0.8692, 0.3171, 0.4470, 0.2431, 0.6499, 0.5239],
  28273. [-0.0289, -0.0041, 0.7068, 0.2008, 0.4259, 0.2260, 0.5732, 0.5807],
  28274. [ 0.5991, 0.3933, 0.8810, 0.4519, 0.3926, 0.5543, 0.6291, 0.5463],
  28275. [ 0.6007, 0.4034, 0.8795, 0.4790, 0.3675, 0.4355, 0.5861, 0.5176]],
  28276. device='cuda:0', grad_fn=<AddmmBackward>)
  28277. landmarks are: tensor([[[0.6286, 0.4078, 0.8062, 0.2267, 0.4787, 0.1533, 0.5953, 0.4913],
  28278. [0.6255, 0.4017, 0.8687, 0.3217, 0.3638, 0.3550, 0.6344, 0.4901],
  28279. [0.6043, 0.4022, 0.6888, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136],
  28280. [0.6199, 0.4102, 0.8950, 0.4417, 0.4013, 0.5367, 0.6112, 0.5967],
  28281. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200],
  28282. [0.0000, 0.0000, 0.6935, 0.1930, 0.4150, 0.2250, 0.5450, 0.5650],
  28283. [0.6193, 0.3930, 0.8949, 0.4437, 0.3852, 0.5435, 0.6263, 0.5263],
  28284. [0.6108, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117]]],
  28285. device='cuda:0')
  28286. loss_train_step before backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  28287. loss_train_step after backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  28288. loss_train: 0.005352006017346866
  28289. step: 14
  28290. running loss: 0.00038228614409620477
  28291.  
  28292. Train Steps: 14/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28293. torch.Size([8, 8])
  28294. tensor([[0.6268, 0.4102, 0.8938, 0.3667, 0.4025, 0.2833, 0.6275, 0.5183],
  28295. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  28296. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5787, 0.5550],
  28297. [0.6026, 0.3979, 0.8550, 0.4233, 0.3613, 0.5233, 0.5582, 0.4967],
  28298. [0.6329, 0.4196, 0.9238, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748],
  28299. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  28300. [0.6246, 0.4008, 0.8757, 0.5088, 0.4101, 0.5392, 0.6644, 0.5133],
  28301. [ nan, nan, 0.8938, 0.2850, 0.4662, 0.3117, 0.7406, 0.5528]],
  28302. device='cuda:0', dtype=torch.float64)
  28303. predictions are: tensor([[0.6304, 0.4434, 0.8698, 0.3517, 0.3991, 0.2967, 0.6378, 0.5092],
  28304. [0.6057, 0.4039, 0.8408, 0.5369, 0.3770, 0.4544, 0.5819, 0.5145],
  28305. [0.6102, 0.4243, 0.8122, 0.2548, 0.3479, 0.3743, 0.5690, 0.5382],
  28306. [0.6166, 0.4169, 0.8540, 0.4114, 0.3589, 0.5225, 0.5643, 0.5030],
  28307. [0.5879, 0.4055, 0.9096, 0.4521, 0.4069, 0.2987, 0.7183, 0.5450],
  28308. [0.5678, 0.3852, 0.8754, 0.3586, 0.4418, 0.2730, 0.6236, 0.5102],
  28309. [0.6362, 0.4257, 0.8712, 0.5049, 0.3944, 0.5445, 0.6412, 0.5012],
  28310. [0.0803, 0.0648, 0.8858, 0.2733, 0.4625, 0.3084, 0.7475, 0.5582]],
  28311. device='cuda:0', grad_fn=<AddmmBackward>)
  28312. landmarks are: tensor([[[0.6268, 0.4102, 0.8938, 0.3667, 0.4025, 0.2833, 0.6275, 0.5183],
  28313. [0.6202, 0.3983, 0.8555, 0.5429, 0.3842, 0.4370, 0.5866, 0.5398],
  28314. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5788, 0.5550],
  28315. [0.6026, 0.3979, 0.8550, 0.4233, 0.3613, 0.5233, 0.5582, 0.4967],
  28316. [0.6329, 0.4196, 0.9237, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748],
  28317. [0.6364, 0.4154, 0.8938, 0.3717, 0.4500, 0.2583, 0.6448, 0.5285],
  28318. [0.6246, 0.4008, 0.8757, 0.5088, 0.4101, 0.5392, 0.6644, 0.5133],
  28319. [0.0000, 0.0000, 0.8938, 0.2850, 0.4663, 0.3117, 0.7406, 0.5528]]],
  28320. device='cuda:0')
  28321. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  28322. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  28323. loss_train: 0.005812907431391068
  28324. step: 15
  28325. running loss: 0.00038752716209273786
  28326. Train Steps: 15/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28327. torch.Size([8, 8])
  28328. tensor([[ nan, nan, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733],
  28329. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  28330. [0.6233, 0.4091, 0.8100, 0.2950, 0.3563, 0.3883, 0.6013, 0.5200],
  28331. [0.6250, 0.3993, 0.9138, 0.4333, 0.3763, 0.5217, 0.6995, 0.5320],
  28332. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  28333. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  28334. [0.6251, 0.4163, 0.8662, 0.4467, 0.3625, 0.3567, 0.6038, 0.5533],
  28335. [0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274]],
  28336. device='cuda:0', dtype=torch.float64)
  28337. predictions are: tensor([[0.0263, 0.0256, 0.6932, 0.1956, 0.4421, 0.2121, 0.5592, 0.5668],
  28338. [0.5857, 0.3967, 0.7539, 0.1916, 0.4034, 0.2748, 0.6176, 0.5288],
  28339. [0.6044, 0.4116, 0.8246, 0.3053, 0.3672, 0.4163, 0.6139, 0.5282],
  28340. [0.6559, 0.4351, 0.9174, 0.4294, 0.3850, 0.5451, 0.6995, 0.5243],
  28341. [0.6037, 0.4098, 0.8734, 0.4678, 0.3612, 0.3758, 0.5414, 0.5704],
  28342. [0.5970, 0.4016, 0.8845, 0.4692, 0.4716, 0.5759, 0.6031, 0.5291],
  28343. [0.6008, 0.4041, 0.8873, 0.4476, 0.3542, 0.3747, 0.5981, 0.5399],
  28344. [0.6103, 0.4158, 0.8827, 0.3146, 0.4869, 0.2184, 0.6485, 0.5120]],
  28345. device='cuda:0', grad_fn=<AddmmBackward>)
  28346. landmarks are: tensor([[[0.0000, 0.0000, 0.6488, 0.1817, 0.4325, 0.1867, 0.5475, 0.5733],
  28347. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  28348. [0.6233, 0.4091, 0.8100, 0.2950, 0.3562, 0.3883, 0.6012, 0.5200],
  28349. [0.6250, 0.3993, 0.9137, 0.4333, 0.3762, 0.5217, 0.6995, 0.5320],
  28350. [0.6130, 0.4072, 0.8550, 0.4567, 0.3638, 0.3667, 0.5290, 0.5892],
  28351. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  28352. [0.6252, 0.4162, 0.8662, 0.4467, 0.3625, 0.3567, 0.6037, 0.5533],
  28353. [0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274]]],
  28354. device='cuda:0')
  28355. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  28356. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  28357. loss_train: 0.006098556055803783
  28358. step: 16
  28359. running loss: 0.00038115975348773645
  28360. Train Steps: 16/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28361. torch.Size([8, 8])
  28362. tensor([[0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  28363. [0.6075, 0.4000, 0.8513, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  28364. [0.6276, 0.4095, 0.8237, 0.2250, 0.4662, 0.1783, 0.6171, 0.4869],
  28365. [0.6143, 0.4040, 0.8237, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  28366. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5787, 0.5550],
  28367. [0.6115, 0.3998, 0.7063, 0.2383, 0.4037, 0.1950, 0.5320, 0.4993],
  28368. [0.6264, 0.4049, 0.8988, 0.4633, 0.3813, 0.4983, 0.6326, 0.4843],
  28369. [0.6240, 0.4217, 0.8150, 0.3133, 0.4425, 0.2650, 0.5650, 0.5817]],
  28370. device='cuda:0', dtype=torch.float64)
  28371. predictions are: tensor([[0.5833, 0.3842, 0.8821, 0.4733, 0.3704, 0.4449, 0.5607, 0.5791],
  28372. [0.5857, 0.3952, 0.8461, 0.5062, 0.4464, 0.5468, 0.5452, 0.5381],
  28373. [0.5606, 0.3757, 0.8263, 0.1889, 0.4795, 0.2077, 0.6426, 0.4731],
  28374. [0.5283, 0.3615, 0.8104, 0.3245, 0.4075, 0.2385, 0.5421, 0.5131],
  28375. [0.5866, 0.3970, 0.8141, 0.2517, 0.3591, 0.3805, 0.5958, 0.5653],
  28376. [0.5578, 0.3751, 0.7063, 0.2147, 0.4096, 0.2207, 0.5562, 0.5014],
  28377. [0.6209, 0.3995, 0.8981, 0.4447, 0.3939, 0.5215, 0.6618, 0.4925],
  28378. [0.5742, 0.4028, 0.8075, 0.3001, 0.4348, 0.2643, 0.5919, 0.5839]],
  28379. device='cuda:0', grad_fn=<AddmmBackward>)
  28380. landmarks are: tensor([[[0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  28381. [0.6075, 0.4000, 0.8512, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  28382. [0.6276, 0.4095, 0.8238, 0.2250, 0.4663, 0.1783, 0.6171, 0.4869],
  28383. [0.6143, 0.4040, 0.8238, 0.3333, 0.4075, 0.2117, 0.5137, 0.4973],
  28384. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5788, 0.5550],
  28385. [0.6115, 0.3998, 0.7063, 0.2383, 0.4038, 0.1950, 0.5320, 0.4993],
  28386. [0.6264, 0.4049, 0.8988, 0.4633, 0.3812, 0.4983, 0.6326, 0.4843],
  28387. [0.6240, 0.4217, 0.8150, 0.3133, 0.4425, 0.2650, 0.5650, 0.5817]]],
  28388. device='cuda:0')
  28389. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  28390. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  28391. loss_train: 0.00670615334820468
  28392. step: 17
  28393. running loss: 0.00039447960871792236
  28394. Train Steps: 17/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28395. torch.Size([8, 8])
  28396. tensor([[0.6075, 0.4007, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280],
  28397. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395],
  28398. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  28399. [0.6111, 0.3995, 0.8788, 0.4567, 0.3813, 0.4833, 0.5450, 0.5700],
  28400. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  28401. [0.6311, 0.3998, 0.7975, 0.5767, 0.3838, 0.4850, 0.7327, 0.5343],
  28402. [0.6289, 0.4024, 0.9088, 0.4567, 0.3937, 0.5633, 0.7058, 0.5609],
  28403. [0.6250, 0.4146, 0.8838, 0.3933, 0.3588, 0.4283, 0.6162, 0.5367]],
  28404. device='cuda:0', dtype=torch.float64)
  28405. predictions are: tensor([[0.6088, 0.4146, 0.8243, 0.4837, 0.4125, 0.5085, 0.4962, 0.5246],
  28406. [0.5454, 0.3526, 0.7500, 0.1850, 0.4105, 0.2336, 0.5899, 0.5240],
  28407. [0.6472, 0.4219, 0.8957, 0.2913, 0.4292, 0.4209, 0.6909, 0.5476],
  28408. [0.6005, 0.4060, 0.8789, 0.4457, 0.3934, 0.4989, 0.5180, 0.5573],
  28409. [0.6391, 0.4281, 0.9070, 0.4751, 0.3749, 0.4296, 0.6496, 0.5202],
  28410. [0.6069, 0.3982, 0.7994, 0.5490, 0.3786, 0.4836, 0.7198, 0.5246],
  28411. [0.5879, 0.3833, 0.8961, 0.4360, 0.4143, 0.5762, 0.6696, 0.5471],
  28412. [0.6208, 0.4172, 0.8873, 0.3753, 0.3730, 0.4114, 0.5894, 0.5254]],
  28413. device='cuda:0', grad_fn=<AddmmBackward>)
  28414. landmarks are: tensor([[[0.6075, 0.4006, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280],
  28415. [0.6186, 0.3967, 0.7337, 0.1992, 0.4120, 0.2508, 0.6105, 0.5395],
  28416. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  28417. [0.6111, 0.3995, 0.8788, 0.4567, 0.3812, 0.4833, 0.5450, 0.5700],
  28418. [0.6271, 0.4081, 0.9081, 0.4894, 0.3700, 0.4283, 0.6661, 0.5274],
  28419. [0.6311, 0.3998, 0.7975, 0.5767, 0.3837, 0.4850, 0.7327, 0.5343],
  28420. [0.6289, 0.4024, 0.9087, 0.4567, 0.3938, 0.5633, 0.7058, 0.5609],
  28421. [0.6250, 0.4146, 0.8838, 0.3933, 0.3587, 0.4283, 0.6162, 0.5367]]],
  28422. device='cuda:0')
  28423. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  28424. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  28425. loss_train: 0.007039182659354992
  28426. step: 18
  28427. running loss: 0.00039106570329749957
  28428.  
  28429. Train Steps: 18/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28430. torch.Size([8, 8])
  28431. tensor([[0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  28432. [0.6228, 0.4119, 0.7938, 0.2233, 0.4674, 0.1773, 0.6188, 0.5433],
  28433. [0.6198, 0.4101, 0.8838, 0.5283, 0.3763, 0.5267, 0.5913, 0.5567],
  28434. [0.6198, 0.3997, 0.8582, 0.5361, 0.4117, 0.5016, 0.5942, 0.5134],
  28435. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  28436. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  28437. [0.6357, 0.4118, 0.8400, 0.2500, 0.5413, 0.1633, 0.6725, 0.5586],
  28438. [ nan, nan, 0.7412, 0.2200, 0.4450, 0.1517, 0.5312, 0.4983]],
  28439. device='cuda:0', dtype=torch.float64)
  28440. predictions are: tensor([[ 0.5953, 0.3860, 0.8574, 0.4379, 0.4426, 0.2795, 0.5448, 0.6258],
  28441. [ 0.6346, 0.3960, 0.8043, 0.1938, 0.4684, 0.1988, 0.6204, 0.5387],
  28442. [ 0.6455, 0.4134, 0.8872, 0.5100, 0.3674, 0.5438, 0.5928, 0.5578],
  28443. [ 0.6155, 0.3911, 0.8685, 0.5148, 0.4183, 0.5489, 0.5997, 0.5189],
  28444. [ 0.5396, 0.3379, 0.7125, 0.1986, 0.4258, 0.1941, 0.5334, 0.5251],
  28445. [ 0.6262, 0.3989, 0.8644, 0.2575, 0.4861, 0.2410, 0.6567, 0.5455],
  28446. [ 0.6301, 0.3905, 0.8583, 0.2206, 0.5306, 0.2024, 0.6591, 0.5488],
  28447. [ 0.0209, -0.0096, 0.7140, 0.2022, 0.4335, 0.1930, 0.5282, 0.5127]],
  28448. device='cuda:0', grad_fn=<AddmmBackward>)
  28449. landmarks are: tensor([[[0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  28450. [0.6228, 0.4119, 0.7937, 0.2233, 0.4674, 0.1773, 0.6187, 0.5433],
  28451. [0.6198, 0.4101, 0.8838, 0.5283, 0.3762, 0.5267, 0.5913, 0.5567],
  28452. [0.6198, 0.3997, 0.8582, 0.5361, 0.4117, 0.5016, 0.5942, 0.5134],
  28453. [0.6154, 0.4048, 0.7100, 0.2067, 0.4338, 0.1667, 0.5413, 0.5220],
  28454. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  28455. [0.6357, 0.4118, 0.8400, 0.2500, 0.5412, 0.1633, 0.6725, 0.5586],
  28456. [0.0000, 0.0000, 0.7412, 0.2200, 0.4450, 0.1517, 0.5312, 0.4983]]],
  28457. device='cuda:0')
  28458. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  28459. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  28460. loss_train: 0.00753981365414802
  28461. step: 19
  28462. running loss: 0.00039683229758673787
  28463. Train Steps: 19/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28464. torch.Size([8, 8])
  28465. tensor([[0.6108, 0.4011, 0.8037, 0.3400, 0.3700, 0.2933, 0.5658, 0.5617],
  28466. [0.6200, 0.4059, 0.8700, 0.4900, 0.4163, 0.5000, 0.6162, 0.5467],
  28467. [0.6097, 0.3988, 0.8650, 0.5250, 0.4213, 0.5200, 0.5675, 0.5050],
  28468. [0.6236, 0.4084, 0.7738, 0.2133, 0.3663, 0.3233, 0.5813, 0.5567],
  28469. [0.6142, 0.4127, 0.7575, 0.3067, 0.3438, 0.4383, 0.5778, 0.5207],
  28470. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  28471. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  28472. [0.6109, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117]],
  28473. device='cuda:0', dtype=torch.float64)
  28474. predictions are: tensor([[0.6439, 0.4007, 0.8235, 0.3386, 0.3773, 0.2824, 0.5642, 0.5577],
  28475. [0.6291, 0.3961, 0.8844, 0.4833, 0.4212, 0.4785, 0.6076, 0.5477],
  28476. [0.6298, 0.3989, 0.8760, 0.5110, 0.4461, 0.4928, 0.5428, 0.5114],
  28477. [0.6189, 0.3918, 0.7877, 0.2282, 0.3809, 0.2835, 0.5803, 0.5645],
  28478. [0.6044, 0.4022, 0.7587, 0.2884, 0.3577, 0.4086, 0.5835, 0.5339],
  28479. [0.6266, 0.3994, 0.8935, 0.4803, 0.4208, 0.5055, 0.5916, 0.5261],
  28480. [0.6269, 0.3987, 0.7843, 0.2327, 0.4596, 0.1697, 0.6159, 0.5375],
  28481. [0.6304, 0.4008, 0.8788, 0.4662, 0.3646, 0.4020, 0.5589, 0.5101]],
  28482. device='cuda:0', grad_fn=<AddmmBackward>)
  28483. landmarks are: tensor([[[0.6108, 0.4011, 0.8037, 0.3400, 0.3700, 0.2933, 0.5658, 0.5617],
  28484. [0.6199, 0.4059, 0.8700, 0.4900, 0.4162, 0.5000, 0.6162, 0.5467],
  28485. [0.6097, 0.3988, 0.8650, 0.5250, 0.4212, 0.5200, 0.5675, 0.5050],
  28486. [0.6236, 0.4084, 0.7738, 0.2133, 0.3663, 0.3233, 0.5813, 0.5567],
  28487. [0.6142, 0.4127, 0.7575, 0.3067, 0.3438, 0.4383, 0.5778, 0.5207],
  28488. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  28489. [0.6214, 0.4116, 0.7750, 0.2317, 0.4487, 0.1883, 0.6200, 0.5400],
  28490. [0.6108, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117]]],
  28491. device='cuda:0')
  28492. loss_train_step before backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  28493. loss_train_step after backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  28494. loss_train: 0.007740936329355463
  28495. step: 20
  28496. running loss: 0.00038704681646777316
  28497. Train Steps: 20/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28498. torch.Size([8, 8])
  28499. tensor([[0.6170, 0.4102, 0.7468, 0.3695, 0.3463, 0.3767, 0.5238, 0.5823],
  28500. [0.6299, 0.4008, 0.8450, 0.5350, 0.4213, 0.5000, 0.6350, 0.5100],
  28501. [0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090],
  28502. [ nan, nan, 0.7625, 0.2433, 0.3713, 0.2867, 0.5235, 0.5220],
  28503. [0.6278, 0.4253, 0.8875, 0.5017, 0.4113, 0.2750, 0.5413, 0.6196],
  28504. [0.6198, 0.4130, 0.8762, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  28505. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  28506. [ nan, nan, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609]],
  28507. device='cuda:0', dtype=torch.float64)
  28508. predictions are: tensor([[ 0.6662, 0.4223, 0.7406, 0.3587, 0.3473, 0.3636, 0.5362, 0.5764],
  28509. [ 0.6988, 0.4409, 0.8367, 0.5339, 0.4200, 0.4791, 0.6176, 0.5213],
  28510. [ 0.6454, 0.4078, 0.8821, 0.3713, 0.4065, 0.2698, 0.6423, 0.5060],
  28511. [ 0.0193, -0.0011, 0.7563, 0.2482, 0.3874, 0.2615, 0.5023, 0.5327],
  28512. [ 0.6599, 0.4391, 0.8735, 0.4986, 0.4344, 0.2519, 0.5618, 0.6114],
  28513. [ 0.6724, 0.4471, 0.8753, 0.4173, 0.3714, 0.4785, 0.5753, 0.5135],
  28514. [ 0.6454, 0.4219, 0.8223, 0.3016, 0.3528, 0.3894, 0.6167, 0.5270],
  28515. [ 0.1428, 0.0750, 0.8914, 0.3256, 0.5149, 0.1905, 0.6792, 0.5704]],
  28516. device='cuda:0', grad_fn=<AddmmBackward>)
  28517. landmarks are: tensor([[[0.6170, 0.4102, 0.7468, 0.3695, 0.3462, 0.3767, 0.5238, 0.5823],
  28518. [0.6299, 0.4008, 0.8450, 0.5350, 0.4212, 0.5000, 0.6350, 0.5100],
  28519. [0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090],
  28520. [0.0000, 0.0000, 0.7625, 0.2433, 0.3713, 0.2867, 0.5235, 0.5220],
  28521. [0.6278, 0.4253, 0.8875, 0.5017, 0.4112, 0.2750, 0.5413, 0.6196],
  28522. [0.6198, 0.4130, 0.8763, 0.4117, 0.3650, 0.4900, 0.5707, 0.5103],
  28523. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  28524. [0.0000, 0.0000, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609]]],
  28525. device='cuda:0')
  28526. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  28527. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  28528. loss_train: 0.008517837704857811
  28529. step: 21
  28530. running loss: 0.0004056113192789434
  28531. Train Steps: 21/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28532. torch.Size([8, 8])
  28533. tensor([[0.6038, 0.3946, 0.8413, 0.4883, 0.3563, 0.4550, 0.5266, 0.4693],
  28534. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  28535. [0.6197, 0.4051, 0.7812, 0.2650, 0.3513, 0.4050, 0.6112, 0.5500],
  28536. [0.6272, 0.4045, 0.8538, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  28537. [0.6273, 0.4110, 0.8900, 0.3817, 0.4188, 0.2167, 0.5858, 0.4835],
  28538. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  28539. [0.6245, 0.4100, 0.7762, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  28540. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209]],
  28541. device='cuda:0', dtype=torch.float64)
  28542. predictions are: tensor([[0.5720, 0.3776, 0.8345, 0.4755, 0.3489, 0.4654, 0.5330, 0.5079],
  28543. [0.5836, 0.3819, 0.8806, 0.5104, 0.3676, 0.3222, 0.6156, 0.5192],
  28544. [0.6031, 0.3888, 0.7692, 0.2578, 0.3459, 0.3895, 0.6086, 0.5580],
  28545. [0.6142, 0.4031, 0.8418, 0.5832, 0.3648, 0.4312, 0.6062, 0.4846],
  28546. [0.6351, 0.4040, 0.8955, 0.3888, 0.4027, 0.2019, 0.5870, 0.4993],
  28547. [0.5938, 0.3881, 0.8417, 0.3796, 0.3461, 0.3908, 0.5894, 0.5815],
  28548. [0.5953, 0.3816, 0.7752, 0.2554, 0.4761, 0.1240, 0.5810, 0.5446],
  28549. [0.5647, 0.3757, 0.8465, 0.4529, 0.4370, 0.2333, 0.5333, 0.6345]],
  28550. device='cuda:0', grad_fn=<AddmmBackward>)
  28551. landmarks are: tensor([[[0.6038, 0.3946, 0.8413, 0.4883, 0.3562, 0.4550, 0.5266, 0.4693],
  28552. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  28553. [0.6197, 0.4051, 0.7812, 0.2650, 0.3512, 0.4050, 0.6112, 0.5500],
  28554. [0.6271, 0.4045, 0.8537, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  28555. [0.6273, 0.4110, 0.8900, 0.3817, 0.4187, 0.2167, 0.5858, 0.4835],
  28556. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  28557. [0.6245, 0.4100, 0.7763, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  28558. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209]]],
  28559. device='cuda:0')
  28560. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28561. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28562. loss_train: 0.008867989556165412
  28563. step: 22
  28564. running loss: 0.0004030904343711551
  28565.  
  28566. Train Steps: 22/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28567. torch.Size([8, 8])
  28568. tensor([[0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  28569. [0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  28570. [0.6200, 0.4112, 0.8862, 0.4100, 0.3638, 0.4917, 0.6088, 0.6050],
  28571. [0.6296, 0.4045, 0.9138, 0.4100, 0.4232, 0.4242, 0.7422, 0.5297],
  28572. [0.6273, 0.4110, 0.8900, 0.3817, 0.4188, 0.2167, 0.5858, 0.4835],
  28573. [0.6219, 0.4114, 0.8175, 0.2817, 0.3925, 0.2783, 0.5900, 0.5350],
  28574. [0.6277, 0.4118, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  28575. [0.6254, 0.4076, 0.8700, 0.3267, 0.4150, 0.3083, 0.7050, 0.5609]],
  28576. device='cuda:0', dtype=torch.float64)
  28577. predictions are: tensor([[0.6024, 0.3981, 0.8128, 0.2590, 0.4556, 0.2491, 0.6595, 0.5584],
  28578. [0.5549, 0.3711, 0.8749, 0.5209, 0.3631, 0.4099, 0.4960, 0.5748],
  28579. [0.5520, 0.3803, 0.8692, 0.4278, 0.3688, 0.4868, 0.5845, 0.5944],
  28580. [0.6277, 0.4208, 0.8766, 0.4481, 0.3989, 0.4240, 0.6891, 0.5475],
  28581. [0.6073, 0.4031, 0.8811, 0.4218, 0.4144, 0.2082, 0.5623, 0.4811],
  28582. [0.5931, 0.3978, 0.7811, 0.3136, 0.3934, 0.2683, 0.5637, 0.5376],
  28583. [0.6119, 0.4190, 0.8744, 0.4121, 0.3878, 0.2531, 0.6042, 0.5076],
  28584. [0.5991, 0.3918, 0.8556, 0.3521, 0.4053, 0.2951, 0.6678, 0.5565]],
  28585. device='cuda:0', grad_fn=<AddmmBackward>)
  28586. landmarks are: tensor([[[0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  28587. [0.6176, 0.4030, 0.8850, 0.4850, 0.3688, 0.4050, 0.5312, 0.5783],
  28588. [0.6200, 0.4112, 0.8863, 0.4100, 0.3638, 0.4917, 0.6087, 0.6050],
  28589. [0.6296, 0.4045, 0.9137, 0.4100, 0.4232, 0.4242, 0.7422, 0.5297],
  28590. [0.6273, 0.4110, 0.8900, 0.3817, 0.4187, 0.2167, 0.5858, 0.4835],
  28591. [0.6219, 0.4114, 0.8175, 0.2817, 0.3925, 0.2783, 0.5900, 0.5350],
  28592. [0.6277, 0.4117, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  28593. [0.6254, 0.4076, 0.8700, 0.3267, 0.4150, 0.3083, 0.7050, 0.5609]]],
  28594. device='cuda:0')
  28595. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  28596. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  28597. loss_train: 0.009468808741075918
  28598. step: 23
  28599. running loss: 0.00041168733656851816
  28600. Train Steps: 23/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28601. torch.Size([8, 8])
  28602. tensor([[0.6268, 0.4052, 0.8175, 0.2250, 0.4688, 0.1917, 0.6375, 0.5267],
  28603. [0.6127, 0.4066, 0.8550, 0.5567, 0.4662, 0.5141, 0.5070, 0.5412],
  28604. [0.6128, 0.4084, 0.8738, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  28605. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  28606. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5837, 0.5500],
  28607. [0.6087, 0.3951, 0.8387, 0.5833, 0.4188, 0.4933, 0.5146, 0.4830],
  28608. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  28609. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200]],
  28610. device='cuda:0', dtype=torch.float64)
  28611. predictions are: tensor([[0.6148, 0.4041, 0.8155, 0.2686, 0.4448, 0.1657, 0.6586, 0.5228],
  28612. [0.6045, 0.4073, 0.8558, 0.5662, 0.4406, 0.4801, 0.5450, 0.5344],
  28613. [0.6265, 0.4334, 0.8728, 0.4929, 0.3422, 0.3535, 0.5344, 0.5387],
  28614. [0.6185, 0.4159, 0.8433, 0.5764, 0.3751, 0.4169, 0.5593, 0.4895],
  28615. [0.5972, 0.3902, 0.8796, 0.4573, 0.4146, 0.5011, 0.6148, 0.5498],
  28616. [0.6242, 0.4118, 0.8383, 0.6012, 0.3855, 0.4754, 0.5471, 0.5078],
  28617. [0.6271, 0.4017, 0.8617, 0.5176, 0.3983, 0.4823, 0.6515, 0.5281],
  28618. [0.6338, 0.4183, 0.8594, 0.3378, 0.4281, 0.2242, 0.6443, 0.5128]],
  28619. device='cuda:0', grad_fn=<AddmmBackward>)
  28620. landmarks are: tensor([[[0.6268, 0.4052, 0.8175, 0.2250, 0.4688, 0.1917, 0.6375, 0.5267],
  28621. [0.6127, 0.4066, 0.8550, 0.5567, 0.4662, 0.5141, 0.5070, 0.5412],
  28622. [0.6127, 0.4084, 0.8737, 0.4683, 0.3613, 0.3700, 0.4960, 0.5397],
  28623. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  28624. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5838, 0.5500],
  28625. [0.6087, 0.3951, 0.8388, 0.5833, 0.4187, 0.4933, 0.5146, 0.4830],
  28626. [0.6250, 0.3961, 0.8672, 0.4929, 0.4199, 0.4972, 0.6312, 0.5301],
  28627. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200]]],
  28628. device='cuda:0')
  28629. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28630. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28631. loss_train: 0.00983630126575008
  28632. step: 24
  28633. running loss: 0.00040984588607292
  28634. Train Steps: 24/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28635. torch.Size([8, 8])
  28636. tensor([[0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167],
  28637. [0.6357, 0.4097, 0.9038, 0.3883, 0.4213, 0.2950, 0.6686, 0.5390],
  28638. [0.6299, 0.4303, 0.7963, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  28639. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  28640. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  28641. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  28642. [0.6100, 0.4071, 0.7601, 0.3444, 0.3400, 0.4117, 0.5625, 0.5617],
  28643. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383]],
  28644. device='cuda:0', dtype=torch.float64)
  28645. predictions are: tensor([[0.5478, 0.3742, 0.7595, 0.3074, 0.3959, 0.2678, 0.5664, 0.5896],
  28646. [0.5971, 0.4027, 0.8927, 0.3894, 0.4067, 0.2946, 0.6613, 0.5152],
  28647. [0.6388, 0.4374, 0.8103, 0.4226, 0.4771, 0.2424, 0.5317, 0.6015],
  28648. [0.6420, 0.4268, 0.8663, 0.2420, 0.5460, 0.2147, 0.7460, 0.5467],
  28649. [0.5977, 0.3995, 0.7534, 0.2410, 0.3814, 0.2896, 0.6142, 0.5233],
  28650. [0.6387, 0.4289, 0.8900, 0.4843, 0.3351, 0.3626, 0.6269, 0.4801],
  28651. [0.5766, 0.3897, 0.7768, 0.3512, 0.3272, 0.4412, 0.5664, 0.5261],
  28652. [0.5779, 0.3848, 0.8936, 0.4680, 0.4346, 0.6008, 0.6111, 0.5183]],
  28653. device='cuda:0', grad_fn=<AddmmBackward>)
  28654. landmarks are: tensor([[[0.6168, 0.4055, 0.7575, 0.2950, 0.4050, 0.2717, 0.5938, 0.6167],
  28655. [0.6357, 0.4097, 0.9038, 0.3883, 0.4212, 0.2950, 0.6686, 0.5390],
  28656. [0.6299, 0.4303, 0.7962, 0.3933, 0.4850, 0.2283, 0.5480, 0.6222],
  28657. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  28658. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  28659. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  28660. [0.6100, 0.4071, 0.7601, 0.3444, 0.3400, 0.4117, 0.5625, 0.5617],
  28661. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383]]],
  28662. device='cuda:0')
  28663. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28664. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28665. loss_train: 0.010191787354415283
  28666. step: 25
  28667. running loss: 0.00040767149417661133
  28668. Train Steps: 25/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28669. torch.Size([8, 8])
  28670. tensor([[0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  28671. [0.6277, 0.4118, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  28672. [0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  28673. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  28674. [0.6138, 0.4101, 0.8800, 0.5083, 0.4637, 0.5950, 0.5587, 0.5077],
  28675. [0.6157, 0.4102, 0.8513, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  28676. [0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  28677. [0.6138, 0.4054, 0.8750, 0.4750, 0.4363, 0.5017, 0.5086, 0.5822]],
  28678. device='cuda:0', dtype=torch.float64)
  28679. predictions are: tensor([[0.6597, 0.4380, 0.9125, 0.4130, 0.4238, 0.3909, 0.7084, 0.5772],
  28680. [0.6539, 0.4436, 0.8975, 0.3660, 0.3861, 0.2707, 0.6218, 0.4967],
  28681. [0.6279, 0.4259, 0.8352, 0.5556, 0.3844, 0.4193, 0.7031, 0.5480],
  28682. [0.6050, 0.4060, 0.8342, 0.5771, 0.4128, 0.4695, 0.5517, 0.5103],
  28683. [0.5663, 0.3743, 0.8819, 0.5145, 0.4564, 0.5949, 0.5561, 0.5101],
  28684. [0.6380, 0.4306, 0.8561, 0.3709, 0.3507, 0.3565, 0.5027, 0.5586],
  28685. [0.6158, 0.4169, 0.9179, 0.3553, 0.4339, 0.3942, 0.7244, 0.5386],
  28686. [0.6019, 0.4118, 0.8663, 0.4772, 0.4382, 0.4870, 0.5078, 0.5601]],
  28687. device='cuda:0', grad_fn=<AddmmBackward>)
  28688. landmarks are: tensor([[[0.6350, 0.4144, 0.9250, 0.4100, 0.4125, 0.3750, 0.7129, 0.5945],
  28689. [0.6277, 0.4117, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  28690. [0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  28691. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  28692. [0.6138, 0.4101, 0.8800, 0.5083, 0.4638, 0.5950, 0.5587, 0.5077],
  28693. [0.6157, 0.4102, 0.8512, 0.3817, 0.3613, 0.3667, 0.5096, 0.5890],
  28694. [0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  28695. [0.6138, 0.4054, 0.8750, 0.4750, 0.4363, 0.5017, 0.5086, 0.5822]]],
  28696. device='cuda:0')
  28697. loss_train_step before backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  28698. loss_train_step after backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  28699. loss_train: 0.010435777367092669
  28700. step: 26
  28701. running loss: 0.0004013760525804873
  28702.  
  28703. Train Steps: 26/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28704. torch.Size([8, 8])
  28705. tensor([[0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  28706. [0.6148, 0.3996, 0.8488, 0.3867, 0.3488, 0.4067, 0.5863, 0.5000],
  28707. [0.6307, 0.4045, 0.8025, 0.5833, 0.3775, 0.4867, 0.6892, 0.5459],
  28708. [0.6109, 0.4041, 0.6975, 0.3167, 0.3513, 0.3383, 0.5153, 0.5319],
  28709. [0.6164, 0.4119, 0.7913, 0.2650, 0.3538, 0.3500, 0.5614, 0.5038],
  28710. [0.6059, 0.4002, 0.7562, 0.2767, 0.3538, 0.3033, 0.5529, 0.5455],
  28711. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050],
  28712. [0.6226, 0.4098, 0.8912, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367]],
  28713. device='cuda:0', dtype=torch.float64)
  28714. predictions are: tensor([[0.6733, 0.4420, 0.9006, 0.4592, 0.3952, 0.4471, 0.6059, 0.5509],
  28715. [0.6782, 0.4406, 0.8676, 0.3867, 0.3818, 0.4237, 0.6056, 0.5240],
  28716. [0.6645, 0.4411, 0.8261, 0.5779, 0.4138, 0.5102, 0.7137, 0.5565],
  28717. [0.6322, 0.4215, 0.7219, 0.3003, 0.3706, 0.3627, 0.5216, 0.5281],
  28718. [0.6336, 0.4272, 0.8046, 0.2458, 0.3731, 0.3714, 0.5971, 0.5173],
  28719. [0.6348, 0.4317, 0.7630, 0.2643, 0.3831, 0.3224, 0.5698, 0.5473],
  28720. [0.6039, 0.4015, 0.9008, 0.5119, 0.4727, 0.5480, 0.5999, 0.5222],
  28721. [0.6702, 0.4475, 0.9149, 0.4061, 0.4231, 0.2565, 0.5952, 0.5360]],
  28722. device='cuda:0', grad_fn=<AddmmBackward>)
  28723. landmarks are: tensor([[[0.6201, 0.4029, 0.8776, 0.4602, 0.3683, 0.4203, 0.5938, 0.5450],
  28724. [0.6148, 0.3996, 0.8487, 0.3867, 0.3487, 0.4067, 0.5863, 0.5000],
  28725. [0.6307, 0.4045, 0.8025, 0.5833, 0.3775, 0.4867, 0.6892, 0.5459],
  28726. [0.6109, 0.4041, 0.6975, 0.3167, 0.3512, 0.3383, 0.5153, 0.5319],
  28727. [0.6164, 0.4119, 0.7912, 0.2650, 0.3537, 0.3500, 0.5614, 0.5038],
  28728. [0.6059, 0.4002, 0.7563, 0.2767, 0.3537, 0.3033, 0.5529, 0.5455],
  28729. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050],
  28730. [0.6226, 0.4098, 0.8913, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367]]],
  28731. device='cuda:0')
  28732. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  28733. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  28734. loss_train: 0.011048498621676117
  28735. step: 27
  28736. running loss: 0.000409203652654671
  28737. Train Steps: 27/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28738. torch.Size([8, 8])
  28739. tensor([[0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  28740. [0.6162, 0.4134, 0.6700, 0.2467, 0.3962, 0.2533, 0.5737, 0.5467],
  28741. [0.6219, 0.3934, 0.8688, 0.5267, 0.4313, 0.4967, 0.5988, 0.4983],
  28742. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  28743. [0.6048, 0.3928, 0.8538, 0.5433, 0.3875, 0.5117, 0.5266, 0.4719],
  28744. [0.6187, 0.4104, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  28745. [0.6149, 0.4054, 0.6713, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  28746. [0.6268, 0.4052, 0.8175, 0.2250, 0.4688, 0.1917, 0.6375, 0.5267]],
  28747. device='cuda:0', dtype=torch.float64)
  28748. predictions are: tensor([[0.6253, 0.3900, 0.8995, 0.5372, 0.4009, 0.5019, 0.6102, 0.5218],
  28749. [0.6573, 0.4399, 0.6953, 0.2518, 0.4098, 0.2716, 0.5794, 0.5534],
  28750. [0.6543, 0.4098, 0.8893, 0.5400, 0.4585, 0.5274, 0.6245, 0.5081],
  28751. [0.6368, 0.4064, 0.7912, 0.2724, 0.4639, 0.2089, 0.5825, 0.5365],
  28752. [0.6131, 0.3953, 0.8726, 0.5570, 0.4134, 0.5330, 0.5629, 0.5069],
  28753. [0.6109, 0.4030, 0.7297, 0.1998, 0.4061, 0.2826, 0.6012, 0.5700],
  28754. [0.6804, 0.4367, 0.6936, 0.2261, 0.4120, 0.2187, 0.5295, 0.5860],
  28755. [0.6497, 0.4117, 0.8310, 0.2360, 0.4834, 0.2101, 0.6582, 0.5428]],
  28756. device='cuda:0', grad_fn=<AddmmBackward>)
  28757. landmarks are: tensor([[[0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  28758. [0.6162, 0.4134, 0.6700, 0.2467, 0.3963, 0.2533, 0.5738, 0.5467],
  28759. [0.6219, 0.3934, 0.8687, 0.5267, 0.4313, 0.4967, 0.5987, 0.4983],
  28760. [0.6175, 0.4013, 0.7734, 0.2783, 0.4475, 0.1786, 0.5790, 0.5351],
  28761. [0.6048, 0.3928, 0.8537, 0.5433, 0.3875, 0.5117, 0.5266, 0.4719],
  28762. [0.6187, 0.4103, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  28763. [0.6149, 0.4054, 0.6712, 0.2333, 0.4025, 0.2017, 0.5213, 0.5695],
  28764. [0.6268, 0.4052, 0.8175, 0.2250, 0.4688, 0.1917, 0.6375, 0.5267]]],
  28765. device='cuda:0')
  28766. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28767. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28768. loss_train: 0.01147236383985728
  28769. step: 28
  28770. running loss: 0.0004097272799949029
  28771. Train Steps: 28/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28772. torch.Size([8, 8])
  28773. tensor([[0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  28774. [0.6200, 0.4070, 0.8938, 0.4183, 0.3538, 0.4567, 0.6175, 0.5400],
  28775. [0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090],
  28776. [0.6250, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6088, 0.5183],
  28777. [0.6175, 0.4093, 0.8800, 0.4433, 0.4075, 0.4367, 0.5128, 0.4901],
  28778. [0.6200, 0.3961, 0.8461, 0.5497, 0.4142, 0.4577, 0.5892, 0.5402],
  28779. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167],
  28780. [ nan, nan, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621]],
  28781. device='cuda:0', dtype=torch.float64)
  28782. predictions are: tensor([[0.6964, 0.4375, 0.8261, 0.2212, 0.4650, 0.2734, 0.7027, 0.5774],
  28783. [0.6946, 0.4328, 0.8898, 0.4038, 0.3624, 0.4662, 0.6211, 0.5375],
  28784. [0.6542, 0.4034, 0.8801, 0.3598, 0.4098, 0.3013, 0.6400, 0.5122],
  28785. [0.6437, 0.4128, 0.8703, 0.4686, 0.4730, 0.5749, 0.6123, 0.5324],
  28786. [0.6431, 0.4101, 0.8832, 0.4444, 0.4106, 0.4485, 0.5275, 0.5117],
  28787. [0.6653, 0.4102, 0.8407, 0.5278, 0.4112, 0.4669, 0.5808, 0.5438],
  28788. [0.6775, 0.4272, 0.8181, 0.3206, 0.3716, 0.3246, 0.5615, 0.5419],
  28789. [0.0300, 0.0195, 0.8411, 0.2661, 0.5341, 0.2151, 0.6922, 0.5651]],
  28790. device='cuda:0', grad_fn=<AddmmBackward>)
  28791. landmarks are: tensor([[[0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  28792. [0.6200, 0.4070, 0.8938, 0.4183, 0.3537, 0.4567, 0.6175, 0.5400],
  28793. [0.6284, 0.4029, 0.8838, 0.3783, 0.3975, 0.2850, 0.6335, 0.5090],
  28794. [0.6251, 0.4116, 0.8700, 0.4850, 0.4650, 0.5567, 0.6087, 0.5183],
  28795. [0.6175, 0.4093, 0.8800, 0.4433, 0.4075, 0.4367, 0.5128, 0.4901],
  28796. [0.6200, 0.3961, 0.8461, 0.5497, 0.4142, 0.4577, 0.5892, 0.5402],
  28797. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167],
  28798. [0.0000, 0.0000, 0.8675, 0.2833, 0.5350, 0.1983, 0.6678, 0.5621]]],
  28799. device='cuda:0')
  28800. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  28801. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  28802. loss_train: 0.011934660928091034
  28803. step: 29
  28804. running loss: 0.00041154003200313913
  28805. Train Steps: 29/90 Loss: 0.0004 torch.Size([8, 600, 800])
  28806. torch.Size([8, 8])
  28807. tensor([[0.6339, 0.4118, 0.7988, 0.5800, 0.3912, 0.4583, 0.7343, 0.5760],
  28808. [0.6205, 0.4081, 0.8950, 0.4017, 0.3788, 0.4700, 0.5963, 0.5667],
  28809. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  28810. [0.6109, 0.4003, 0.8650, 0.4883, 0.4775, 0.4867, 0.5175, 0.5683],
  28811. [ nan, nan, 0.8363, 0.3317, 0.3563, 0.3367, 0.5329, 0.5142],
  28812. [0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726],
  28813. [ nan, nan, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  28814. [0.6156, 0.4125, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084]],
  28815. device='cuda:0', dtype=torch.float64)
  28816. predictions are: tensor([[0.7444, 0.4550, 0.7838, 0.5614, 0.3935, 0.4228, 0.7239, 0.5780],
  28817. [0.7114, 0.4385, 0.8891, 0.3816, 0.3997, 0.4399, 0.6062, 0.5688],
  28818. [0.6936, 0.4127, 0.8160, 0.2357, 0.4294, 0.2486, 0.6374, 0.5281],
  28819. [0.6937, 0.4294, 0.8518, 0.4698, 0.5013, 0.4684, 0.5379, 0.5678],
  28820. [0.1745, 0.0923, 0.8023, 0.3025, 0.3639, 0.3292, 0.5370, 0.5189],
  28821. [0.6800, 0.4200, 0.7463, 0.2655, 0.3864, 0.2591, 0.5379, 0.4902],
  28822. [0.0922, 0.0367, 0.7440, 0.2449, 0.3967, 0.2344, 0.4970, 0.5574],
  28823. [0.6934, 0.4381, 0.8870, 0.4631, 0.4642, 0.5533, 0.5820, 0.5214]],
  28824. device='cuda:0', grad_fn=<AddmmBackward>)
  28825. landmarks are: tensor([[[0.6339, 0.4118, 0.7987, 0.5800, 0.3913, 0.4583, 0.7343, 0.5760],
  28826. [0.6205, 0.4081, 0.8950, 0.4017, 0.3787, 0.4700, 0.5962, 0.5667],
  28827. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  28828. [0.6109, 0.4003, 0.8650, 0.4883, 0.4775, 0.4867, 0.5175, 0.5683],
  28829. [0.0000, 0.0000, 0.8363, 0.3317, 0.3562, 0.3367, 0.5329, 0.5142],
  28830. [0.6031, 0.3986, 0.7462, 0.2833, 0.3638, 0.2717, 0.5253, 0.4726],
  28831. [0.0000, 0.0000, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  28832. [0.6155, 0.4124, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084]]],
  28833. device='cuda:0')
  28834. loss_train_step before backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  28835. loss_train_step after backward: tensor(0.0017, device='cuda:0', grad_fn=<MseLossBackward>)
  28836. loss_train: 0.013682589953532442
  28837. step: 30
  28838. running loss: 0.00045608633178441474
  28839.  
  28840. Train Steps: 30/90 Loss: 0.0005 torch.Size([8, 600, 800])
  28841. torch.Size([8, 8])
  28842. tensor([[0.6199, 0.4071, 0.7600, 0.2117, 0.4037, 0.2767, 0.6138, 0.5550],
  28843. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  28844. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901],
  28845. [0.6140, 0.4070, 0.8700, 0.5000, 0.4612, 0.4900, 0.5260, 0.5852],
  28846. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  28847. [0.6087, 0.3951, 0.8387, 0.5833, 0.4188, 0.4933, 0.5146, 0.4830],
  28848. [0.6165, 0.4106, 0.7575, 0.1733, 0.3838, 0.2650, 0.5680, 0.5116],
  28849. [0.6222, 0.4172, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633]],
  28850. device='cuda:0', dtype=torch.float64)
  28851. predictions are: tensor([[0.6379, 0.3909, 0.7579, 0.2245, 0.4173, 0.2712, 0.6210, 0.5547],
  28852. [0.6233, 0.3771, 0.8362, 0.4750, 0.3749, 0.4593, 0.5709, 0.5568],
  28853. [0.5827, 0.3658, 0.7819, 0.3057, 0.3846, 0.2551, 0.5367, 0.5088],
  28854. [0.6191, 0.3872, 0.8765, 0.5207, 0.4641, 0.4718, 0.5482, 0.5878],
  28855. [0.6211, 0.3721, 0.8128, 0.2735, 0.4169, 0.2571, 0.6458, 0.5189],
  28856. [0.6149, 0.3738, 0.8328, 0.6057, 0.4157, 0.4810, 0.5209, 0.4928],
  28857. [0.6080, 0.3842, 0.7540, 0.1936, 0.3971, 0.2233, 0.5777, 0.4945],
  28858. [0.5913, 0.3738, 0.8921, 0.5489, 0.3768, 0.4280, 0.5715, 0.5629]],
  28859. device='cuda:0', grad_fn=<AddmmBackward>)
  28860. landmarks are: tensor([[[0.6199, 0.4071, 0.7600, 0.2117, 0.4038, 0.2767, 0.6137, 0.5550],
  28861. [0.6128, 0.4116, 0.8450, 0.4583, 0.3675, 0.4867, 0.5337, 0.5446],
  28862. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901],
  28863. [0.6140, 0.4070, 0.8700, 0.5000, 0.4613, 0.4900, 0.5260, 0.5852],
  28864. [0.6275, 0.4008, 0.8047, 0.2484, 0.4037, 0.2777, 0.6506, 0.5212],
  28865. [0.6087, 0.3951, 0.8388, 0.5833, 0.4187, 0.4933, 0.5146, 0.4830],
  28866. [0.6165, 0.4106, 0.7575, 0.1733, 0.3837, 0.2650, 0.5680, 0.5116],
  28867. [0.6222, 0.4171, 0.8850, 0.5217, 0.3738, 0.4600, 0.5700, 0.5633]]],
  28868. device='cuda:0')
  28869. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28870. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28871. loss_train: 0.014052209124201909
  28872. step: 31
  28873. running loss: 0.0004532970685226422
  28874. Train Steps: 31/90 Loss: 0.0005 torch.Size([8, 600, 800])
  28875. torch.Size([8, 8])
  28876. tensor([[0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  28877. [0.6198, 0.4101, 0.8838, 0.5283, 0.3763, 0.5267, 0.5913, 0.5567],
  28878. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5963, 0.6217],
  28879. [0.6186, 0.4060, 0.8750, 0.5050, 0.3538, 0.4367, 0.5813, 0.6083],
  28880. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  28881. [ nan, nan, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  28882. [0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  28883. [0.6239, 0.4206, 0.8750, 0.5400, 0.3688, 0.4850, 0.5737, 0.5700]],
  28884. device='cuda:0', dtype=torch.float64)
  28885. predictions are: tensor([[ 0.6495, 0.4011, 0.8692, 0.5537, 0.3627, 0.4384, 0.6152, 0.4814],
  28886. [ 0.5820, 0.3635, 0.8739, 0.5224, 0.3707, 0.5128, 0.5727, 0.5245],
  28887. [ 0.6267, 0.4012, 0.7082, 0.2621, 0.4269, 0.2425, 0.5802, 0.5959],
  28888. [ 0.6077, 0.3814, 0.8682, 0.5002, 0.3490, 0.4299, 0.5728, 0.5763],
  28889. [ 0.6103, 0.3933, 0.8777, 0.4750, 0.4192, 0.5271, 0.5952, 0.5091],
  28890. [-0.0732, -0.0449, 0.8521, 0.2557, 0.5387, 0.2475, 0.7223, 0.5523],
  28891. [ 0.5827, 0.3624, 0.8614, 0.5299, 0.3894, 0.4689, 0.5755, 0.4910],
  28892. [ 0.6196, 0.3949, 0.8621, 0.5466, 0.3767, 0.4803, 0.5679, 0.5404]],
  28893. device='cuda:0', grad_fn=<AddmmBackward>)
  28894. landmarks are: tensor([[[0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  28895. [0.6198, 0.4101, 0.8838, 0.5283, 0.3762, 0.5267, 0.5913, 0.5567],
  28896. [0.6180, 0.4064, 0.7200, 0.2467, 0.4275, 0.2367, 0.5962, 0.6217],
  28897. [0.6186, 0.4060, 0.8750, 0.5050, 0.3537, 0.4367, 0.5813, 0.6083],
  28898. [0.6246, 0.4126, 0.8850, 0.4833, 0.4200, 0.5350, 0.6112, 0.5250],
  28899. [0.0000, 0.0000, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  28900. [0.6199, 0.4015, 0.8716, 0.5228, 0.3833, 0.4772, 0.5883, 0.5148],
  28901. [0.6239, 0.4206, 0.8750, 0.5400, 0.3688, 0.4850, 0.5738, 0.5700]]],
  28902. device='cuda:0')
  28903. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28904. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  28905. loss_train: 0.014462272200034931
  28906. step: 32
  28907. running loss: 0.0004519460062510916
  28908. Train Steps: 32/90 Loss: 0.0005 torch.Size([8, 600, 800])
  28909. torch.Size([8, 8])
  28910. tensor([[0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767],
  28911. [0.6109, 0.4015, 0.7668, 0.3639, 0.3513, 0.3667, 0.5200, 0.5641],
  28912. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  28913. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  28914. [0.6276, 0.4002, 0.8800, 0.5533, 0.3575, 0.4400, 0.6132, 0.4672],
  28915. [0.6248, 0.4185, 0.8500, 0.5767, 0.4463, 0.4550, 0.5613, 0.5917],
  28916. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  28917. [0.6168, 0.4029, 0.8523, 0.3417, 0.3588, 0.5000, 0.6125, 0.5400]],
  28918. device='cuda:0', dtype=torch.float64)
  28919. predictions are: tensor([[0.5832, 0.3911, 0.8822, 0.4977, 0.3909, 0.4613, 0.5244, 0.5735],
  28920. [0.5820, 0.3953, 0.7600, 0.3603, 0.3348, 0.3778, 0.5203, 0.5651],
  28921. [0.5541, 0.3639, 0.8830, 0.5053, 0.3868, 0.5514, 0.7251, 0.5361],
  28922. [0.5904, 0.3968, 0.8927, 0.4745, 0.3561, 0.3798, 0.6333, 0.5047],
  28923. [0.6538, 0.4274, 0.8706, 0.5482, 0.3517, 0.4422, 0.6079, 0.4647],
  28924. [0.5155, 0.3559, 0.8503, 0.5779, 0.4409, 0.4510, 0.5537, 0.5804],
  28925. [0.5931, 0.3919, 0.7732, 0.4311, 0.3554, 0.4521, 0.5282, 0.5202],
  28926. [0.5779, 0.3841, 0.8554, 0.3345, 0.3512, 0.5036, 0.6164, 0.5412]],
  28927. device='cuda:0', grad_fn=<AddmmBackward>)
  28928. landmarks are: tensor([[[0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767],
  28929. [0.6109, 0.4015, 0.7668, 0.3639, 0.3512, 0.3667, 0.5200, 0.5641],
  28930. [0.6339, 0.4112, 0.8838, 0.5067, 0.4000, 0.5433, 0.7549, 0.5378],
  28931. [0.6292, 0.4010, 0.8988, 0.4800, 0.3638, 0.3817, 0.6357, 0.5051],
  28932. [0.6276, 0.4002, 0.8800, 0.5533, 0.3575, 0.4400, 0.6132, 0.4672],
  28933. [0.6248, 0.4185, 0.8500, 0.5767, 0.4462, 0.4550, 0.5612, 0.5917],
  28934. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  28935. [0.6168, 0.4029, 0.8523, 0.3417, 0.3587, 0.5000, 0.6125, 0.5400]]],
  28936. device='cuda:0')
  28937. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  28938. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  28939. loss_train: 0.015006558765890077
  28940. step: 33
  28941. running loss: 0.000454744205026972
  28942. Train Steps: 33/90 Loss: 0.0005 torch.Size([8, 600, 800])
  28943. torch.Size([8, 8])
  28944. tensor([[0.6236, 0.3967, 0.8675, 0.5400, 0.3862, 0.4517, 0.5825, 0.5200],
  28945. [0.6104, 0.4029, 0.8738, 0.4900, 0.4088, 0.4533, 0.5070, 0.5510],
  28946. [0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  28947. [0.6204, 0.4049, 0.7975, 0.2700, 0.3937, 0.2567, 0.5700, 0.5183],
  28948. [0.6200, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  28949. [0.6286, 0.4086, 0.8408, 0.2801, 0.4163, 0.2800, 0.6725, 0.5393],
  28950. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6038, 0.6167],
  28951. [ nan, nan, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819]],
  28952. device='cuda:0', dtype=torch.float64)
  28953. predictions are: tensor([[ 0.6451, 0.4259, 0.8675, 0.5698, 0.3666, 0.4369, 0.5717, 0.5125],
  28954. [ 0.5645, 0.3908, 0.8689, 0.5002, 0.3998, 0.4359, 0.5041, 0.5314],
  28955. [ 0.6088, 0.4146, 0.7864, 0.1810, 0.4185, 0.2331, 0.6403, 0.5081],
  28956. [ 0.5629, 0.3920, 0.7945, 0.2791, 0.3713, 0.2506, 0.5538, 0.5200],
  28957. [ 0.6072, 0.3994, 0.8853, 0.4779, 0.3450, 0.4178, 0.5725, 0.5193],
  28958. [ 0.5482, 0.3731, 0.8415, 0.2738, 0.3922, 0.2913, 0.6542, 0.5502],
  28959. [ 0.5835, 0.4134, 0.8025, 0.3365, 0.3402, 0.3861, 0.5850, 0.6057],
  28960. [-0.0904, -0.0457, 0.7108, 0.2879, 0.3720, 0.2219, 0.5418, 0.5538]],
  28961. device='cuda:0', grad_fn=<AddmmBackward>)
  28962. landmarks are: tensor([[[0.6236, 0.3967, 0.8675, 0.5400, 0.3862, 0.4517, 0.5825, 0.5200],
  28963. [0.6104, 0.4029, 0.8737, 0.4900, 0.4087, 0.4533, 0.5070, 0.5510],
  28964. [0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  28965. [0.6204, 0.4049, 0.7975, 0.2700, 0.3938, 0.2567, 0.5700, 0.5183],
  28966. [0.6201, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  28967. [0.6286, 0.4086, 0.8408, 0.2801, 0.4162, 0.2800, 0.6725, 0.5393],
  28968. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6037, 0.6167],
  28969. [0.0000, 0.0000, 0.6992, 0.2791, 0.3950, 0.2383, 0.5483, 0.5819]]],
  28970. device='cuda:0')
  28971. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  28972. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  28973. loss_train: 0.015572562260786071
  28974. step: 34
  28975. running loss: 0.00045801653708194326
  28976.  
  28977. Train Steps: 34/90 Loss: 0.0005 torch.Size([8, 600, 800])
  28978. torch.Size([8, 8])
  28979. tensor([[0.6120, 0.4014, 0.6863, 0.2817, 0.3700, 0.2783, 0.5513, 0.5667],
  28980. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  28981. [0.6135, 0.3994, 0.7913, 0.3050, 0.3625, 0.3050, 0.5837, 0.5050],
  28982. [0.6162, 0.4134, 0.6700, 0.2467, 0.3962, 0.2533, 0.5737, 0.5467],
  28983. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5463, 0.5800],
  28984. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250],
  28985. [0.6241, 0.4143, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  28986. [ nan, nan, 0.7425, 0.2117, 0.3937, 0.2433, 0.5438, 0.5567]],
  28987. device='cuda:0', dtype=torch.float64)
  28988. predictions are: tensor([[ 0.5648, 0.3873, 0.7104, 0.2988, 0.3594, 0.2853, 0.5460, 0.5725],
  28989. [ 0.5962, 0.4024, 0.8461, 0.3272, 0.3354, 0.3844, 0.6203, 0.5423],
  28990. [ 0.5860, 0.4008, 0.7996, 0.3109, 0.3559, 0.3106, 0.5874, 0.5184],
  28991. [ 0.5606, 0.4088, 0.6864, 0.2589, 0.3830, 0.2525, 0.5607, 0.5531],
  28992. [ 0.5616, 0.4096, 0.7585, 0.3136, 0.3572, 0.2629, 0.5283, 0.5779],
  28993. [ 0.5532, 0.3874, 0.9154, 0.4124, 0.3580, 0.3366, 0.6117, 0.5302],
  28994. [ 0.5849, 0.4117, 0.9025, 0.4614, 0.3938, 0.5506, 0.6109, 0.5616],
  28995. [-0.0348, -0.0089, 0.7443, 0.2201, 0.3803, 0.2465, 0.5251, 0.5509]],
  28996. device='cuda:0', grad_fn=<AddmmBackward>)
  28997. landmarks are: tensor([[[0.6120, 0.4013, 0.6862, 0.2817, 0.3700, 0.2783, 0.5512, 0.5667],
  28998. [0.6200, 0.4024, 0.8390, 0.3139, 0.3525, 0.3833, 0.6162, 0.5383],
  28999. [0.6135, 0.3994, 0.7912, 0.3050, 0.3625, 0.3050, 0.5838, 0.5050],
  29000. [0.6162, 0.4134, 0.6700, 0.2467, 0.3963, 0.2533, 0.5738, 0.5467],
  29001. [0.6150, 0.4097, 0.7468, 0.3194, 0.3825, 0.2633, 0.5462, 0.5800],
  29002. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250],
  29003. [0.6241, 0.4142, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  29004. [0.0000, 0.0000, 0.7425, 0.2117, 0.3938, 0.2433, 0.5437, 0.5567]]],
  29005. device='cuda:0')
  29006. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  29007. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  29008. loss_train: 0.015958790871081874
  29009. step: 35
  29010. running loss: 0.0004559654534594821
  29011. Train Steps: 35/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29012. torch.Size([8, 8])
  29013. tensor([[0.6135, 0.3994, 0.7913, 0.3050, 0.3625, 0.3050, 0.5837, 0.5050],
  29014. [0.6140, 0.4070, 0.8700, 0.5000, 0.4612, 0.4900, 0.5260, 0.5852],
  29015. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  29016. [0.6239, 0.4174, 0.8425, 0.5733, 0.4825, 0.4500, 0.5625, 0.5933],
  29017. [0.6185, 0.4129, 0.8900, 0.4567, 0.3937, 0.5417, 0.5734, 0.5110],
  29018. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  29019. [0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  29020. [0.6260, 0.4153, 0.9000, 0.4533, 0.4025, 0.2633, 0.6223, 0.4967]],
  29021. device='cuda:0', dtype=torch.float64)
  29022. predictions are: tensor([[0.5715, 0.3983, 0.7859, 0.2862, 0.3385, 0.2973, 0.5770, 0.5222],
  29023. [0.5618, 0.4052, 0.8710, 0.4820, 0.4305, 0.4784, 0.5196, 0.5981],
  29024. [0.6019, 0.4226, 0.8774, 0.4308, 0.4082, 0.5659, 0.6071, 0.5512],
  29025. [0.5632, 0.4007, 0.8402, 0.5524, 0.4494, 0.4242, 0.5551, 0.6019],
  29026. [0.5532, 0.4036, 0.8765, 0.4356, 0.3600, 0.5315, 0.5501, 0.5206],
  29027. [0.5350, 0.3746, 0.8441, 0.2485, 0.4228, 0.2560, 0.6971, 0.5770],
  29028. [0.5611, 0.3938, 0.8491, 0.5134, 0.3984, 0.5053, 0.6425, 0.5280],
  29029. [0.5484, 0.3925, 0.8883, 0.4308, 0.3827, 0.2655, 0.6115, 0.5072]],
  29030. device='cuda:0', grad_fn=<AddmmBackward>)
  29031. landmarks are: tensor([[[0.6135, 0.3994, 0.7912, 0.3050, 0.3625, 0.3050, 0.5838, 0.5050],
  29032. [0.6140, 0.4070, 0.8700, 0.5000, 0.4613, 0.4900, 0.5260, 0.5852],
  29033. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  29034. [0.6239, 0.4174, 0.8425, 0.5733, 0.4825, 0.4500, 0.5625, 0.5933],
  29035. [0.6186, 0.4129, 0.8900, 0.4567, 0.3938, 0.5417, 0.5734, 0.5110],
  29036. [0.6264, 0.4055, 0.8425, 0.2767, 0.4425, 0.2767, 0.7050, 0.5586],
  29037. [0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  29038. [0.6260, 0.4153, 0.9000, 0.4533, 0.4025, 0.2633, 0.6223, 0.4967]]],
  29039. device='cuda:0')
  29040. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  29041. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  29042. loss_train: 0.016739433369366452
  29043. step: 36
  29044. running loss: 0.00046498426026017923
  29045. Train Steps: 36/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29046. torch.Size([8, 8])
  29047. tensor([[0.6185, 0.4080, 0.8625, 0.3483, 0.3788, 0.2650, 0.5320, 0.5272],
  29048. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  29049. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  29050. [0.6179, 0.4008, 0.7505, 0.2678, 0.4368, 0.1891, 0.5831, 0.5263],
  29051. [0.6058, 0.3986, 0.8324, 0.4626, 0.3838, 0.4983, 0.5147, 0.5466],
  29052. [0.6265, 0.4071, 0.8875, 0.3367, 0.3975, 0.3350, 0.6312, 0.5250],
  29053. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  29054. [0.6122, 0.3993, 0.8738, 0.4667, 0.4517, 0.4879, 0.5155, 0.4927]],
  29055. device='cuda:0', dtype=torch.float64)
  29056. predictions are: tensor([[0.5706, 0.3897, 0.8698, 0.3285, 0.3751, 0.2562, 0.5224, 0.5354],
  29057. [0.5487, 0.3692, 0.7964, 0.2755, 0.3574, 0.3253, 0.5934, 0.5550],
  29058. [0.5685, 0.3815, 0.7108, 0.2668, 0.4292, 0.2223, 0.5417, 0.6050],
  29059. [0.6088, 0.4080, 0.7604, 0.2438, 0.4274, 0.1796, 0.5675, 0.5478],
  29060. [0.5814, 0.3926, 0.8298, 0.4636, 0.3864, 0.4965, 0.5183, 0.5379],
  29061. [0.5431, 0.3703, 0.8917, 0.3262, 0.3925, 0.3486, 0.6546, 0.5392],
  29062. [0.5391, 0.3666, 0.8971, 0.5011, 0.3574, 0.3801, 0.6286, 0.4892],
  29063. [0.5493, 0.3695, 0.8689, 0.4610, 0.4409, 0.4958, 0.5160, 0.4922]],
  29064. device='cuda:0', grad_fn=<AddmmBackward>)
  29065. landmarks are: tensor([[[0.6186, 0.4080, 0.8625, 0.3483, 0.3787, 0.2650, 0.5320, 0.5272],
  29066. [0.6182, 0.3987, 0.7878, 0.2889, 0.3699, 0.3260, 0.6086, 0.5367],
  29067. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  29068. [0.6179, 0.4008, 0.7505, 0.2678, 0.4368, 0.1891, 0.5831, 0.5263],
  29069. [0.6058, 0.3986, 0.8324, 0.4626, 0.3837, 0.4983, 0.5147, 0.5466],
  29070. [0.6265, 0.4071, 0.8875, 0.3367, 0.3975, 0.3350, 0.6313, 0.5250],
  29071. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  29072. [0.6122, 0.3993, 0.8737, 0.4667, 0.4517, 0.4879, 0.5155, 0.4927]]],
  29073. device='cuda:0')
  29074. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  29075. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  29076. loss_train: 0.017385684681357816
  29077. step: 37
  29078. running loss: 0.0004698833697664275
  29079. Train Steps: 37/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29080. torch.Size([8, 8])
  29081. tensor([[0.6132, 0.4118, 0.8200, 0.3633, 0.3563, 0.5400, 0.5787, 0.5136],
  29082. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5413, 0.5683],
  29083. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  29084. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  29085. [0.6115, 0.4005, 0.8838, 0.3867, 0.3763, 0.4700, 0.5800, 0.5550],
  29086. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250],
  29087. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  29088. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117]],
  29089. device='cuda:0', dtype=torch.float64)
  29090. predictions are: tensor([[0.6185, 0.4146, 0.8208, 0.3592, 0.3667, 0.5381, 0.5866, 0.5155],
  29091. [0.5871, 0.3943, 0.8878, 0.5392, 0.3985, 0.3881, 0.5449, 0.5594],
  29092. [0.6026, 0.4076, 0.6866, 0.2655, 0.3988, 0.2523, 0.5521, 0.5697],
  29093. [0.6158, 0.4022, 0.7173, 0.2858, 0.4572, 0.2319, 0.5530, 0.5862],
  29094. [0.6032, 0.4139, 0.8797, 0.3877, 0.3929, 0.4822, 0.5769, 0.5445],
  29095. [0.6120, 0.4118, 0.9116, 0.4091, 0.3926, 0.3357, 0.6214, 0.5160],
  29096. [0.6625, 0.4409, 0.7361, 0.2061, 0.4297, 0.2643, 0.6194, 0.5430],
  29097. [0.5817, 0.3856, 0.8910, 0.3248, 0.3996, 0.3122, 0.5976, 0.5089]],
  29098. device='cuda:0', grad_fn=<AddmmBackward>)
  29099. landmarks are: tensor([[[0.6132, 0.4118, 0.8200, 0.3633, 0.3562, 0.5400, 0.5787, 0.5136],
  29100. [0.6200, 0.4101, 0.8838, 0.5317, 0.3825, 0.3800, 0.5412, 0.5683],
  29101. [0.6169, 0.4108, 0.6821, 0.2722, 0.3825, 0.2550, 0.5550, 0.5717],
  29102. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  29103. [0.6115, 0.4005, 0.8838, 0.3867, 0.3762, 0.4700, 0.5800, 0.5550],
  29104. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250],
  29105. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  29106. [0.6205, 0.4012, 0.8675, 0.3283, 0.3713, 0.3050, 0.5813, 0.5117]]],
  29107. device='cuda:0')
  29108. loss_train_step before backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  29109. loss_train_step after backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  29110. loss_train: 0.017598614853341132
  29111. step: 38
  29112. running loss: 0.00046312144350897714
  29113.  
  29114. Train Steps: 38/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29115. torch.Size([8, 8])
  29116. tensor([[0.6043, 0.4022, 0.6887, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136],
  29117. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  29118. [0.6162, 0.4014, 0.8800, 0.5333, 0.3750, 0.4817, 0.5988, 0.5283],
  29119. [0.6131, 0.4064, 0.8638, 0.5200, 0.4788, 0.4783, 0.5258, 0.5867],
  29120. [ nan, nan, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  29121. [0.6273, 0.4100, 0.7137, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633],
  29122. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  29123. [0.6304, 0.4024, 0.8925, 0.4800, 0.3937, 0.4817, 0.7485, 0.5297]],
  29124. device='cuda:0', dtype=torch.float64)
  29125. predictions are: tensor([[0.6718, 0.4344, 0.6978, 0.2278, 0.3973, 0.2618, 0.5353, 0.5026],
  29126. [0.6814, 0.4551, 0.8863, 0.4626, 0.3983, 0.4597, 0.5437, 0.5441],
  29127. [0.6454, 0.4267, 0.8699, 0.5458, 0.3998, 0.5120, 0.5762, 0.5138],
  29128. [0.6757, 0.4400, 0.8705, 0.5272, 0.4855, 0.4925, 0.5310, 0.5669],
  29129. [0.0959, 0.0477, 0.7053, 0.2470, 0.4459, 0.1847, 0.5273, 0.5702],
  29130. [0.6634, 0.4301, 0.7144, 0.2299, 0.4122, 0.2717, 0.6030, 0.5523],
  29131. [0.7147, 0.4760, 0.8374, 0.5446, 0.4280, 0.5935, 0.7018, 0.5490],
  29132. [0.6632, 0.4300, 0.8874, 0.5051, 0.4079, 0.4964, 0.7220, 0.5106]],
  29133. device='cuda:0', grad_fn=<AddmmBackward>)
  29134. landmarks are: tensor([[[0.6043, 0.4022, 0.6888, 0.1983, 0.3775, 0.2483, 0.5480, 0.5136],
  29135. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  29136. [0.6162, 0.4014, 0.8800, 0.5333, 0.3750, 0.4817, 0.5987, 0.5283],
  29137. [0.6132, 0.4063, 0.8637, 0.5200, 0.4787, 0.4783, 0.5258, 0.5867],
  29138. [0.0000, 0.0000, 0.6859, 0.2194, 0.4150, 0.1867, 0.5153, 0.5729],
  29139. [0.6273, 0.4099, 0.7138, 0.2133, 0.4000, 0.2650, 0.6075, 0.5633],
  29140. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  29141. [0.6304, 0.4024, 0.8925, 0.4800, 0.3938, 0.4817, 0.7485, 0.5297]]],
  29142. device='cuda:0')
  29143. loss_train_step before backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  29144. loss_train_step after backward: tensor(0.0010, device='cuda:0', grad_fn=<MseLossBackward>)
  29145. loss_train: 0.018550915468949825
  29146. step: 39
  29147. running loss: 0.00047566449920384167
  29148. Train Steps: 39/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29149. torch.Size([8, 8])
  29150. tensor([[0.6053, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  29151. [0.6339, 0.4118, 0.7988, 0.5800, 0.3912, 0.4583, 0.7343, 0.5760],
  29152. [0.6185, 0.4129, 0.8900, 0.4567, 0.3937, 0.5417, 0.5734, 0.5110],
  29153. [0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  29154. [0.6072, 0.4029, 0.7037, 0.2150, 0.3912, 0.2267, 0.5516, 0.5507],
  29155. [ nan, nan, 0.6412, 0.1900, 0.4238, 0.1883, 0.5487, 0.5700],
  29156. [0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153],
  29157. [0.6185, 0.4067, 0.8838, 0.4450, 0.4037, 0.4733, 0.5213, 0.5142]],
  29158. device='cuda:0', dtype=torch.float64)
  29159. predictions are: tensor([[0.6169, 0.3839, 0.6956, 0.2080, 0.4152, 0.2001, 0.5619, 0.5252],
  29160. [0.7022, 0.4459, 0.7833, 0.5770, 0.3878, 0.4553, 0.7285, 0.5615],
  29161. [0.7060, 0.4618, 0.8837, 0.4516, 0.4031, 0.5463, 0.5737, 0.5116],
  29162. [0.6692, 0.4372, 0.8901, 0.4432, 0.3769, 0.3541, 0.5819, 0.5860],
  29163. [0.6585, 0.4263, 0.7016, 0.2302, 0.4021, 0.2140, 0.5644, 0.5644],
  29164. [0.1082, 0.0597, 0.6819, 0.2193, 0.4518, 0.2031, 0.5519, 0.5844],
  29165. [0.6855, 0.4326, 0.8580, 0.5093, 0.4684, 0.5414, 0.5962, 0.5084],
  29166. [0.6360, 0.4014, 0.8779, 0.4440, 0.4092, 0.4681, 0.5291, 0.5039]],
  29167. device='cuda:0', grad_fn=<AddmmBackward>)
  29168. landmarks are: tensor([[[0.6054, 0.4035, 0.6897, 0.1916, 0.4000, 0.2117, 0.5440, 0.5168],
  29169. [0.6339, 0.4118, 0.7987, 0.5800, 0.3913, 0.4583, 0.7343, 0.5760],
  29170. [0.6186, 0.4129, 0.8900, 0.4567, 0.3938, 0.5417, 0.5734, 0.5110],
  29171. [0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  29172. [0.6072, 0.4029, 0.7038, 0.2150, 0.3913, 0.2267, 0.5516, 0.5507],
  29173. [0.0000, 0.0000, 0.6413, 0.1900, 0.4238, 0.1883, 0.5487, 0.5700],
  29174. [0.6150, 0.3935, 0.8696, 0.5158, 0.4647, 0.5329, 0.6041, 0.5153],
  29175. [0.6185, 0.4067, 0.8838, 0.4450, 0.4038, 0.4733, 0.5213, 0.5142]]],
  29176. device='cuda:0')
  29177. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  29178. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  29179. loss_train: 0.01935047988081351
  29180. step: 40
  29181. running loss: 0.00048376199702033774
  29182. Train Steps: 40/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29183. torch.Size([8, 8])
  29184. tensor([[0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  29185. [0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510],
  29186. [0.6250, 0.4054, 0.8770, 0.4723, 0.4662, 0.5367, 0.6162, 0.5433],
  29187. [0.6200, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  29188. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  29189. [0.6198, 0.4115, 0.7762, 0.2717, 0.3713, 0.3200, 0.5837, 0.5683],
  29190. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  29191. [0.6200, 0.3993, 0.8519, 0.4923, 0.3962, 0.4717, 0.6013, 0.5433]],
  29192. device='cuda:0', dtype=torch.float64)
  29193. predictions are: tensor([[0.6516, 0.4067, 0.7639, 0.2336, 0.3768, 0.3196, 0.6163, 0.5172],
  29194. [0.5634, 0.3469, 0.7085, 0.2859, 0.3648, 0.3074, 0.5218, 0.5476],
  29195. [0.6775, 0.4278, 0.8840, 0.4648, 0.4622, 0.5190, 0.6203, 0.5268],
  29196. [0.6685, 0.4191, 0.8493, 0.2953, 0.4208, 0.2197, 0.5933, 0.5217],
  29197. [0.6808, 0.4332, 0.8404, 0.5141, 0.4195, 0.5606, 0.7106, 0.5578],
  29198. [0.6479, 0.4090, 0.7783, 0.2634, 0.3727, 0.3091, 0.5903, 0.5485],
  29199. [0.6854, 0.4254, 0.8617, 0.4614, 0.4806, 0.5480, 0.6056, 0.5317],
  29200. [0.6326, 0.3945, 0.8462, 0.4798, 0.3954, 0.4619, 0.6060, 0.5428]],
  29201. device='cuda:0', grad_fn=<AddmmBackward>)
  29202. landmarks are: tensor([[[0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  29203. [0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510],
  29204. [0.6250, 0.4054, 0.8770, 0.4723, 0.4663, 0.5367, 0.6162, 0.5433],
  29205. [0.6199, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  29206. [0.6304, 0.4029, 0.8413, 0.5217, 0.4125, 0.5617, 0.7089, 0.5679],
  29207. [0.6198, 0.4115, 0.7763, 0.2717, 0.3713, 0.3200, 0.5838, 0.5683],
  29208. [0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  29209. [0.6200, 0.3993, 0.8519, 0.4923, 0.3963, 0.4717, 0.6012, 0.5433]]],
  29210. device='cuda:0')
  29211. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  29212. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  29213. loss_train: 0.019759502029046416
  29214. step: 41
  29215. running loss: 0.0004819390738791809
  29216. Train Steps: 41/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29217. torch.Size([8, 8])
  29218. tensor([[0.6187, 0.4104, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  29219. [0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  29220. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467],
  29221. [0.6213, 0.4001, 0.7712, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183],
  29222. [0.6248, 0.4185, 0.8500, 0.5767, 0.4463, 0.4550, 0.5613, 0.5917],
  29223. [0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6088, 0.5583],
  29224. [0.6204, 0.4013, 0.8075, 0.2400, 0.4313, 0.2050, 0.5800, 0.5150],
  29225. [0.6332, 0.4165, 0.9100, 0.3350, 0.4188, 0.3683, 0.7438, 0.5528]],
  29226. device='cuda:0', dtype=torch.float64)
  29227. predictions are: tensor([[0.6332, 0.3977, 0.6971, 0.2190, 0.3922, 0.2525, 0.5969, 0.5592],
  29228. [0.6500, 0.3900, 0.7688, 0.1750, 0.4375, 0.2283, 0.6539, 0.5060],
  29229. [0.5769, 0.3600, 0.8426, 0.3437, 0.3481, 0.4058, 0.5943, 0.5408],
  29230. [0.6526, 0.3892, 0.7577, 0.2210, 0.4365, 0.1818, 0.5951, 0.5200],
  29231. [0.7137, 0.4460, 0.8387, 0.6048, 0.4453, 0.4621, 0.5559, 0.5927],
  29232. [0.6569, 0.4075, 0.6810, 0.2255, 0.3898, 0.2605, 0.5880, 0.5565],
  29233. [0.6181, 0.3658, 0.7769, 0.2475, 0.4378, 0.2099, 0.5868, 0.5283],
  29234. [0.6375, 0.4019, 0.9070, 0.3575, 0.4146, 0.3607, 0.7427, 0.5510]],
  29235. device='cuda:0', grad_fn=<AddmmBackward>)
  29236. landmarks are: tensor([[[0.6187, 0.4103, 0.7126, 0.2069, 0.3950, 0.2517, 0.5913, 0.5683],
  29237. [0.6264, 0.4069, 0.7900, 0.1650, 0.4275, 0.2267, 0.6290, 0.4947],
  29238. [0.6136, 0.3955, 0.8400, 0.3267, 0.3500, 0.4200, 0.5863, 0.5467],
  29239. [0.6213, 0.4001, 0.7713, 0.2117, 0.4338, 0.1800, 0.5850, 0.5183],
  29240. [0.6248, 0.4185, 0.8500, 0.5767, 0.4462, 0.4550, 0.5612, 0.5917],
  29241. [0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6087, 0.5583],
  29242. [0.6204, 0.4013, 0.8075, 0.2400, 0.4313, 0.2050, 0.5800, 0.5150],
  29243. [0.6332, 0.4165, 0.9100, 0.3350, 0.4187, 0.3683, 0.7438, 0.5528]]],
  29244. device='cuda:0')
  29245. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  29246. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  29247. loss_train: 0.02012144192121923
  29248. step: 42
  29249. running loss: 0.00047908195050521974
  29250.  
  29251. Train Steps: 42/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29252. torch.Size([8, 8])
  29253. tensor([[ nan, nan, 0.8850, 0.2817, 0.5112, 0.2183, 0.7184, 0.5436],
  29254. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290],
  29255. [0.6109, 0.4003, 0.8650, 0.4883, 0.4775, 0.4867, 0.5175, 0.5683],
  29256. [0.6075, 0.4000, 0.8513, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  29257. [0.6213, 0.4131, 0.8438, 0.3550, 0.3513, 0.4400, 0.5716, 0.5123],
  29258. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901],
  29259. [0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  29260. [0.6245, 0.4115, 0.8700, 0.4883, 0.4625, 0.5517, 0.6100, 0.5217]],
  29261. device='cuda:0', dtype=torch.float64)
  29262. predictions are: tensor([[ 0.0542, -0.0027, 0.8669, 0.2556, 0.5190, 0.1917, 0.7471, 0.5670],
  29263. [ 0.6373, 0.3760, 0.7465, 0.2069, 0.4317, 0.2301, 0.6611, 0.5499],
  29264. [ 0.6509, 0.4078, 0.8469, 0.4730, 0.4775, 0.4878, 0.5511, 0.5893],
  29265. [ 0.6428, 0.3937, 0.8409, 0.5153, 0.4562, 0.5230, 0.5367, 0.5416],
  29266. [ 0.6606, 0.4011, 0.8424, 0.3529, 0.3611, 0.4268, 0.5778, 0.5345],
  29267. [ 0.6471, 0.4007, 0.7889, 0.2801, 0.3805, 0.2498, 0.5351, 0.5312],
  29268. [ 0.6499, 0.3934, 0.8639, 0.4930, 0.4056, 0.5281, 0.6079, 0.5312],
  29269. [ 0.6847, 0.4187, 0.8458, 0.4770, 0.4485, 0.5440, 0.6391, 0.5610]],
  29270. device='cuda:0', grad_fn=<AddmmBackward>)
  29271. landmarks are: tensor([[[0.0000, 0.0000, 0.8850, 0.2817, 0.5113, 0.2183, 0.7184, 0.5436],
  29272. [0.6275, 0.4024, 0.7722, 0.2080, 0.4392, 0.2234, 0.6435, 0.5290],
  29273. [0.6109, 0.4003, 0.8650, 0.4883, 0.4775, 0.4867, 0.5175, 0.5683],
  29274. [0.6075, 0.4000, 0.8512, 0.5183, 0.4510, 0.5329, 0.5180, 0.5280],
  29275. [0.6213, 0.4131, 0.8438, 0.3550, 0.3512, 0.4400, 0.5716, 0.5123],
  29276. [0.6143, 0.4055, 0.8150, 0.2767, 0.3825, 0.2567, 0.5173, 0.4901],
  29277. [0.6176, 0.4017, 0.8788, 0.5100, 0.4075, 0.5250, 0.5913, 0.5033],
  29278. [0.6245, 0.4115, 0.8700, 0.4883, 0.4625, 0.5517, 0.6100, 0.5217]]],
  29279. device='cuda:0')
  29280. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  29281. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  29282. loss_train: 0.020595025533111766
  29283. step: 43
  29284. running loss: 0.0004789540821653899
  29285. Train Steps: 43/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29286. torch.Size([8, 8])
  29287. tensor([[0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  29288. [0.6127, 0.4118, 0.8650, 0.5083, 0.4088, 0.5367, 0.5300, 0.5456],
  29289. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  29290. [0.6113, 0.4006, 0.8700, 0.5350, 0.3638, 0.3767, 0.5097, 0.4882],
  29291. [0.6259, 0.4133, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  29292. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  29293. [0.6339, 0.4102, 0.9088, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  29294. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333]],
  29295. device='cuda:0', dtype=torch.float64)
  29296. predictions are: tensor([[0.5879, 0.3822, 0.8430, 0.3862, 0.3465, 0.4548, 0.5848, 0.5375],
  29297. [0.6455, 0.4225, 0.8797, 0.4847, 0.4201, 0.5532, 0.5451, 0.5790],
  29298. [0.5972, 0.3725, 0.7776, 0.4046, 0.3612, 0.4352, 0.5180, 0.5511],
  29299. [0.5742, 0.3705, 0.8650, 0.5016, 0.3802, 0.4013, 0.5332, 0.5245],
  29300. [0.6719, 0.4335, 0.8126, 0.2123, 0.5088, 0.1532, 0.6522, 0.5176],
  29301. [0.6547, 0.4174, 0.8908, 0.4526, 0.4380, 0.5870, 0.6292, 0.5493],
  29302. [0.6425, 0.4106, 0.8989, 0.4601, 0.4023, 0.5418, 0.7542, 0.5687],
  29303. [0.6464, 0.4193, 0.7281, 0.1871, 0.4320, 0.2155, 0.6112, 0.5420]],
  29304. device='cuda:0', grad_fn=<AddmmBackward>)
  29305. landmarks are: tensor([[[0.6157, 0.3956, 0.8323, 0.4138, 0.3479, 0.4431, 0.5914, 0.5131],
  29306. [0.6127, 0.4118, 0.8650, 0.5083, 0.4087, 0.5367, 0.5300, 0.5456],
  29307. [0.6086, 0.4019, 0.7782, 0.4278, 0.3625, 0.4350, 0.5150, 0.5285],
  29308. [0.6113, 0.4006, 0.8700, 0.5350, 0.3638, 0.3767, 0.5097, 0.4882],
  29309. [0.6259, 0.4132, 0.8200, 0.2317, 0.5025, 0.1533, 0.6250, 0.4947],
  29310. [0.6251, 0.4108, 0.8888, 0.4700, 0.4325, 0.5817, 0.6075, 0.5150],
  29311. [0.6339, 0.4102, 0.9087, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  29312. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333]]],
  29313. device='cuda:0')
  29314. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  29315. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  29316. loss_train: 0.02098468688200228
  29317. step: 44
  29318. running loss: 0.00047692470186368814
  29319. Train Steps: 44/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29320. torch.Size([8, 8])
  29321. tensor([[0.6250, 0.4106, 0.8700, 0.3717, 0.3588, 0.4967, 0.6038, 0.5167],
  29322. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  29323. [0.6284, 0.4127, 0.8538, 0.5867, 0.4363, 0.5083, 0.6038, 0.5433],
  29324. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5837, 0.5500],
  29325. [0.6252, 0.4158, 0.8988, 0.4083, 0.3788, 0.4783, 0.6225, 0.5633],
  29326. [0.6199, 0.4065, 0.7598, 0.2385, 0.4317, 0.1981, 0.5933, 0.5221],
  29327. [0.6300, 0.4102, 0.9088, 0.4433, 0.4088, 0.3067, 0.6820, 0.5540],
  29328. [0.6276, 0.4095, 0.8237, 0.2250, 0.4662, 0.1783, 0.6171, 0.4869]],
  29329. device='cuda:0', dtype=torch.float64)
  29330. predictions are: tensor([[0.5912, 0.3878, 0.8807, 0.3672, 0.3595, 0.4962, 0.6225, 0.5206],
  29331. [0.5541, 0.3758, 0.7213, 0.2091, 0.3880, 0.2429, 0.5415, 0.5559],
  29332. [0.5959, 0.3878, 0.8543, 0.5744, 0.4372, 0.5065, 0.5773, 0.5287],
  29333. [0.5881, 0.3708, 0.8818, 0.4274, 0.4238, 0.5098, 0.5767, 0.5491],
  29334. [0.6098, 0.4083, 0.8911, 0.3873, 0.3684, 0.4666, 0.6218, 0.5463],
  29335. [0.6152, 0.3969, 0.7550, 0.2310, 0.4111, 0.2159, 0.5823, 0.5025],
  29336. [0.5669, 0.3730, 0.9082, 0.4236, 0.4116, 0.3081, 0.6940, 0.5416],
  29337. [0.6773, 0.4370, 0.8223, 0.2095, 0.4636, 0.1752, 0.6199, 0.4743]],
  29338. device='cuda:0', grad_fn=<AddmmBackward>)
  29339. landmarks are: tensor([[[0.6250, 0.4105, 0.8700, 0.3717, 0.3587, 0.4967, 0.6037, 0.5167],
  29340. [0.6090, 0.4045, 0.7250, 0.2100, 0.4075, 0.2300, 0.5476, 0.5663],
  29341. [0.6284, 0.4127, 0.8537, 0.5867, 0.4363, 0.5083, 0.6037, 0.5433],
  29342. [0.6162, 0.3949, 0.8838, 0.4517, 0.4250, 0.5183, 0.5838, 0.5500],
  29343. [0.6252, 0.4158, 0.8988, 0.4083, 0.3787, 0.4783, 0.6225, 0.5633],
  29344. [0.6199, 0.4065, 0.7598, 0.2385, 0.4317, 0.1981, 0.5933, 0.5221],
  29345. [0.6300, 0.4102, 0.9087, 0.4433, 0.4087, 0.3067, 0.6820, 0.5540],
  29346. [0.6276, 0.4095, 0.8238, 0.2250, 0.4663, 0.1783, 0.6171, 0.4869]]],
  29347. device='cuda:0')
  29348. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  29349. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  29350. loss_train: 0.021359838166972622
  29351. step: 45
  29352. running loss: 0.0004746630703771694
  29353. Train Steps: 45/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29354. torch.Size([8, 8])
  29355. tensor([[0.6329, 0.4055, 0.9050, 0.4783, 0.3613, 0.3917, 0.6464, 0.5019],
  29356. [0.6218, 0.4137, 0.7263, 0.2233, 0.4075, 0.2650, 0.6212, 0.5783],
  29357. [0.6145, 0.4007, 0.8775, 0.4533, 0.4562, 0.5533, 0.6088, 0.5533],
  29358. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5637, 0.5633],
  29359. [ nan, nan, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  29360. [0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767],
  29361. [0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6138, 0.5400],
  29362. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896]],
  29363. device='cuda:0', dtype=torch.float64)
  29364. predictions are: tensor([[0.5748, 0.3760, 0.9063, 0.4830, 0.3700, 0.4068, 0.6425, 0.4730],
  29365. [0.6236, 0.4329, 0.7296, 0.2235, 0.3907, 0.2747, 0.6246, 0.5548],
  29366. [0.5967, 0.4017, 0.8822, 0.4524, 0.4647, 0.5588, 0.6056, 0.5269],
  29367. [0.6249, 0.4424, 0.8794, 0.4925, 0.3538, 0.3837, 0.5687, 0.5527],
  29368. [0.0318, 0.0366, 0.8491, 0.2514, 0.5056, 0.2510, 0.7441, 0.5336],
  29369. [0.6129, 0.4187, 0.8801, 0.5066, 0.4024, 0.4690, 0.5349, 0.5547],
  29370. [0.5513, 0.3737, 0.8868, 0.4114, 0.3565, 0.4728, 0.6038, 0.5205],
  29371. [0.6103, 0.4197, 0.8110, 0.4017, 0.3478, 0.3106, 0.5409, 0.5645]],
  29372. device='cuda:0', grad_fn=<AddmmBackward>)
  29373. landmarks are: tensor([[[0.6329, 0.4055, 0.9050, 0.4783, 0.3613, 0.3917, 0.6464, 0.5019],
  29374. [0.6218, 0.4137, 0.7262, 0.2233, 0.4075, 0.2650, 0.6212, 0.5783],
  29375. [0.6145, 0.4007, 0.8775, 0.4533, 0.4563, 0.5533, 0.6087, 0.5533],
  29376. [0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5638, 0.5633],
  29377. [0.0000, 0.0000, 0.8750, 0.2467, 0.5138, 0.2617, 0.7382, 0.5552],
  29378. [0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767],
  29379. [0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6137, 0.5400],
  29380. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896]]],
  29381. device='cuda:0')
  29382. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  29383. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  29384. loss_train: 0.021688233740860596
  29385. step: 46
  29386. running loss: 0.00047148334219262165
  29387.  
  29388. Train Steps: 46/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29389. torch.Size([8, 8])
  29390. tensor([[ nan, nan, 0.7268, 0.2333, 0.4125, 0.1933, 0.5112, 0.5383],
  29391. [ nan, nan, 0.8488, 0.2300, 0.5563, 0.2100, 0.7390, 0.5679],
  29392. [0.6332, 0.4165, 0.9100, 0.3350, 0.4188, 0.3683, 0.7438, 0.5528],
  29393. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  29394. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  29395. [0.6064, 0.4019, 0.8650, 0.4517, 0.4037, 0.5367, 0.5703, 0.5609],
  29396. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  29397. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317]],
  29398. device='cuda:0', dtype=torch.float64)
  29399. predictions are: tensor([[0.0505, 0.0629, 0.7221, 0.2172, 0.3990, 0.1907, 0.4836, 0.5072],
  29400. [0.0711, 0.0606, 0.8532, 0.2203, 0.5276, 0.2427, 0.7032, 0.5434],
  29401. [0.6273, 0.4385, 0.9236, 0.3451, 0.4138, 0.3583, 0.7116, 0.5344],
  29402. [0.6543, 0.4615, 0.8308, 0.5804, 0.3941, 0.4709, 0.5442, 0.5924],
  29403. [0.6330, 0.4136, 0.8432, 0.5356, 0.4166, 0.5505, 0.7111, 0.5536],
  29404. [0.6060, 0.4188, 0.8725, 0.4506, 0.4011, 0.5457, 0.5526, 0.5252],
  29405. [0.6341, 0.4384, 0.9139, 0.5257, 0.3681, 0.3829, 0.6115, 0.4767],
  29406. [0.6443, 0.4431, 0.8885, 0.5408, 0.3817, 0.3726, 0.5642, 0.5151]],
  29407. device='cuda:0', grad_fn=<AddmmBackward>)
  29408. landmarks are: tensor([[[0.0000, 0.0000, 0.7268, 0.2333, 0.4125, 0.1933, 0.5113, 0.5383],
  29409. [0.0000, 0.0000, 0.8487, 0.2300, 0.5562, 0.2100, 0.7390, 0.5679],
  29410. [0.6332, 0.4165, 0.9100, 0.3350, 0.4187, 0.3683, 0.7438, 0.5528],
  29411. [0.6274, 0.4117, 0.8100, 0.5801, 0.4000, 0.4583, 0.5582, 0.6118],
  29412. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690],
  29413. [0.6064, 0.4019, 0.8650, 0.4517, 0.4038, 0.5367, 0.5703, 0.5609],
  29414. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  29415. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317]]],
  29416. device='cuda:0')
  29417. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  29418. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  29419. loss_train: 0.022245831292821094
  29420. step: 47
  29421. running loss: 0.0004733155594217254
  29422. Train Steps: 47/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29423. torch.Size([8, 8])
  29424. tensor([[0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  29425. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  29426. [0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950],
  29427. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  29428. [0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297],
  29429. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  29430. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  29431. [0.6198, 0.4115, 0.7762, 0.2717, 0.3713, 0.3200, 0.5837, 0.5683]],
  29432. device='cuda:0', dtype=torch.float64)
  29433. predictions are: tensor([[0.5713, 0.3957, 0.8536, 0.5315, 0.3781, 0.5025, 0.6913, 0.5320],
  29434. [0.5430, 0.3901, 0.7948, 0.3414, 0.3455, 0.3305, 0.5656, 0.5079],
  29435. [0.5416, 0.3661, 0.8745, 0.5353, 0.4398, 0.4958, 0.5801, 0.4915],
  29436. [0.6007, 0.4279, 0.9035, 0.5517, 0.3764, 0.4135, 0.5504, 0.5678],
  29437. [0.6196, 0.4281, 0.8863, 0.5627, 0.3712, 0.4381, 0.6375, 0.5271],
  29438. [0.6077, 0.4280, 0.9010, 0.4555, 0.3897, 0.4254, 0.5202, 0.5663],
  29439. [0.5874, 0.4125, 0.7613, 0.2292, 0.3627, 0.3174, 0.5922, 0.5303],
  29440. [0.5564, 0.3966, 0.7779, 0.2761, 0.3650, 0.3021, 0.5742, 0.5564]],
  29441. device='cuda:0', grad_fn=<AddmmBackward>)
  29442. landmarks are: tensor([[[0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  29443. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  29444. [0.6209, 0.3920, 0.8650, 0.5367, 0.4400, 0.5067, 0.6025, 0.4950],
  29445. [0.6226, 0.4185, 0.8850, 0.5500, 0.3800, 0.4250, 0.5625, 0.5617],
  29446. [0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297],
  29447. [0.6137, 0.4035, 0.8850, 0.4417, 0.3900, 0.4283, 0.5449, 0.5617],
  29448. [0.6189, 0.3961, 0.7589, 0.2256, 0.3759, 0.3280, 0.6184, 0.5334],
  29449. [0.6198, 0.4115, 0.7763, 0.2717, 0.3713, 0.3200, 0.5838, 0.5683]]],
  29450. device='cuda:0')
  29451. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  29452. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  29453. loss_train: 0.02273709830478765
  29454. step: 48
  29455. running loss: 0.0004736895480164094
  29456. Train Steps: 48/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29457. torch.Size([8, 8])
  29458. tensor([[0.6286, 0.4086, 0.8408, 0.2801, 0.4163, 0.2800, 0.6725, 0.5393],
  29459. [0.6239, 0.4061, 0.8850, 0.4600, 0.4225, 0.5200, 0.6138, 0.5450],
  29460. [0.6246, 0.4008, 0.8757, 0.5088, 0.4101, 0.5392, 0.6644, 0.5133],
  29461. [0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  29462. [0.6262, 0.4052, 0.8888, 0.4700, 0.3675, 0.5117, 0.6350, 0.5233],
  29463. [0.6127, 0.4115, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  29464. [0.6206, 0.4001, 0.8900, 0.3933, 0.3588, 0.3567, 0.5837, 0.5083],
  29465. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459]],
  29466. device='cuda:0', dtype=torch.float64)
  29467. predictions are: tensor([[0.5733, 0.3989, 0.8587, 0.2864, 0.4182, 0.2777, 0.6666, 0.5564],
  29468. [0.5894, 0.4048, 0.8967, 0.4735, 0.4157, 0.5211, 0.6128, 0.5577],
  29469. [0.5565, 0.3758, 0.9014, 0.5329, 0.4087, 0.5500, 0.6531, 0.5286],
  29470. [0.5870, 0.3902, 0.8603, 0.6248, 0.4068, 0.4734, 0.6039, 0.4739],
  29471. [0.5898, 0.4098, 0.9070, 0.4832, 0.3771, 0.5000, 0.6268, 0.5352],
  29472. [0.5793, 0.4194, 0.7333, 0.2957, 0.3599, 0.2952, 0.5088, 0.5769],
  29473. [0.5803, 0.4017, 0.9111, 0.4044, 0.3606, 0.3576, 0.5664, 0.5195],
  29474. [0.5550, 0.3854, 0.8797, 0.2995, 0.4887, 0.2015, 0.6591, 0.5446]],
  29475. device='cuda:0', grad_fn=<AddmmBackward>)
  29476. landmarks are: tensor([[[0.6286, 0.4086, 0.8408, 0.2801, 0.4162, 0.2800, 0.6725, 0.5393],
  29477. [0.6239, 0.4061, 0.8850, 0.4600, 0.4225, 0.5200, 0.6137, 0.5450],
  29478. [0.6246, 0.4008, 0.8757, 0.5088, 0.4101, 0.5392, 0.6644, 0.5133],
  29479. [0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  29480. [0.6262, 0.4052, 0.8888, 0.4700, 0.3675, 0.5117, 0.6350, 0.5233],
  29481. [0.6127, 0.4114, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  29482. [0.6206, 0.4001, 0.8900, 0.3933, 0.3587, 0.3567, 0.5838, 0.5083],
  29483. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459]]],
  29484. device='cuda:0')
  29485. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  29486. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  29487. loss_train: 0.023219668451929465
  29488. step: 49
  29489. running loss: 0.0004738707847332544
  29490. Train Steps: 49/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29491. torch.Size([8, 8])
  29492. tensor([[0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  29493. [0.6200, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  29494. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  29495. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  29496. [0.6200, 0.4049, 0.8638, 0.5617, 0.4125, 0.5100, 0.6013, 0.5317],
  29497. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385],
  29498. [0.6226, 0.4103, 0.8575, 0.3450, 0.4388, 0.2067, 0.5787, 0.5383],
  29499. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117]],
  29500. device='cuda:0', dtype=torch.float64)
  29501. predictions are: tensor([[0.5143, 0.3352, 0.7564, 0.1950, 0.4566, 0.1685, 0.6002, 0.4866],
  29502. [0.6141, 0.3955, 0.8951, 0.4780, 0.3630, 0.4330, 0.5919, 0.5264],
  29503. [0.6263, 0.4158, 0.8824, 0.5108, 0.3786, 0.3296, 0.6314, 0.5002],
  29504. [0.6027, 0.3875, 0.8526, 0.3889, 0.3475, 0.4036, 0.6088, 0.5667],
  29505. [0.5613, 0.3617, 0.8650, 0.5623, 0.4191, 0.5066, 0.6014, 0.5359],
  29506. [0.5758, 0.3717, 0.8225, 0.2557, 0.4555, 0.2425, 0.6712, 0.5450],
  29507. [0.5658, 0.3698, 0.8539, 0.3462, 0.4375, 0.2139, 0.5617, 0.5394],
  29508. [0.5803, 0.3703, 0.8256, 0.5701, 0.4086, 0.4908, 0.5831, 0.6090]],
  29509. device='cuda:0', grad_fn=<AddmmBackward>)
  29510. landmarks are: tensor([[[0.6211, 0.3993, 0.7650, 0.1933, 0.4575, 0.1550, 0.5965, 0.4895],
  29511. [0.6201, 0.4039, 0.8880, 0.4799, 0.3625, 0.4285, 0.5866, 0.5148],
  29512. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  29513. [0.6196, 0.4090, 0.8450, 0.3900, 0.3575, 0.3933, 0.5975, 0.5583],
  29514. [0.6199, 0.4049, 0.8637, 0.5617, 0.4125, 0.5100, 0.6012, 0.5317],
  29515. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385],
  29516. [0.6226, 0.4103, 0.8575, 0.3450, 0.4387, 0.2067, 0.5788, 0.5383],
  29517. [0.6274, 0.4087, 0.8375, 0.5700, 0.4025, 0.4800, 0.5700, 0.6117]]],
  29518. device='cuda:0')
  29519. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  29520. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  29521. loss_train: 0.023799304181011394
  29522. step: 50
  29523. running loss: 0.0004759860836202279
  29524.  
  29525. Train Steps: 50/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29526. torch.Size([8, 8])
  29527. tensor([[0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6138, 0.5400],
  29528. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6188, 0.5283],
  29529. [0.6248, 0.4185, 0.8500, 0.5767, 0.4463, 0.4550, 0.5613, 0.5917],
  29530. [0.6336, 0.4086, 0.8900, 0.3950, 0.3900, 0.2950, 0.6504, 0.5066],
  29531. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  29532. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  29533. [0.6204, 0.4110, 0.7913, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367],
  29534. [0.6201, 0.4064, 0.8688, 0.5050, 0.4225, 0.5100, 0.6138, 0.5500]],
  29535. device='cuda:0', dtype=torch.float64)
  29536. predictions are: tensor([[0.6089, 0.3839, 0.8779, 0.4035, 0.3542, 0.4681, 0.6114, 0.5448],
  29537. [0.6044, 0.3894, 0.8943, 0.3586, 0.4072, 0.2674, 0.6360, 0.5405],
  29538. [0.5996, 0.3847, 0.8351, 0.5734, 0.4571, 0.4574, 0.5717, 0.5991],
  29539. [0.6905, 0.4306, 0.9103, 0.3863, 0.3879, 0.2945, 0.6742, 0.5114],
  29540. [0.6187, 0.3879, 0.8670, 0.5505, 0.3616, 0.3690, 0.5881, 0.5461],
  29541. [0.4900, 0.3153, 0.6603, 0.3017, 0.3806, 0.3047, 0.5667, 0.5835],
  29542. [0.6272, 0.4046, 0.7958, 0.2784, 0.4054, 0.2550, 0.6150, 0.5347],
  29543. [0.5972, 0.3814, 0.8602, 0.5074, 0.4322, 0.5232, 0.6212, 0.5594]],
  29544. device='cuda:0', grad_fn=<AddmmBackward>)
  29545. landmarks are: tensor([[[0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6137, 0.5400],
  29546. [0.6261, 0.4131, 0.8938, 0.3550, 0.4000, 0.2683, 0.6187, 0.5283],
  29547. [0.6248, 0.4185, 0.8500, 0.5767, 0.4462, 0.4550, 0.5612, 0.5917],
  29548. [0.6336, 0.4086, 0.8900, 0.3950, 0.3900, 0.2950, 0.6504, 0.5066],
  29549. [0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  29550. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  29551. [0.6204, 0.4110, 0.7912, 0.2667, 0.4062, 0.2500, 0.6225, 0.5367],
  29552. [0.6201, 0.4064, 0.8687, 0.5050, 0.4225, 0.5100, 0.6137, 0.5500]]],
  29553. device='cuda:0')
  29554. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  29555. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  29556. loss_train: 0.024365292192669585
  29557. step: 51
  29558. running loss: 0.0004777508273072468
  29559. Train Steps: 51/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29560. torch.Size([8, 8])
  29561. tensor([[0.6261, 0.4045, 0.8865, 0.5369, 0.3895, 0.4859, 0.6683, 0.5249],
  29562. [0.6364, 0.4165, 0.9088, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  29563. [0.6102, 0.4020, 0.8638, 0.3717, 0.3625, 0.5017, 0.6038, 0.5500],
  29564. [0.6263, 0.4030, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  29565. [0.6202, 0.4053, 0.8638, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  29566. [0.6265, 0.4251, 0.7113, 0.3550, 0.4375, 0.2117, 0.5587, 0.6118],
  29567. [0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921],
  29568. [0.6277, 0.4083, 0.8350, 0.2717, 0.4562, 0.1800, 0.5918, 0.4878]],
  29569. device='cuda:0', dtype=torch.float64)
  29570. predictions are: tensor([[0.6439, 0.4036, 0.8803, 0.5353, 0.3712, 0.4872, 0.6681, 0.5360],
  29571. [0.6116, 0.3821, 0.8909, 0.4335, 0.4142, 0.3066, 0.6441, 0.5334],
  29572. [0.5969, 0.3855, 0.8576, 0.3654, 0.3585, 0.4943, 0.6255, 0.5415],
  29573. [0.6176, 0.3881, 0.8856, 0.4795, 0.3741, 0.5054, 0.6522, 0.4894],
  29574. [0.6183, 0.3785, 0.8550, 0.5274, 0.4576, 0.5067, 0.5971, 0.5188],
  29575. [0.5650, 0.3853, 0.7067, 0.3604, 0.4218, 0.2029, 0.5690, 0.6241],
  29576. [0.6409, 0.4029, 0.8242, 0.2356, 0.3934, 0.2916, 0.6437, 0.5114],
  29577. [0.5672, 0.3577, 0.8080, 0.2659, 0.4671, 0.1860, 0.5837, 0.4921]],
  29578. device='cuda:0', grad_fn=<AddmmBackward>)
  29579. landmarks are: tensor([[[0.6261, 0.4045, 0.8865, 0.5369, 0.3895, 0.4859, 0.6683, 0.5249],
  29580. [0.6364, 0.4165, 0.9087, 0.4367, 0.4075, 0.3150, 0.6448, 0.5297],
  29581. [0.6102, 0.4020, 0.8637, 0.3717, 0.3625, 0.5017, 0.6037, 0.5500],
  29582. [0.6263, 0.4029, 0.9000, 0.4767, 0.3800, 0.5167, 0.6415, 0.4771],
  29583. [0.6202, 0.4053, 0.8637, 0.5283, 0.4546, 0.5108, 0.5900, 0.5133],
  29584. [0.6265, 0.4251, 0.7113, 0.3550, 0.4375, 0.2117, 0.5587, 0.6118],
  29585. [0.6266, 0.4101, 0.8350, 0.2333, 0.3950, 0.2950, 0.6264, 0.4921],
  29586. [0.6277, 0.4083, 0.8350, 0.2717, 0.4563, 0.1800, 0.5918, 0.4878]]],
  29587. device='cuda:0')
  29588. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  29589. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  29590. loss_train: 0.02468289455282502
  29591. step: 52
  29592. running loss: 0.00047467104909278883
  29593. Train Steps: 52/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29594. torch.Size([8, 8])
  29595. tensor([[0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  29596. [ nan, nan, 0.7225, 0.2167, 0.3987, 0.2283, 0.5427, 0.5181],
  29597. [0.6336, 0.4191, 0.8938, 0.5167, 0.3937, 0.3517, 0.7343, 0.5748],
  29598. [0.6250, 0.4131, 0.8688, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  29599. [0.6200, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913],
  29600. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  29601. [0.6156, 0.4125, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  29602. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200]],
  29603. device='cuda:0', dtype=torch.float64)
  29604. predictions are: tensor([[0.6829, 0.4331, 0.8681, 0.4453, 0.3672, 0.3511, 0.5824, 0.5813],
  29605. [0.0488, 0.0061, 0.7005, 0.2047, 0.4013, 0.2191, 0.5391, 0.5093],
  29606. [0.6749, 0.4260, 0.8833, 0.5126, 0.4011, 0.3471, 0.7377, 0.5690],
  29607. [0.6430, 0.4035, 0.8453, 0.2920, 0.4334, 0.2236, 0.6163, 0.5345],
  29608. [0.6359, 0.3727, 0.8428, 0.5143, 0.4045, 0.5073, 0.6094, 0.4836],
  29609. [0.6329, 0.4247, 0.7100, 0.2810, 0.4323, 0.2096, 0.5824, 0.5666],
  29610. [0.6682, 0.4187, 0.8638, 0.4755, 0.4442, 0.5742, 0.5900, 0.5133],
  29611. [0.6326, 0.3849, 0.8501, 0.2994, 0.4474, 0.2098, 0.6189, 0.5257]],
  29612. device='cuda:0', grad_fn=<AddmmBackward>)
  29613. landmarks are: tensor([[[0.6260, 0.4253, 0.8938, 0.4400, 0.3675, 0.3650, 0.5675, 0.5767],
  29614. [0.0000, 0.0000, 0.7225, 0.2167, 0.3988, 0.2283, 0.5427, 0.5181],
  29615. [0.6336, 0.4191, 0.8938, 0.5167, 0.3938, 0.3517, 0.7343, 0.5748],
  29616. [0.6250, 0.4131, 0.8687, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  29617. [0.6199, 0.3913, 0.8788, 0.5217, 0.4075, 0.5100, 0.6060, 0.4913],
  29618. [0.6214, 0.4175, 0.7300, 0.2883, 0.4338, 0.2167, 0.5698, 0.5773],
  29619. [0.6155, 0.4124, 0.8850, 0.4833, 0.4550, 0.5817, 0.5765, 0.5084],
  29620. [0.6267, 0.4094, 0.8712, 0.3083, 0.4400, 0.2267, 0.6250, 0.5200]]],
  29621. device='cuda:0')
  29622. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  29623. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  29624. loss_train: 0.024981963797472417
  29625. step: 53
  29626. running loss: 0.00047135780749947955
  29627. Train Steps: 53/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29628. torch.Size([8, 8])
  29629. tensor([[0.6175, 0.4091, 0.7863, 0.2800, 0.3638, 0.3583, 0.6188, 0.5433],
  29630. [0.6135, 0.4115, 0.8838, 0.4667, 0.4288, 0.6050, 0.5778, 0.5097],
  29631. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  29632. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  29633. [0.6127, 0.4115, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  29634. [ nan, nan, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  29635. [0.6164, 0.4119, 0.7913, 0.2650, 0.3538, 0.3500, 0.5614, 0.5038],
  29636. [0.6264, 0.4049, 0.8988, 0.4633, 0.3813, 0.4983, 0.6326, 0.4843]],
  29637. device='cuda:0', dtype=torch.float64)
  29638. predictions are: tensor([[ 0.6558, 0.4323, 0.7786, 0.2756, 0.3678, 0.3484, 0.6330, 0.5481],
  29639. [ 0.6911, 0.4423, 0.8657, 0.4683, 0.4272, 0.5781, 0.5810, 0.5236],
  29640. [ 0.6710, 0.4255, 0.8808, 0.4174, 0.3541, 0.3507, 0.6189, 0.5552],
  29641. [ 0.6651, 0.4252, 0.8548, 0.4724, 0.3746, 0.5103, 0.6280, 0.5095],
  29642. [ 0.6851, 0.4552, 0.7107, 0.2866, 0.3654, 0.2804, 0.5467, 0.5620],
  29643. [-0.0314, -0.0217, 0.8840, 0.3413, 0.5100, 0.1859, 0.7191, 0.5681],
  29644. [ 0.6652, 0.4315, 0.7833, 0.2691, 0.3491, 0.3422, 0.5961, 0.5102],
  29645. [ 0.6593, 0.4110, 0.8902, 0.4746, 0.3892, 0.4822, 0.6407, 0.4775]],
  29646. device='cuda:0', grad_fn=<AddmmBackward>)
  29647. landmarks are: tensor([[[0.6175, 0.4091, 0.7862, 0.2800, 0.3638, 0.3583, 0.6187, 0.5433],
  29648. [0.6135, 0.4115, 0.8838, 0.4667, 0.4288, 0.6050, 0.5778, 0.5097],
  29649. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  29650. [0.6132, 0.3930, 0.8672, 0.4754, 0.3712, 0.5222, 0.5974, 0.5098],
  29651. [0.6127, 0.4114, 0.7163, 0.2883, 0.3625, 0.2950, 0.5327, 0.5495],
  29652. [0.0000, 0.0000, 0.8900, 0.3217, 0.5038, 0.2233, 0.6694, 0.5609],
  29653. [0.6164, 0.4119, 0.7912, 0.2650, 0.3537, 0.3500, 0.5614, 0.5038],
  29654. [0.6264, 0.4049, 0.8988, 0.4633, 0.3812, 0.4983, 0.6326, 0.4843]]],
  29655. device='cuda:0')
  29656. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  29657. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  29658. loss_train: 0.025597798521630466
  29659. step: 54
  29660. running loss: 0.00047403330595611977
  29661.  
  29662. Train Steps: 54/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29663. torch.Size([8, 8])
  29664. tensor([[0.6127, 0.4084, 0.8700, 0.4467, 0.3987, 0.4317, 0.5013, 0.5471],
  29665. [0.6296, 0.4008, 0.9150, 0.4317, 0.4263, 0.3050, 0.7256, 0.5413],
  29666. [0.6192, 0.4128, 0.8513, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  29667. [ nan, nan, 0.7335, 0.2569, 0.3788, 0.2667, 0.5066, 0.5578],
  29668. [0.6109, 0.4015, 0.7668, 0.3639, 0.3513, 0.3667, 0.5200, 0.5641],
  29669. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  29670. [0.6226, 0.4098, 0.8912, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  29671. [0.6072, 0.4029, 0.7037, 0.2150, 0.3912, 0.2267, 0.5516, 0.5507]],
  29672. device='cuda:0', dtype=torch.float64)
  29673. predictions are: tensor([[ 0.6535, 0.4354, 0.8624, 0.4512, 0.3896, 0.4346, 0.5115, 0.5233],
  29674. [ 0.6379, 0.4002, 0.9202, 0.4149, 0.4190, 0.3058, 0.7154, 0.5308],
  29675. [ 0.6623, 0.4254, 0.8586, 0.5727, 0.4205, 0.5445, 0.6047, 0.5441],
  29676. [-0.0117, 0.0061, 0.7192, 0.2681, 0.3944, 0.2608, 0.5313, 0.5637],
  29677. [ 0.6767, 0.4417, 0.7500, 0.3647, 0.3468, 0.3627, 0.5505, 0.5424],
  29678. [ 0.6308, 0.4081, 0.8827, 0.4676, 0.3585, 0.3575, 0.6272, 0.4758],
  29679. [ 0.6622, 0.4342, 0.8869, 0.4127, 0.4088, 0.2478, 0.5990, 0.5257],
  29680. [ 0.6462, 0.4288, 0.6884, 0.2136, 0.3919, 0.2280, 0.5715, 0.5308]],
  29681. device='cuda:0', grad_fn=<AddmmBackward>)
  29682. landmarks are: tensor([[[0.6127, 0.4084, 0.8700, 0.4467, 0.3988, 0.4317, 0.5013, 0.5471],
  29683. [0.6296, 0.4008, 0.9150, 0.4317, 0.4263, 0.3050, 0.7256, 0.5413],
  29684. [0.6192, 0.4128, 0.8512, 0.5617, 0.4200, 0.5267, 0.5850, 0.5633],
  29685. [0.0000, 0.0000, 0.7335, 0.2569, 0.3787, 0.2667, 0.5066, 0.5578],
  29686. [0.6109, 0.4015, 0.7668, 0.3639, 0.3512, 0.3667, 0.5200, 0.5641],
  29687. [0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  29688. [0.6226, 0.4098, 0.8913, 0.4100, 0.4025, 0.2383, 0.5763, 0.5367],
  29689. [0.6072, 0.4029, 0.7038, 0.2150, 0.3913, 0.2267, 0.5516, 0.5507]]],
  29690. device='cuda:0')
  29691. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  29692. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  29693. loss_train: 0.02595448528882116
  29694. step: 55
  29695. running loss: 0.0004718997325240211
  29696. Train Steps: 55/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29697. torch.Size([8, 8])
  29698. tensor([[0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  29699. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  29700. [0.6277, 0.4057, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  29701. [0.6277, 0.4118, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  29702. [0.6147, 0.4026, 0.6600, 0.2467, 0.4088, 0.2150, 0.5489, 0.5773],
  29703. [0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  29704. [0.6102, 0.4005, 0.8688, 0.5100, 0.4813, 0.5400, 0.5404, 0.5064],
  29705. [0.6289, 0.4024, 0.9088, 0.4567, 0.3937, 0.5633, 0.7058, 0.5609]],
  29706. device='cuda:0', dtype=torch.float64)
  29707. predictions are: tensor([[0.6168, 0.4003, 0.8846, 0.4705, 0.4641, 0.5626, 0.5990, 0.5645],
  29708. [0.6576, 0.4380, 0.8644, 0.3584, 0.3492, 0.3211, 0.5855, 0.5366],
  29709. [0.6508, 0.4180, 0.8447, 0.2560, 0.4321, 0.1896, 0.6037, 0.4935],
  29710. [0.6707, 0.4402, 0.9167, 0.3831, 0.3854, 0.2592, 0.6097, 0.5180],
  29711. [0.6497, 0.4374, 0.6726, 0.2335, 0.3940, 0.2255, 0.5302, 0.5871],
  29712. [0.6735, 0.4251, 0.8556, 0.6107, 0.3848, 0.4843, 0.5976, 0.4758],
  29713. [0.6407, 0.4308, 0.8873, 0.5028, 0.4670, 0.5341, 0.5162, 0.5112],
  29714. [0.6604, 0.4275, 0.9052, 0.4598, 0.3730, 0.5711, 0.7126, 0.5736]],
  29715. device='cuda:0', grad_fn=<AddmmBackward>)
  29716. landmarks are: tensor([[[0.6173, 0.4013, 0.8713, 0.4640, 0.4800, 0.5700, 0.6025, 0.5600],
  29717. [0.6200, 0.3993, 0.8639, 0.3687, 0.3658, 0.3139, 0.6002, 0.5374],
  29718. [0.6277, 0.4056, 0.8300, 0.2650, 0.4363, 0.1850, 0.6140, 0.4823],
  29719. [0.6277, 0.4117, 0.8988, 0.3833, 0.3950, 0.2650, 0.6290, 0.4938],
  29720. [0.6147, 0.4026, 0.6600, 0.2467, 0.4087, 0.2150, 0.5489, 0.5773],
  29721. [0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  29722. [0.6102, 0.4005, 0.8687, 0.5100, 0.4812, 0.5400, 0.5404, 0.5064],
  29723. [0.6289, 0.4024, 0.9087, 0.4567, 0.3938, 0.5633, 0.7058, 0.5609]]],
  29724. device='cuda:0')
  29725. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  29726. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  29727. loss_train: 0.026293839560821652
  29728. step: 56
  29729. running loss: 0.0004695328493003866
  29730. Train Steps: 56/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29731. torch.Size([8, 8])
  29732. tensor([[0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5637, 0.5633],
  29733. [0.6261, 0.4066, 0.8325, 0.2150, 0.4763, 0.2667, 0.7002, 0.5633],
  29734. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  29735. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6138, 0.5333],
  29736. [ nan, nan, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  29737. [0.6266, 0.4070, 0.8712, 0.5600, 0.3713, 0.4783, 0.5775, 0.6100],
  29738. [0.6165, 0.4106, 0.7575, 0.1733, 0.3838, 0.2650, 0.5680, 0.5116],
  29739. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517]],
  29740. device='cuda:0', dtype=torch.float64)
  29741. predictions are: tensor([[0.6809, 0.4527, 0.8958, 0.5070, 0.3730, 0.3764, 0.5636, 0.5649],
  29742. [0.6687, 0.4385, 0.8463, 0.2286, 0.4897, 0.2646, 0.7110, 0.5588],
  29743. [0.6743, 0.4415, 0.8559, 0.4503, 0.3849, 0.4493, 0.5528, 0.5542],
  29744. [0.6415, 0.4214, 0.9230, 0.4754, 0.3986, 0.4939, 0.6279, 0.5129],
  29745. [0.0709, 0.0313, 0.8239, 0.2984, 0.3913, 0.2874, 0.5317, 0.5538],
  29746. [0.6992, 0.4456, 0.8838, 0.5829, 0.3848, 0.4956, 0.5796, 0.5727],
  29747. [0.6580, 0.4399, 0.7690, 0.1786, 0.3969, 0.2561, 0.5709, 0.4883],
  29748. [0.7060, 0.4708, 0.7031, 0.2578, 0.4217, 0.2420, 0.5853, 0.5384]],
  29749. device='cuda:0', grad_fn=<AddmmBackward>)
  29750. landmarks are: tensor([[[0.6227, 0.4193, 0.8838, 0.4933, 0.3663, 0.3733, 0.5638, 0.5633],
  29751. [0.6261, 0.4066, 0.8325, 0.2150, 0.4762, 0.2667, 0.7002, 0.5633],
  29752. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  29753. [0.6249, 0.4138, 0.9038, 0.4517, 0.3862, 0.4917, 0.6137, 0.5333],
  29754. [0.0000, 0.0000, 0.8213, 0.2700, 0.3775, 0.2817, 0.5425, 0.5533],
  29755. [0.6266, 0.4070, 0.8712, 0.5600, 0.3713, 0.4783, 0.5775, 0.6100],
  29756. [0.6165, 0.4106, 0.7575, 0.1733, 0.3837, 0.2650, 0.5680, 0.5116],
  29757. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517]]],
  29758. device='cuda:0')
  29759. loss_train_step before backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  29760. loss_train_step after backward: tensor(0.0008, device='cuda:0', grad_fn=<MseLossBackward>)
  29761. loss_train: 0.02706561255035922
  29762. step: 57
  29763. running loss: 0.0004748353079010389
  29764. Train Steps: 57/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29765. torch.Size([8, 8])
  29766. tensor([[ nan, nan, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  29767. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  29768. [0.6125, 0.4035, 0.7825, 0.3100, 0.3463, 0.4900, 0.5832, 0.5637],
  29769. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  29770. [0.6197, 0.4091, 0.8800, 0.4783, 0.3538, 0.4767, 0.5950, 0.5550],
  29771. [0.6127, 0.4066, 0.8550, 0.5567, 0.4662, 0.5141, 0.5070, 0.5412],
  29772. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  29773. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690]],
  29774. device='cuda:0', dtype=torch.float64)
  29775. predictions are: tensor([[0.0304, 0.0195, 0.8875, 0.2379, 0.5391, 0.2203, 0.7275, 0.5593],
  29776. [0.6255, 0.4086, 0.8686, 0.5094, 0.4391, 0.5338, 0.5083, 0.5156],
  29777. [0.6054, 0.4096, 0.7995, 0.3114, 0.3484, 0.4842, 0.5791, 0.5237],
  29778. [0.6546, 0.4381, 0.7531, 0.3264, 0.4934, 0.1947, 0.5435, 0.6094],
  29779. [0.6182, 0.4169, 0.8919, 0.4782, 0.3639, 0.4761, 0.5869, 0.5366],
  29780. [0.6371, 0.4349, 0.8747, 0.5479, 0.4648, 0.4981, 0.5032, 0.5233],
  29781. [0.6280, 0.4171, 0.7957, 0.3231, 0.3506, 0.3399, 0.5864, 0.4858],
  29782. [0.6626, 0.4187, 0.8417, 0.5371, 0.4203, 0.5553, 0.7206, 0.5507]],
  29783. device='cuda:0', grad_fn=<AddmmBackward>)
  29784. landmarks are: tensor([[[0.0000, 0.0000, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  29785. [0.6070, 0.3979, 0.8575, 0.5083, 0.4350, 0.5400, 0.5180, 0.5466],
  29786. [0.6125, 0.4035, 0.7825, 0.3100, 0.3462, 0.4900, 0.5832, 0.5637],
  29787. [0.6283, 0.4283, 0.7477, 0.3542, 0.5125, 0.1917, 0.5500, 0.6167],
  29788. [0.6197, 0.4091, 0.8800, 0.4783, 0.3537, 0.4767, 0.5950, 0.5550],
  29789. [0.6127, 0.4066, 0.8550, 0.5567, 0.4662, 0.5141, 0.5070, 0.5412],
  29790. [0.6164, 0.3972, 0.7818, 0.3381, 0.3599, 0.3387, 0.5880, 0.5153],
  29791. [0.6314, 0.4050, 0.8227, 0.5431, 0.4150, 0.5517, 0.7121, 0.5690]]],
  29792. device='cuda:0')
  29793. loss_train_step before backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  29794. loss_train_step after backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  29795. loss_train: 0.0273065606597811
  29796. step: 58
  29797. running loss: 0.00047080276999622583
  29798.  
  29799. Train Steps: 58/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29800. torch.Size([8, 8])
  29801. tensor([[0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  29802. [0.6203, 0.4073, 0.8189, 0.2398, 0.4400, 0.2054, 0.5929, 0.5501],
  29803. [0.6085, 0.4005, 0.8400, 0.4317, 0.3763, 0.4750, 0.5476, 0.5058],
  29804. [0.6201, 0.4017, 0.8871, 0.4621, 0.3517, 0.4675, 0.5999, 0.5106],
  29805. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795],
  29806. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  29807. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  29808. [0.6222, 0.4169, 0.8638, 0.5650, 0.4313, 0.4783, 0.5637, 0.5633]],
  29809. device='cuda:0', dtype=torch.float64)
  29810. predictions are: tensor([[0.5931, 0.4097, 0.8874, 0.4887, 0.3945, 0.4566, 0.4961, 0.5504],
  29811. [0.6113, 0.4209, 0.8224, 0.2378, 0.4481, 0.2116, 0.5767, 0.5303],
  29812. [0.6100, 0.4114, 0.8547, 0.4254, 0.3745, 0.4647, 0.5253, 0.5002],
  29813. [0.5575, 0.3701, 0.8966, 0.4368, 0.3579, 0.4684, 0.5695, 0.5236],
  29814. [0.6361, 0.4070, 0.8598, 0.5326, 0.4002, 0.5371, 0.7005, 0.5858],
  29815. [0.5959, 0.3916, 0.8418, 0.5647, 0.4152, 0.4711, 0.5487, 0.5289],
  29816. [0.6106, 0.4112, 0.8561, 0.2199, 0.4688, 0.2690, 0.6868, 0.5519],
  29817. [0.6009, 0.4062, 0.8732, 0.5490, 0.4295, 0.4702, 0.5407, 0.5784]],
  29818. device='cuda:0', grad_fn=<AddmmBackward>)
  29819. landmarks are: tensor([[[0.6146, 0.4124, 0.8800, 0.4867, 0.3800, 0.4500, 0.5106, 0.5524],
  29820. [0.6203, 0.4073, 0.8189, 0.2398, 0.4400, 0.2054, 0.5929, 0.5501],
  29821. [0.6084, 0.4005, 0.8400, 0.4317, 0.3762, 0.4750, 0.5476, 0.5058],
  29822. [0.6201, 0.4017, 0.8871, 0.4621, 0.3517, 0.4675, 0.5999, 0.5106],
  29823. [0.6325, 0.4066, 0.8438, 0.5350, 0.3925, 0.5267, 0.7113, 0.5795],
  29824. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267],
  29825. [0.6257, 0.4060, 0.8300, 0.2333, 0.4688, 0.2583, 0.7050, 0.5633],
  29826. [0.6222, 0.4169, 0.8637, 0.5650, 0.4313, 0.4783, 0.5638, 0.5633]]],
  29827. device='cuda:0')
  29828. loss_train_step before backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  29829. loss_train_step after backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  29830. loss_train: 0.027555422129807994
  29831. step: 59
  29832. running loss: 0.00046704105304759314
  29833. Train Steps: 59/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29834. torch.Size([8, 8])
  29835. tensor([[0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  29836. [0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274],
  29837. [0.6270, 0.4267, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  29838. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  29839. [0.6138, 0.4054, 0.8750, 0.4750, 0.4363, 0.5017, 0.5086, 0.5822],
  29840. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385],
  29841. [0.6142, 0.4127, 0.7575, 0.3067, 0.3438, 0.4383, 0.5778, 0.5207],
  29842. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411]],
  29843. device='cuda:0', dtype=torch.float64)
  29844. predictions are: tensor([[0.5805, 0.3695, 0.8528, 0.5969, 0.3916, 0.4936, 0.5851, 0.4726],
  29845. [0.6133, 0.4126, 0.8841, 0.3074, 0.4825, 0.2126, 0.6232, 0.5347],
  29846. [0.5386, 0.3626, 0.7153, 0.2838, 0.4685, 0.2080, 0.5480, 0.6325],
  29847. [0.5736, 0.3836, 0.8924, 0.5182, 0.4034, 0.5241, 0.5703, 0.5236],
  29848. [0.5928, 0.4034, 0.8754, 0.4717, 0.4411, 0.5010, 0.4946, 0.5865],
  29849. [0.6145, 0.3952, 0.8317, 0.2403, 0.4494, 0.2615, 0.6516, 0.5482],
  29850. [0.5552, 0.3814, 0.7775, 0.2858, 0.3488, 0.4323, 0.5621, 0.5333],
  29851. [0.5896, 0.3951, 0.8348, 0.2885, 0.4179, 0.2225, 0.5542, 0.5421]],
  29852. device='cuda:0', grad_fn=<AddmmBackward>)
  29853. landmarks are: tensor([[[0.6271, 0.4020, 0.8375, 0.6083, 0.3925, 0.4867, 0.6037, 0.4626],
  29854. [0.6364, 0.4144, 0.8625, 0.3083, 0.4913, 0.2000, 0.6448, 0.5274],
  29855. [0.6270, 0.4266, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  29856. [0.6211, 0.4069, 0.8750, 0.5117, 0.4150, 0.4900, 0.5875, 0.5167],
  29857. [0.6138, 0.4054, 0.8750, 0.4750, 0.4363, 0.5017, 0.5086, 0.5822],
  29858. [0.6286, 0.4097, 0.8107, 0.2414, 0.4425, 0.2483, 0.6745, 0.5385],
  29859. [0.6142, 0.4127, 0.7575, 0.3067, 0.3438, 0.4383, 0.5778, 0.5207],
  29860. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411]]],
  29861. device='cuda:0')
  29862. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  29863. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  29864. loss_train: 0.028138416208093986
  29865. step: 60
  29866. running loss: 0.0004689736034682331
  29867. Train Steps: 60/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29868. torch.Size([8, 8])
  29869. tensor([[0.6268, 0.4061, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  29870. [ nan, nan, 0.7412, 0.2200, 0.4450, 0.1517, 0.5312, 0.4983],
  29871. [0.6122, 0.4006, 0.8850, 0.4217, 0.4088, 0.5517, 0.6063, 0.5517],
  29872. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323],
  29873. [0.6198, 0.4101, 0.8838, 0.5283, 0.3763, 0.5267, 0.5913, 0.5567],
  29874. [0.6122, 0.3993, 0.8738, 0.4667, 0.4517, 0.4879, 0.5155, 0.4927],
  29875. [0.6229, 0.4066, 0.7612, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  29876. [0.6274, 0.4270, 0.8938, 0.4967, 0.3550, 0.4283, 0.5700, 0.5733]],
  29877. device='cuda:0', dtype=torch.float64)
  29878. predictions are: tensor([[ 0.6601, 0.4249, 0.8365, 0.2621, 0.4635, 0.2275, 0.6326, 0.5491],
  29879. [-0.0086, -0.0145, 0.7348, 0.2205, 0.4374, 0.1838, 0.5444, 0.5287],
  29880. [ 0.5796, 0.3725, 0.8702, 0.4454, 0.4151, 0.5646, 0.6027, 0.5698],
  29881. [ 0.6180, 0.3984, 0.7739, 0.2203, 0.4497, 0.2328, 0.6333, 0.5667],
  29882. [ 0.5950, 0.3945, 0.8707, 0.5517, 0.3807, 0.5406, 0.5929, 0.5902],
  29883. [ 0.5702, 0.3789, 0.8715, 0.4924, 0.4610, 0.5067, 0.5206, 0.5155],
  29884. [ 0.5951, 0.3871, 0.7522, 0.2959, 0.4304, 0.2437, 0.5927, 0.5768],
  29885. [ 0.5827, 0.3908, 0.8806, 0.5121, 0.3746, 0.4485, 0.5799, 0.5850]],
  29886. device='cuda:0', grad_fn=<AddmmBackward>)
  29887. landmarks are: tensor([[[0.6268, 0.4060, 0.8350, 0.2433, 0.4575, 0.2283, 0.6350, 0.5300],
  29888. [0.0000, 0.0000, 0.7412, 0.2200, 0.4450, 0.1517, 0.5312, 0.4983],
  29889. [0.6122, 0.4006, 0.8850, 0.4217, 0.4087, 0.5517, 0.6062, 0.5517],
  29890. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323],
  29891. [0.6198, 0.4101, 0.8838, 0.5283, 0.3762, 0.5267, 0.5913, 0.5567],
  29892. [0.6122, 0.3993, 0.8737, 0.4667, 0.4517, 0.4879, 0.5155, 0.4927],
  29893. [0.6229, 0.4066, 0.7613, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  29894. [0.6274, 0.4270, 0.8938, 0.4967, 0.3550, 0.4283, 0.5700, 0.5733]]],
  29895. device='cuda:0')
  29896. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  29897. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  29898. loss_train: 0.028534956247312948
  29899. step: 61
  29900. running loss: 0.00046778616798873684
  29901. Train Steps: 61/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29902. torch.Size([8, 8])
  29903. tensor([[0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  29904. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  29905. [0.6234, 0.4179, 0.7825, 0.3450, 0.3813, 0.2867, 0.5675, 0.5617],
  29906. [ nan, nan, 0.7425, 0.2117, 0.3937, 0.2433, 0.5438, 0.5567],
  29907. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  29908. [0.6250, 0.4131, 0.8688, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  29909. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  29910. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482]],
  29911. device='cuda:0', dtype=torch.float64)
  29912. predictions are: tensor([[0.5669, 0.3619, 0.8556, 0.5292, 0.4412, 0.5751, 0.6005, 0.5261],
  29913. [0.6083, 0.3899, 0.8528, 0.4225, 0.3767, 0.4475, 0.6131, 0.5482],
  29914. [0.5707, 0.3839, 0.7689, 0.3624, 0.4002, 0.2881, 0.5756, 0.5738],
  29915. [0.0472, 0.0215, 0.7341, 0.2395, 0.3982, 0.2607, 0.5287, 0.5618],
  29916. [0.6124, 0.3909, 0.8916, 0.4302, 0.3773, 0.3933, 0.6496, 0.5194],
  29917. [0.6038, 0.3976, 0.8544, 0.3207, 0.4519, 0.2539, 0.6068, 0.5451],
  29918. [0.6097, 0.3940, 0.8519, 0.4202, 0.4003, 0.4898, 0.5916, 0.5555],
  29919. [0.5586, 0.3608, 0.8751, 0.4247, 0.3675, 0.3782, 0.5871, 0.5745]],
  29920. device='cuda:0', grad_fn=<AddmmBackward>)
  29921. landmarks are: tensor([[[0.6129, 0.3930, 0.8769, 0.5105, 0.4146, 0.5642, 0.6000, 0.5082],
  29922. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  29923. [0.6234, 0.4179, 0.7825, 0.3450, 0.3812, 0.2867, 0.5675, 0.5617],
  29924. [0.0000, 0.0000, 0.7425, 0.2117, 0.3938, 0.2433, 0.5437, 0.5567],
  29925. [0.6296, 0.3989, 0.9000, 0.4150, 0.3613, 0.3867, 0.6400, 0.5100],
  29926. [0.6250, 0.4131, 0.8687, 0.2983, 0.4275, 0.2367, 0.6162, 0.5367],
  29927. [0.6150, 0.3949, 0.8800, 0.4033, 0.3825, 0.4900, 0.5875, 0.5483],
  29928. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482]]],
  29929. device='cuda:0')
  29930. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  29931. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  29932. loss_train: 0.028984240110730752
  29933. step: 62
  29934. running loss: 0.00046748774372146375
  29935.  
  29936. Train Steps: 62/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29937. torch.Size([8, 8])
  29938. tensor([[0.6126, 0.3954, 0.8538, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  29939. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  29940. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  29941. [0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  29942. [0.6068, 0.3963, 0.8650, 0.4317, 0.4037, 0.5083, 0.5253, 0.4999],
  29943. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6362, 0.5200],
  29944. [0.6286, 0.4086, 0.8408, 0.2801, 0.4163, 0.2800, 0.6725, 0.5393],
  29945. [0.6182, 0.4099, 0.7812, 0.3000, 0.3937, 0.2367, 0.5325, 0.5750]],
  29946. device='cuda:0', dtype=torch.float64)
  29947. predictions are: tensor([[0.5621, 0.3566, 0.8446, 0.5022, 0.4237, 0.4701, 0.5275, 0.5495],
  29948. [0.6145, 0.3894, 0.8666, 0.4223, 0.3447, 0.4410, 0.6383, 0.5094],
  29949. [0.5641, 0.3690, 0.8509, 0.5467, 0.3611, 0.4495, 0.6190, 0.5186],
  29950. [0.5908, 0.3902, 0.8255, 0.5633, 0.3920, 0.4061, 0.7199, 0.5706],
  29951. [0.6059, 0.3850, 0.8353, 0.4464, 0.4018, 0.5076, 0.5215, 0.4927],
  29952. [0.5895, 0.3806, 0.8615, 0.3665, 0.3606, 0.5174, 0.6403, 0.5171],
  29953. [0.6403, 0.4124, 0.8280, 0.2962, 0.4103, 0.2930, 0.6851, 0.5467],
  29954. [0.6094, 0.4098, 0.7718, 0.3154, 0.4044, 0.2346, 0.5285, 0.5751]],
  29955. device='cuda:0', grad_fn=<AddmmBackward>)
  29956. landmarks are: tensor([[[0.6126, 0.3954, 0.8537, 0.4983, 0.4250, 0.4700, 0.5355, 0.5350],
  29957. [0.6250, 0.4008, 0.8950, 0.4183, 0.3550, 0.4383, 0.6361, 0.4927],
  29958. [0.6302, 0.4007, 0.8850, 0.5500, 0.3613, 0.4517, 0.6335, 0.5012],
  29959. [0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  29960. [0.6068, 0.3963, 0.8650, 0.4317, 0.4038, 0.5083, 0.5253, 0.4999],
  29961. [0.6214, 0.4040, 0.8838, 0.3500, 0.3600, 0.5183, 0.6363, 0.5200],
  29962. [0.6286, 0.4086, 0.8408, 0.2801, 0.4162, 0.2800, 0.6725, 0.5393],
  29963. [0.6182, 0.4099, 0.7812, 0.3000, 0.3938, 0.2367, 0.5325, 0.5750]]],
  29964. device='cuda:0')
  29965. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  29966. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  29967. loss_train: 0.029316391359316185
  29968. step: 63
  29969. running loss: 0.0004653395453859712
  29970. Train Steps: 63/90 Loss: 0.0005 torch.Size([8, 600, 800])
  29971. torch.Size([8, 8])
  29972. tensor([[0.6147, 0.4081, 0.8538, 0.3400, 0.3663, 0.3150, 0.5142, 0.4875],
  29973. [0.6229, 0.4066, 0.7612, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  29974. [0.6163, 0.4006, 0.8788, 0.4683, 0.3663, 0.4883, 0.5887, 0.5017],
  29975. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  29976. [0.6111, 0.4019, 0.8350, 0.3867, 0.3500, 0.4283, 0.5480, 0.5719],
  29977. [0.6261, 0.4029, 0.8720, 0.3364, 0.3665, 0.3753, 0.6531, 0.5183],
  29978. [0.6097, 0.4024, 0.8488, 0.3717, 0.3875, 0.5517, 0.5836, 0.5591],
  29979. [0.6082, 0.4024, 0.8738, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136]],
  29980. device='cuda:0', dtype=torch.float64)
  29981. predictions are: tensor([[0.6310, 0.4094, 0.8501, 0.3549, 0.3542, 0.2914, 0.5601, 0.4993],
  29982. [0.6054, 0.3872, 0.7358, 0.2707, 0.4196, 0.2048, 0.6096, 0.5439],
  29983. [0.6054, 0.3931, 0.8595, 0.4581, 0.3593, 0.4813, 0.5988, 0.5172],
  29984. [0.5903, 0.3818, 0.8267, 0.5510, 0.3844, 0.4066, 0.5926, 0.5354],
  29985. [0.6094, 0.3951, 0.8160, 0.3809, 0.3438, 0.4054, 0.5520, 0.5542],
  29986. [0.6233, 0.3981, 0.8425, 0.3481, 0.3685, 0.3532, 0.6469, 0.5067],
  29987. [0.5660, 0.3539, 0.8270, 0.3669, 0.3908, 0.5373, 0.6171, 0.5336],
  29988. [0.5670, 0.3656, 0.8518, 0.3878, 0.3705, 0.3865, 0.5457, 0.4974]],
  29989. device='cuda:0', grad_fn=<AddmmBackward>)
  29990. landmarks are: tensor([[[0.6147, 0.4081, 0.8537, 0.3400, 0.3663, 0.3150, 0.5142, 0.4875],
  29991. [0.6229, 0.4066, 0.7613, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  29992. [0.6163, 0.4006, 0.8788, 0.4683, 0.3663, 0.4883, 0.5888, 0.5017],
  29993. [0.6226, 0.4001, 0.8438, 0.5733, 0.3862, 0.4250, 0.5750, 0.5283],
  29994. [0.6111, 0.4019, 0.8350, 0.3867, 0.3500, 0.4283, 0.5480, 0.5719],
  29995. [0.6261, 0.4029, 0.8720, 0.3364, 0.3665, 0.3753, 0.6531, 0.5183],
  29996. [0.6097, 0.4024, 0.8487, 0.3717, 0.3875, 0.5517, 0.5836, 0.5591],
  29997. [0.6082, 0.4024, 0.8737, 0.4017, 0.3688, 0.3950, 0.5306, 0.5136]]],
  29998. device='cuda:0')
  29999. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  30000. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  30001. loss_train: 0.029691978357732296
  30002. step: 64
  30003. running loss: 0.0004639371618395671
  30004. Train Steps: 64/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30005. torch.Size([8, 8])
  30006. tensor([[0.6329, 0.4196, 0.9238, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748],
  30007. [0.6075, 0.4007, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280],
  30008. [0.6189, 0.4049, 0.8888, 0.4417, 0.4213, 0.5200, 0.5988, 0.5633],
  30009. [0.6267, 0.4065, 0.8313, 0.2467, 0.4788, 0.1733, 0.6312, 0.5133],
  30010. [0.6339, 0.4102, 0.8588, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089],
  30011. [0.6198, 0.3997, 0.8582, 0.5361, 0.4117, 0.5016, 0.5942, 0.5134],
  30012. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  30013. [0.6135, 0.4115, 0.8838, 0.4667, 0.4288, 0.6050, 0.5778, 0.5097]],
  30014. device='cuda:0', dtype=torch.float64)
  30015. predictions are: tensor([[0.6060, 0.4027, 0.9025, 0.4582, 0.3996, 0.2778, 0.7325, 0.5489],
  30016. [0.6204, 0.4067, 0.8110, 0.4773, 0.3780, 0.4878, 0.5226, 0.5209],
  30017. [0.5794, 0.3736, 0.8835, 0.4338, 0.3948, 0.5220, 0.6071, 0.5618],
  30018. [0.6142, 0.4049, 0.8165, 0.2311, 0.4544, 0.1651, 0.6256, 0.5015],
  30019. [0.6073, 0.3961, 0.8408, 0.3007, 0.4270, 0.2088, 0.6565, 0.4918],
  30020. [0.5828, 0.3903, 0.8484, 0.5090, 0.3845, 0.5092, 0.5891, 0.4947],
  30021. [0.6020, 0.3937, 0.8769, 0.4152, 0.3334, 0.3141, 0.5947, 0.5279],
  30022. [0.5650, 0.3774, 0.8745, 0.4475, 0.3991, 0.5974, 0.5709, 0.5031]],
  30023. device='cuda:0', grad_fn=<AddmmBackward>)
  30024. landmarks are: tensor([[[0.6329, 0.4196, 0.9237, 0.4583, 0.4263, 0.2933, 0.7343, 0.5748],
  30025. [0.6075, 0.4006, 0.8275, 0.4917, 0.4050, 0.5100, 0.5167, 0.5280],
  30026. [0.6189, 0.4049, 0.8888, 0.4417, 0.4212, 0.5200, 0.5987, 0.5633],
  30027. [0.6266, 0.4065, 0.8313, 0.2467, 0.4787, 0.1733, 0.6313, 0.5133],
  30028. [0.6339, 0.4102, 0.8587, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089],
  30029. [0.6198, 0.3997, 0.8582, 0.5361, 0.4117, 0.5016, 0.5942, 0.5134],
  30030. [0.6202, 0.4054, 0.8892, 0.4428, 0.3642, 0.3249, 0.5840, 0.5426],
  30031. [0.6135, 0.4115, 0.8838, 0.4667, 0.4288, 0.6050, 0.5778, 0.5097]]],
  30032. device='cuda:0')
  30033. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  30034. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  30035. loss_train: 0.030056098359636962
  30036. step: 65
  30037. running loss: 0.000462401513225184
  30038. Train Steps: 65/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30039. torch.Size([8, 8])
  30040. tensor([[0.6201, 0.4017, 0.8871, 0.4621, 0.3517, 0.4675, 0.5999, 0.5106],
  30041. [0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6138, 0.5400],
  30042. [0.6219, 0.4097, 0.8738, 0.3400, 0.3563, 0.4117, 0.5975, 0.5683],
  30043. [0.6112, 0.4029, 0.8638, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  30044. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  30045. [0.6198, 0.4115, 0.7762, 0.2717, 0.3713, 0.3200, 0.5837, 0.5683],
  30046. [0.6137, 0.4038, 0.8563, 0.4050, 0.3813, 0.2550, 0.5106, 0.4954],
  30047. [0.6228, 0.4119, 0.7938, 0.2233, 0.4674, 0.1773, 0.6188, 0.5433]],
  30048. device='cuda:0', dtype=torch.float64)
  30049. predictions are: tensor([[0.6325, 0.4188, 0.8790, 0.4454, 0.3357, 0.4470, 0.5937, 0.5070],
  30050. [0.6276, 0.4159, 0.8831, 0.3919, 0.3389, 0.4420, 0.6121, 0.5288],
  30051. [0.5823, 0.4039, 0.8797, 0.3453, 0.3316, 0.3823, 0.6009, 0.5428],
  30052. [0.6027, 0.3982, 0.8725, 0.4607, 0.4668, 0.4718, 0.5631, 0.5381],
  30053. [0.5903, 0.3948, 0.8057, 0.2344, 0.4171, 0.2533, 0.6727, 0.5183],
  30054. [0.6415, 0.4335, 0.7790, 0.2666, 0.3548, 0.2899, 0.5766, 0.5454],
  30055. [0.6410, 0.4316, 0.8549, 0.4109, 0.3794, 0.2458, 0.5019, 0.4821],
  30056. [0.6050, 0.4117, 0.7940, 0.2302, 0.4581, 0.1521, 0.6242, 0.5132]],
  30057. device='cuda:0', grad_fn=<AddmmBackward>)
  30058. landmarks are: tensor([[[0.6201, 0.4017, 0.8871, 0.4621, 0.3517, 0.4675, 0.5999, 0.5106],
  30059. [0.6200, 0.3998, 0.8850, 0.3950, 0.3500, 0.4650, 0.6137, 0.5400],
  30060. [0.6219, 0.4097, 0.8737, 0.3400, 0.3562, 0.4117, 0.5975, 0.5683],
  30061. [0.6112, 0.4029, 0.8637, 0.4800, 0.4875, 0.5083, 0.5763, 0.5567],
  30062. [0.6279, 0.4050, 0.8025, 0.2200, 0.4313, 0.2733, 0.6820, 0.5320],
  30063. [0.6198, 0.4115, 0.7763, 0.2717, 0.3713, 0.3200, 0.5838, 0.5683],
  30064. [0.6137, 0.4038, 0.8562, 0.4050, 0.3812, 0.2550, 0.5106, 0.4954],
  30065. [0.6228, 0.4119, 0.7937, 0.2233, 0.4674, 0.1773, 0.6187, 0.5433]]],
  30066. device='cuda:0')
  30067. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  30068. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  30069. loss_train: 0.030345589388161898
  30070. step: 66
  30071. running loss: 0.0004597816573963924
  30072.  
  30073. Train Steps: 66/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30074. torch.Size([8, 8])
  30075. tensor([[0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  30076. [0.6173, 0.4114, 0.7325, 0.2500, 0.4213, 0.1917, 0.5338, 0.5700],
  30077. [0.6058, 0.3978, 0.8287, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461],
  30078. [0.6182, 0.3930, 0.8841, 0.3892, 0.3556, 0.4967, 0.6222, 0.5279],
  30079. [ nan, nan, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  30080. [0.6300, 0.4133, 0.8538, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  30081. [0.6141, 0.4038, 0.8650, 0.4833, 0.4839, 0.5176, 0.5787, 0.5600],
  30082. [ nan, nan, 0.7625, 0.2433, 0.3713, 0.2867, 0.5235, 0.5220]],
  30083. device='cuda:0', dtype=torch.float64)
  30084. predictions are: tensor([[0.6730, 0.4373, 0.9066, 0.5274, 0.3590, 0.4573, 0.6407, 0.4906],
  30085. [0.6628, 0.4471, 0.7531, 0.2596, 0.4204, 0.1885, 0.5333, 0.5555],
  30086. [0.6626, 0.4496, 0.8338, 0.3639, 0.3408, 0.4037, 0.5647, 0.5396],
  30087. [0.6586, 0.4142, 0.8963, 0.4003, 0.3388, 0.4788, 0.6314, 0.5172],
  30088. [0.0469, 0.0414, 0.8846, 0.2554, 0.5381, 0.2105, 0.7217, 0.5614],
  30089. [0.6127, 0.3974, 0.8840, 0.2309, 0.5523, 0.2475, 0.7283, 0.5359],
  30090. [0.6800, 0.4512, 0.8846, 0.4860, 0.4886, 0.5012, 0.5734, 0.5537],
  30091. [0.0506, 0.0346, 0.7934, 0.2526, 0.3742, 0.2824, 0.5256, 0.5156]],
  30092. device='cuda:0', grad_fn=<AddmmBackward>)
  30093. landmarks are: tensor([[[0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  30094. [0.6173, 0.4114, 0.7325, 0.2500, 0.4212, 0.1917, 0.5337, 0.5700],
  30095. [0.6058, 0.3978, 0.8288, 0.3600, 0.3400, 0.4050, 0.5529, 0.5461],
  30096. [0.6182, 0.3930, 0.8841, 0.3892, 0.3556, 0.4967, 0.6222, 0.5279],
  30097. [0.0000, 0.0000, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  30098. [0.6300, 0.4133, 0.8537, 0.2167, 0.5587, 0.2250, 0.7390, 0.5413],
  30099. [0.6141, 0.4038, 0.8650, 0.4833, 0.4839, 0.5176, 0.5788, 0.5600],
  30100. [0.0000, 0.0000, 0.7625, 0.2433, 0.3713, 0.2867, 0.5235, 0.5220]]],
  30101. device='cuda:0')
  30102. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  30103. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  30104. loss_train: 0.03091353503987193
  30105. step: 67
  30106. running loss: 0.0004613960453712229
  30107. Train Steps: 67/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30108. torch.Size([8, 8])
  30109. tensor([[0.6162, 0.4134, 0.6700, 0.2467, 0.3962, 0.2533, 0.5737, 0.5467],
  30110. [0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  30111. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6063, 0.5617],
  30112. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  30113. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  30114. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  30115. [0.6339, 0.4102, 0.9088, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  30116. [0.6204, 0.4055, 0.8438, 0.5733, 0.4574, 0.4801, 0.5487, 0.5617]],
  30117. device='cuda:0', dtype=torch.float64)
  30118. predictions are: tensor([[0.6072, 0.4156, 0.6845, 0.2488, 0.3885, 0.2646, 0.5822, 0.5506],
  30119. [0.6824, 0.4430, 0.9033, 0.3734, 0.3597, 0.3630, 0.6155, 0.4861],
  30120. [0.6485, 0.4382, 0.9266, 0.4681, 0.3643, 0.4627, 0.5905, 0.5646],
  30121. [0.6429, 0.4136, 0.9213, 0.3295, 0.4761, 0.3300, 0.7252, 0.5254],
  30122. [0.6812, 0.4458, 0.8530, 0.3079, 0.3447, 0.4176, 0.6092, 0.5309],
  30123. [0.6150, 0.4112, 0.8791, 0.5605, 0.3931, 0.4561, 0.5007, 0.4900],
  30124. [0.6644, 0.4421, 0.9267, 0.4800, 0.3993, 0.5388, 0.7369, 0.5406],
  30125. [0.6674, 0.4306, 0.8511, 0.5840, 0.4571, 0.4663, 0.5463, 0.5880]],
  30126. device='cuda:0', grad_fn=<AddmmBackward>)
  30127. landmarks are: tensor([[[0.6162, 0.4134, 0.6700, 0.2467, 0.3963, 0.2533, 0.5738, 0.5467],
  30128. [0.6263, 0.4057, 0.8800, 0.3833, 0.3650, 0.3717, 0.6375, 0.4804],
  30129. [0.6204, 0.4091, 0.8950, 0.4783, 0.3613, 0.4617, 0.6062, 0.5617],
  30130. [0.6296, 0.4060, 0.9100, 0.3267, 0.4726, 0.3367, 0.7446, 0.5355],
  30131. [0.6182, 0.3967, 0.8263, 0.3065, 0.3526, 0.4161, 0.6192, 0.5290],
  30132. [0.6101, 0.3977, 0.8550, 0.5667, 0.3862, 0.4417, 0.5128, 0.4869],
  30133. [0.6339, 0.4102, 0.9087, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  30134. [0.6204, 0.4055, 0.8438, 0.5733, 0.4574, 0.4801, 0.5487, 0.5617]]],
  30135. device='cuda:0')
  30136. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  30137. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  30138. loss_train: 0.03130897891242057
  30139. step: 68
  30140. running loss: 0.0004604261604767731
  30141. Train Steps: 68/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30142. torch.Size([8, 8])
  30143. tensor([[0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6088, 0.5583],
  30144. [0.6241, 0.4143, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  30145. [0.6239, 0.4174, 0.8425, 0.5733, 0.4825, 0.4500, 0.5625, 0.5933],
  30146. [0.6239, 0.4206, 0.8750, 0.5400, 0.3688, 0.4850, 0.5737, 0.5700],
  30147. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600],
  30148. [0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  30149. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  30150. [0.6183, 0.4076, 0.8838, 0.4517, 0.3813, 0.4483, 0.5775, 0.5633]],
  30151. device='cuda:0', dtype=torch.float64)
  30152. predictions are: tensor([[0.6401, 0.4268, 0.7226, 0.2012, 0.4088, 0.2647, 0.6165, 0.5383],
  30153. [0.6213, 0.4168, 0.9257, 0.4448, 0.4271, 0.5342, 0.6339, 0.5451],
  30154. [0.6416, 0.4361, 0.8674, 0.5556, 0.4946, 0.4437, 0.5999, 0.5797],
  30155. [0.6356, 0.4275, 0.9069, 0.5226, 0.3884, 0.4892, 0.5855, 0.5514],
  30156. [0.6988, 0.4739, 0.9074, 0.5321, 0.4143, 0.4907, 0.5857, 0.5554],
  30157. [0.6401, 0.4200, 0.8829, 0.5230, 0.4142, 0.4624, 0.6204, 0.5285],
  30158. [0.5978, 0.3988, 0.7610, 0.2069, 0.4505, 0.2119, 0.6215, 0.5091],
  30159. [0.6463, 0.4257, 0.9085, 0.4362, 0.4016, 0.4561, 0.5870, 0.5596]],
  30160. device='cuda:0', grad_fn=<AddmmBackward>)
  30161. landmarks are: tensor([[[0.6246, 0.4090, 0.6964, 0.2027, 0.3925, 0.2683, 0.6087, 0.5583],
  30162. [0.6241, 0.4142, 0.8938, 0.4650, 0.4075, 0.5350, 0.6250, 0.5550],
  30163. [0.6239, 0.4174, 0.8425, 0.5733, 0.4825, 0.4500, 0.5625, 0.5933],
  30164. [0.6239, 0.4206, 0.8750, 0.5400, 0.3688, 0.4850, 0.5738, 0.5700],
  30165. [0.6223, 0.4171, 0.8750, 0.5500, 0.4050, 0.4817, 0.5675, 0.5600],
  30166. [0.6199, 0.3952, 0.8573, 0.5374, 0.4075, 0.4687, 0.5942, 0.5377],
  30167. [0.6216, 0.4100, 0.7350, 0.2067, 0.4325, 0.2050, 0.5950, 0.5333],
  30168. [0.6183, 0.4076, 0.8838, 0.4517, 0.3812, 0.4483, 0.5775, 0.5633]]],
  30169. device='cuda:0')
  30170. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  30171. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  30172. loss_train: 0.031762625847477466
  30173. step: 69
  30174. running loss: 0.00046032791083300674
  30175. Train Steps: 69/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30176. torch.Size([8, 8])
  30177. tensor([[0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  30178. [0.6200, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  30179. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  30180. [0.6215, 0.4119, 0.7688, 0.2300, 0.4200, 0.2283, 0.5925, 0.5317],
  30181. [0.6151, 0.4058, 0.7068, 0.2680, 0.3400, 0.4083, 0.5775, 0.5733],
  30182. [0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  30183. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  30184. [0.6109, 0.4041, 0.6975, 0.3167, 0.3513, 0.3383, 0.5153, 0.5319]],
  30185. device='cuda:0', dtype=torch.float64)
  30186. predictions are: tensor([[0.6460, 0.4203, 0.9000, 0.5420, 0.3821, 0.3687, 0.5508, 0.5364],
  30187. [0.6151, 0.4056, 0.8694, 0.2938, 0.4477, 0.2250, 0.5831, 0.5344],
  30188. [0.5969, 0.3859, 0.7443, 0.1821, 0.4409, 0.2535, 0.6052, 0.5541],
  30189. [0.5619, 0.3696, 0.7780, 0.2227, 0.4654, 0.2409, 0.6000, 0.5386],
  30190. [0.6365, 0.4150, 0.7459, 0.2567, 0.3699, 0.4079, 0.5746, 0.5771],
  30191. [0.6522, 0.4290, 0.8946, 0.5280, 0.4635, 0.5333, 0.6704, 0.5245],
  30192. [0.6457, 0.4102, 0.9029, 0.4656, 0.4647, 0.5256, 0.6216, 0.5263],
  30193. [0.6640, 0.4348, 0.7400, 0.3104, 0.3679, 0.3367, 0.5077, 0.5379]],
  30194. device='cuda:0', grad_fn=<AddmmBackward>)
  30195. landmarks are: tensor([[[0.6182, 0.3972, 0.8720, 0.5527, 0.3638, 0.3582, 0.5696, 0.5395],
  30196. [0.6199, 0.4086, 0.8414, 0.2974, 0.4117, 0.2274, 0.5869, 0.5391],
  30197. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  30198. [0.6215, 0.4119, 0.7688, 0.2300, 0.4200, 0.2283, 0.5925, 0.5317],
  30199. [0.6151, 0.4058, 0.7068, 0.2680, 0.3400, 0.4083, 0.5775, 0.5733],
  30200. [0.6257, 0.4024, 0.8612, 0.5352, 0.4361, 0.5253, 0.6680, 0.5166],
  30201. [0.6211, 0.3935, 0.8636, 0.4841, 0.4417, 0.5126, 0.6331, 0.5268],
  30202. [0.6109, 0.4041, 0.6975, 0.3167, 0.3512, 0.3383, 0.5153, 0.5319]]],
  30203. device='cuda:0')
  30204. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  30205. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  30206. loss_train: 0.03225781302899122
  30207. step: 70
  30208. running loss: 0.0004608259004141603
  30209.  
  30210. Train Steps: 70/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30211. torch.Size([8, 8])
  30212. tensor([[0.6310, 0.4017, 0.8563, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006],
  30213. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  30214. [0.6142, 0.4127, 0.7575, 0.3067, 0.3438, 0.4383, 0.5778, 0.5207],
  30215. [0.6227, 0.4083, 0.8938, 0.4800, 0.3800, 0.2950, 0.5737, 0.5350],
  30216. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  30217. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6038, 0.6167],
  30218. [0.6200, 0.4049, 0.8638, 0.5617, 0.4125, 0.5100, 0.6013, 0.5317],
  30219. [0.6201, 0.4102, 0.7288, 0.2417, 0.4150, 0.2383, 0.6100, 0.5500]],
  30220. device='cuda:0', dtype=torch.float64)
  30221. predictions are: tensor([[0.6148, 0.3716, 0.8605, 0.5585, 0.3871, 0.4737, 0.6259, 0.5169],
  30222. [0.6468, 0.4023, 0.8991, 0.4805, 0.4321, 0.5419, 0.6411, 0.5354],
  30223. [0.6433, 0.4191, 0.7664, 0.3155, 0.3697, 0.4264, 0.5722, 0.5404],
  30224. [0.6585, 0.4212, 0.8732, 0.4792, 0.3988, 0.3011, 0.5767, 0.5431],
  30225. [0.6219, 0.3866, 0.9016, 0.4525, 0.4625, 0.5749, 0.6231, 0.5492],
  30226. [0.6424, 0.4062, 0.7986, 0.3269, 0.3891, 0.3860, 0.6075, 0.6323],
  30227. [0.6266, 0.4012, 0.8681, 0.5553, 0.4246, 0.4999, 0.5951, 0.5453],
  30228. [0.6402, 0.4050, 0.7469, 0.2468, 0.4276, 0.2472, 0.5946, 0.5733]],
  30229. device='cuda:0', grad_fn=<AddmmBackward>)
  30230. landmarks are: tensor([[[0.6310, 0.4017, 0.8562, 0.5800, 0.3688, 0.4750, 0.6326, 0.5006],
  30231. [0.6197, 0.3930, 0.8793, 0.4736, 0.4152, 0.5464, 0.6308, 0.5235],
  30232. [0.6142, 0.4127, 0.7575, 0.3067, 0.3438, 0.4383, 0.5778, 0.5207],
  30233. [0.6227, 0.4083, 0.8938, 0.4800, 0.3800, 0.2950, 0.5738, 0.5350],
  30234. [0.6175, 0.4013, 0.8900, 0.4500, 0.4375, 0.5850, 0.6175, 0.5383],
  30235. [0.6198, 0.4105, 0.7950, 0.3267, 0.3675, 0.3767, 0.6037, 0.6167],
  30236. [0.6199, 0.4049, 0.8637, 0.5617, 0.4125, 0.5100, 0.6012, 0.5317],
  30237. [0.6201, 0.4102, 0.7287, 0.2417, 0.4150, 0.2383, 0.6100, 0.5500]]],
  30238. device='cuda:0')
  30239. loss_train_step before backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  30240. loss_train_step after backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  30241. loss_train: 0.032472519553266466
  30242. step: 71
  30243. running loss: 0.0004573594303276967
  30244. Train Steps: 71/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30245. torch.Size([8, 8])
  30246. tensor([[0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  30247. [0.6277, 0.4083, 0.8350, 0.2717, 0.4562, 0.1800, 0.5918, 0.4878],
  30248. [0.6339, 0.4102, 0.9088, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  30249. [0.6218, 0.4185, 0.7338, 0.2650, 0.4625, 0.1950, 0.5687, 0.5800],
  30250. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  30251. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  30252. [0.6193, 0.4050, 0.7313, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  30253. [0.6090, 0.4010, 0.7838, 0.3483, 0.3538, 0.3783, 0.5462, 0.5077]],
  30254. device='cuda:0', dtype=torch.float64)
  30255. predictions are: tensor([[0.5940, 0.3746, 0.8675, 0.4779, 0.3429, 0.3852, 0.6126, 0.4978],
  30256. [0.6416, 0.4011, 0.7967, 0.2772, 0.4555, 0.2147, 0.5898, 0.5085],
  30257. [0.6110, 0.3931, 0.8839, 0.4816, 0.3927, 0.5575, 0.7458, 0.5450],
  30258. [0.5519, 0.3622, 0.6993, 0.2598, 0.4502, 0.2145, 0.5542, 0.5974],
  30259. [0.6182, 0.4113, 0.8660, 0.4374, 0.4177, 0.5403, 0.5742, 0.5472],
  30260. [0.6018, 0.4055, 0.8068, 0.4443, 0.4412, 0.2809, 0.5477, 0.6292],
  30261. [0.5763, 0.3665, 0.7078, 0.2500, 0.4051, 0.2317, 0.5698, 0.5809],
  30262. [0.5822, 0.3735, 0.7703, 0.3540, 0.3481, 0.3917, 0.5402, 0.5193]],
  30263. device='cuda:0', grad_fn=<AddmmBackward>)
  30264. landmarks are: tensor([[[0.6277, 0.4013, 0.8888, 0.4767, 0.3600, 0.3567, 0.6148, 0.4672],
  30265. [0.6277, 0.4083, 0.8350, 0.2717, 0.4563, 0.1800, 0.5918, 0.4878],
  30266. [0.6339, 0.4102, 0.9087, 0.4767, 0.3925, 0.5283, 0.7509, 0.5390],
  30267. [0.6218, 0.4185, 0.7337, 0.2650, 0.4625, 0.1950, 0.5688, 0.5800],
  30268. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  30269. [0.6286, 0.4274, 0.8500, 0.4500, 0.4525, 0.2583, 0.5440, 0.6209],
  30270. [0.6193, 0.4050, 0.7312, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  30271. [0.6090, 0.4010, 0.7837, 0.3483, 0.3537, 0.3783, 0.5462, 0.5077]]],
  30272. device='cuda:0')
  30273. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  30274. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  30275. loss_train: 0.03298366616945714
  30276. step: 72
  30277. running loss: 0.0004581064745757936
  30278. Train Steps: 72/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30279. torch.Size([8, 8])
  30280. tensor([[0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533],
  30281. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667],
  30282. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  30283. [0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5413, 0.5433],
  30284. [0.6109, 0.4041, 0.6975, 0.3167, 0.3513, 0.3383, 0.5153, 0.5319],
  30285. [0.6109, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  30286. [0.6097, 0.4024, 0.8488, 0.3717, 0.3875, 0.5517, 0.5836, 0.5591],
  30287. [0.6219, 0.4097, 0.8738, 0.3400, 0.3563, 0.4117, 0.5975, 0.5683]],
  30288. device='cuda:0', dtype=torch.float64)
  30289. predictions are: tensor([[0.6499, 0.4078, 0.7000, 0.2581, 0.3769, 0.3111, 0.6247, 0.5697],
  30290. [0.6281, 0.4074, 0.8035, 0.3519, 0.3447, 0.3648, 0.5826, 0.5726],
  30291. [0.6364, 0.4119, 0.8539, 0.5459, 0.3686, 0.3947, 0.6423, 0.4841],
  30292. [0.6095, 0.3974, 0.8234, 0.3717, 0.3595, 0.3201, 0.5477, 0.5440],
  30293. [0.6284, 0.3995, 0.6904, 0.3434, 0.3354, 0.3422, 0.5280, 0.5376],
  30294. [0.5998, 0.3851, 0.8500, 0.4988, 0.3631, 0.4099, 0.5674, 0.5175],
  30295. [0.6158, 0.3904, 0.8237, 0.4010, 0.3881, 0.5619, 0.6200, 0.5523],
  30296. [0.6299, 0.4122, 0.8502, 0.3714, 0.3427, 0.4069, 0.6090, 0.5711]],
  30297. device='cuda:0', grad_fn=<AddmmBackward>)
  30298. landmarks are: tensor([[[0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533],
  30299. [0.6198, 0.4114, 0.8263, 0.3283, 0.3550, 0.3583, 0.5813, 0.5667],
  30300. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  30301. [0.6225, 0.4196, 0.8788, 0.3467, 0.3750, 0.3400, 0.5412, 0.5433],
  30302. [0.6109, 0.4041, 0.6975, 0.3167, 0.3512, 0.3383, 0.5153, 0.5319],
  30303. [0.6108, 0.4009, 0.8725, 0.4700, 0.3550, 0.4167, 0.5650, 0.5117],
  30304. [0.6097, 0.4024, 0.8487, 0.3717, 0.3875, 0.5517, 0.5836, 0.5591],
  30305. [0.6219, 0.4097, 0.8737, 0.3400, 0.3562, 0.4117, 0.5975, 0.5683]]],
  30306. device='cuda:0')
  30307. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  30308. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  30309. loss_train: 0.033311496314127
  30310. step: 73
  30311. running loss: 0.0004563218673168082
  30312. Train Steps: 73/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30313. torch.Size([8, 8])
  30314. tensor([[0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  30315. [0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510],
  30316. [ nan, nan, 0.7648, 0.2722, 0.3962, 0.2183, 0.5060, 0.5422],
  30317. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  30318. [0.6129, 0.4069, 0.8750, 0.5067, 0.3875, 0.4233, 0.5235, 0.5881],
  30319. [0.6278, 0.4253, 0.8875, 0.5017, 0.4113, 0.2750, 0.5413, 0.6196],
  30320. [0.6296, 0.4076, 0.8400, 0.5583, 0.3700, 0.4367, 0.6876, 0.5494],
  30321. [0.6166, 0.4008, 0.8563, 0.5667, 0.4388, 0.4933, 0.5575, 0.5567]],
  30322. device='cuda:0', dtype=torch.float64)
  30323. predictions are: tensor([[0.6001, 0.3739, 0.8487, 0.4550, 0.4385, 0.5025, 0.5561, 0.5098],
  30324. [0.6396, 0.3958, 0.6744, 0.2837, 0.3477, 0.3114, 0.5517, 0.5619],
  30325. [0.0268, 0.0122, 0.7260, 0.2481, 0.3726, 0.2221, 0.5261, 0.5384],
  30326. [0.6357, 0.3967, 0.8064, 0.5874, 0.3649, 0.5221, 0.6275, 0.4840],
  30327. [0.5734, 0.3575, 0.8358, 0.5049, 0.3809, 0.4387, 0.5359, 0.5817],
  30328. [0.6598, 0.4303, 0.8430, 0.4853, 0.4221, 0.2899, 0.5880, 0.6172],
  30329. [0.6129, 0.3875, 0.8133, 0.5460, 0.3682, 0.4424, 0.7284, 0.5532],
  30330. [0.6262, 0.3910, 0.8388, 0.5480, 0.4184, 0.5080, 0.5845, 0.5708]],
  30331. device='cuda:0', grad_fn=<AddmmBackward>)
  30332. landmarks are: tensor([[[0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960],
  30333. [0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510],
  30334. [0.0000, 0.0000, 0.7648, 0.2722, 0.3963, 0.2183, 0.5060, 0.5422],
  30335. [0.6271, 0.4005, 0.8450, 0.6067, 0.3850, 0.4983, 0.6069, 0.4649],
  30336. [0.6129, 0.4069, 0.8750, 0.5067, 0.3875, 0.4233, 0.5235, 0.5881],
  30337. [0.6278, 0.4253, 0.8875, 0.5017, 0.4112, 0.2750, 0.5413, 0.6196],
  30338. [0.6296, 0.4076, 0.8400, 0.5583, 0.3700, 0.4367, 0.6876, 0.5494],
  30339. [0.6166, 0.4008, 0.8562, 0.5667, 0.4387, 0.4933, 0.5575, 0.5567]]],
  30340. device='cuda:0')
  30341. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  30342. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  30343. loss_train: 0.03379115313873626
  30344. step: 74
  30345. running loss: 0.00045663720457751706
  30346.  
  30347. Train Steps: 74/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30348. torch.Size([8, 8])
  30349. tensor([[0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  30350. [0.6201, 0.4151, 0.8588, 0.5467, 0.3700, 0.3950, 0.5637, 0.5933],
  30351. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  30352. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167],
  30353. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791],
  30354. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  30355. [0.6307, 0.4029, 0.8650, 0.5200, 0.3763, 0.4017, 0.7311, 0.5366],
  30356. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986]],
  30357. device='cuda:0', dtype=torch.float64)
  30358. predictions are: tensor([[0.6089, 0.3922, 0.7377, 0.3172, 0.3326, 0.3313, 0.5328, 0.5695],
  30359. [0.6255, 0.4055, 0.8288, 0.5502, 0.3399, 0.3870, 0.5390, 0.5697],
  30360. [0.6152, 0.4073, 0.6935, 0.2694, 0.3570, 0.2763, 0.5415, 0.5174],
  30361. [0.5788, 0.3750, 0.7982, 0.3405, 0.3481, 0.3064, 0.5433, 0.5249],
  30362. [0.6074, 0.3984, 0.8524, 0.4886, 0.3761, 0.5244, 0.6152, 0.4800],
  30363. [0.6091, 0.3986, 0.8160, 0.4387, 0.3483, 0.4726, 0.5366, 0.5699],
  30364. [0.6254, 0.4090, 0.8349, 0.5263, 0.3610, 0.4049, 0.7104, 0.5167],
  30365. [0.6412, 0.4223, 0.8794, 0.3954, 0.4289, 0.2125, 0.5985, 0.4951]],
  30366. device='cuda:0', grad_fn=<AddmmBackward>)
  30367. landmarks are: tensor([[[0.6122, 0.4048, 0.7506, 0.3014, 0.3475, 0.3333, 0.5487, 0.5749],
  30368. [0.6202, 0.4151, 0.8587, 0.5467, 0.3700, 0.3950, 0.5638, 0.5933],
  30369. [0.6106, 0.4022, 0.7125, 0.2650, 0.3713, 0.2700, 0.5431, 0.5123],
  30370. [0.6136, 0.4029, 0.8263, 0.3350, 0.3625, 0.3067, 0.5675, 0.5167],
  30371. [0.6264, 0.4035, 0.8888, 0.4883, 0.4050, 0.5217, 0.6361, 0.4791],
  30372. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  30373. [0.6307, 0.4029, 0.8650, 0.5200, 0.3762, 0.4017, 0.7311, 0.5366],
  30374. [0.6260, 0.4161, 0.9000, 0.3833, 0.4450, 0.2133, 0.6237, 0.4986]]],
  30375. device='cuda:0')
  30376. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  30377. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  30378. loss_train: 0.03405617133830674
  30379. step: 75
  30380. running loss: 0.00045408228451075655
  30381. Train Steps: 75/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30382. torch.Size([8, 8])
  30383. tensor([[0.6102, 0.4001, 0.7738, 0.3583, 0.3463, 0.3800, 0.5524, 0.5689],
  30384. [0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  30385. [0.6293, 0.3982, 0.8700, 0.5300, 0.3763, 0.4717, 0.7050, 0.5297],
  30386. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  30387. [0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  30388. [0.6163, 0.4114, 0.7650, 0.2017, 0.3763, 0.2867, 0.5631, 0.5071],
  30389. [0.6097, 0.3988, 0.8650, 0.5250, 0.4213, 0.5200, 0.5675, 0.5050],
  30390. [0.6199, 0.4060, 0.8888, 0.4667, 0.3800, 0.5050, 0.6188, 0.5433]],
  30391. device='cuda:0', dtype=torch.float64)
  30392. predictions are: tensor([[0.6041, 0.3908, 0.7807, 0.3757, 0.3441, 0.3718, 0.5173, 0.5518],
  30393. [0.5819, 0.3833, 0.8373, 0.4099, 0.3521, 0.4983, 0.5535, 0.5542],
  30394. [0.6308, 0.3993, 0.8683, 0.5513, 0.3575, 0.4532, 0.6896, 0.5235],
  30395. [0.6291, 0.4198, 0.8922, 0.5122, 0.3483, 0.4118, 0.6168, 0.5032],
  30396. [0.6506, 0.4303, 0.8431, 0.5729, 0.3834, 0.3998, 0.7119, 0.5599],
  30397. [0.6258, 0.4162, 0.7446, 0.2005, 0.3655, 0.2652, 0.5402, 0.5004],
  30398. [0.6046, 0.4005, 0.8637, 0.5245, 0.4355, 0.5057, 0.5201, 0.4984],
  30399. [0.6070, 0.4025, 0.8941, 0.4713, 0.3727, 0.5104, 0.5979, 0.5312]],
  30400. device='cuda:0', grad_fn=<AddmmBackward>)
  30401. landmarks are: tensor([[[0.6102, 0.4001, 0.7738, 0.3583, 0.3462, 0.3800, 0.5524, 0.5689],
  30402. [0.6037, 0.4020, 0.8300, 0.4033, 0.3575, 0.4883, 0.5647, 0.5631],
  30403. [0.6293, 0.3982, 0.8700, 0.5300, 0.3762, 0.4717, 0.7050, 0.5297],
  30404. [0.6269, 0.4073, 0.8900, 0.4933, 0.3625, 0.4183, 0.6288, 0.5150],
  30405. [0.6339, 0.4159, 0.8400, 0.5617, 0.3825, 0.4150, 0.7343, 0.5748],
  30406. [0.6163, 0.4114, 0.7650, 0.2017, 0.3762, 0.2867, 0.5631, 0.5071],
  30407. [0.6097, 0.3988, 0.8650, 0.5250, 0.4212, 0.5200, 0.5675, 0.5050],
  30408. [0.6199, 0.4060, 0.8888, 0.4667, 0.3800, 0.5050, 0.6187, 0.5433]]],
  30409. device='cuda:0')
  30410. loss_train_step before backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  30411. loss_train_step after backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  30412. loss_train: 0.03425430311472155
  30413. step: 76
  30414. running loss: 0.00045071451466738885
  30415. Train Steps: 76/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30416. torch.Size([8, 8])
  30417. tensor([[0.6133, 0.4094, 0.8495, 0.4028, 0.3588, 0.3200, 0.5003, 0.5407],
  30418. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  30419. [0.6122, 0.3993, 0.8738, 0.4667, 0.4517, 0.4879, 0.5155, 0.4927],
  30420. [0.6275, 0.4111, 0.8463, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  30421. [0.6272, 0.4071, 0.8738, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  30422. [ nan, nan, 0.7625, 0.2433, 0.3713, 0.2867, 0.5235, 0.5220],
  30423. [0.6102, 0.4020, 0.8638, 0.3717, 0.3625, 0.5017, 0.6038, 0.5500],
  30424. [0.6339, 0.4102, 0.8588, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089]],
  30425. device='cuda:0', dtype=torch.float64)
  30426. predictions are: tensor([[ 0.5782, 0.3960, 0.8596, 0.4026, 0.3638, 0.2947, 0.5104, 0.5605],
  30427. [ 0.5868, 0.3921, 0.8928, 0.5126, 0.3761, 0.3818, 0.6223, 0.5027],
  30428. [ 0.5819, 0.3889, 0.8771, 0.4751, 0.4446, 0.4962, 0.5090, 0.5008],
  30429. [ 0.6055, 0.4110, 0.8534, 0.2530, 0.4695, 0.1872, 0.6282, 0.5183],
  30430. [ 0.5992, 0.3971, 0.8742, 0.5657, 0.3681, 0.3804, 0.6038, 0.4837],
  30431. [-0.0291, -0.0051, 0.7582, 0.2469, 0.3822, 0.2729, 0.5228, 0.5483],
  30432. [ 0.5652, 0.3814, 0.8680, 0.3559, 0.3709, 0.4956, 0.6065, 0.5715],
  30433. [ 0.6148, 0.4079, 0.8644, 0.3119, 0.4575, 0.2078, 0.6471, 0.5255]],
  30434. device='cuda:0', grad_fn=<AddmmBackward>)
  30435. landmarks are: tensor([[[0.6133, 0.4094, 0.8495, 0.4028, 0.3587, 0.3200, 0.5003, 0.5407],
  30436. [0.6307, 0.4060, 0.8950, 0.5183, 0.3750, 0.3850, 0.6338, 0.4938],
  30437. [0.6122, 0.3993, 0.8737, 0.4667, 0.4517, 0.4879, 0.5155, 0.4927],
  30438. [0.6275, 0.4111, 0.8462, 0.2617, 0.4487, 0.1983, 0.6211, 0.4904],
  30439. [0.6272, 0.4071, 0.8737, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  30440. [0.0000, 0.0000, 0.7625, 0.2433, 0.3713, 0.2867, 0.5235, 0.5220],
  30441. [0.6102, 0.4020, 0.8637, 0.3717, 0.3625, 0.5017, 0.6037, 0.5500],
  30442. [0.6339, 0.4102, 0.8587, 0.3133, 0.4425, 0.2117, 0.6417, 0.5089]]],
  30443. device='cuda:0')
  30444. loss_train_step before backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  30445. loss_train_step after backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  30446. loss_train: 0.03449908675975166
  30447. step: 77
  30448. running loss: 0.00044804008778898265
  30449. Train Steps: 77/90 Loss: 0.0004 torch.Size([8, 600, 800])
  30450. torch.Size([8, 8])
  30451. tensor([[0.6346, 0.4144, 0.9088, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  30452. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  30453. [0.6262, 0.4052, 0.8888, 0.4700, 0.3675, 0.5117, 0.6350, 0.5233],
  30454. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  30455. [0.6161, 0.4024, 0.8662, 0.4683, 0.4935, 0.5364, 0.6063, 0.5567],
  30456. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  30457. [0.6346, 0.4092, 0.7712, 0.5917, 0.4037, 0.4767, 0.7343, 0.5725],
  30458. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960]],
  30459. device='cuda:0', dtype=torch.float64)
  30460. predictions are: tensor([[0.5928, 0.4148, 0.9317, 0.4549, 0.3857, 0.4069, 0.6875, 0.5749],
  30461. [0.6258, 0.4227, 0.8987, 0.4046, 0.3453, 0.4238, 0.5952, 0.5166],
  30462. [0.5839, 0.4023, 0.9205, 0.4549, 0.3527, 0.4937, 0.6135, 0.5055],
  30463. [0.6059, 0.4120, 0.7293, 0.2632, 0.4128, 0.1986, 0.5330, 0.5595],
  30464. [0.6360, 0.4297, 0.9069, 0.4480, 0.4612, 0.5162, 0.5732, 0.5343],
  30465. [0.5389, 0.3747, 0.9047, 0.5002, 0.3684, 0.3171, 0.6194, 0.4796],
  30466. [0.5922, 0.4071, 0.8178, 0.5521, 0.3789, 0.4638, 0.7159, 0.5648],
  30467. [0.5700, 0.3894, 0.9051, 0.4533, 0.4485, 0.4864, 0.5142, 0.4808]],
  30468. device='cuda:0', grad_fn=<AddmmBackward>)
  30469. landmarks are: tensor([[[0.6346, 0.4144, 0.9087, 0.4667, 0.3850, 0.4333, 0.7121, 0.5899],
  30470. [0.6182, 0.3998, 0.8793, 0.4191, 0.3552, 0.4285, 0.6038, 0.5312],
  30471. [0.6262, 0.4052, 0.8888, 0.4700, 0.3675, 0.5117, 0.6350, 0.5233],
  30472. [0.6192, 0.3980, 0.7078, 0.2750, 0.4250, 0.2100, 0.5450, 0.5783],
  30473. [0.6161, 0.4024, 0.8662, 0.4683, 0.4935, 0.5364, 0.6062, 0.5567],
  30474. [0.6260, 0.4133, 0.8800, 0.5117, 0.3713, 0.3283, 0.6223, 0.4967],
  30475. [0.6346, 0.4092, 0.7713, 0.5917, 0.4038, 0.4767, 0.7343, 0.5725],
  30476. [0.6076, 0.3958, 0.8700, 0.4667, 0.4546, 0.5046, 0.5231, 0.4960]]],
  30477. device='cuda:0')
  30478. loss_train_step before backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  30479. loss_train_step after backward: tensor(0.0006, device='cuda:0', grad_fn=<MseLossBackward>)
  30480. loss_train: 0.03509918806958012
  30481. step: 78
  30482. running loss: 0.0004499895906356426
  30483.  
  30484. Train Steps: 78/90 Loss: 0.0004 torch.Size([8, 600, 800])
  30485. torch.Size([8, 8])
  30486. tensor([[0.6202, 0.4064, 0.7879, 0.2179, 0.4567, 0.1725, 0.5955, 0.5478],
  30487. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  30488. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  30489. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  30490. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  30491. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  30492. [0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717],
  30493. [0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510]],
  30494. device='cuda:0', dtype=torch.float64)
  30495. predictions are: tensor([[0.5432, 0.3973, 0.8100, 0.2312, 0.4705, 0.1608, 0.5825, 0.5160],
  30496. [0.5358, 0.3683, 0.9406, 0.3140, 0.4232, 0.4218, 0.7012, 0.5485],
  30497. [0.5776, 0.4162, 0.9353, 0.4523, 0.4308, 0.5172, 0.5749, 0.5158],
  30498. [0.6260, 0.4445, 0.7304, 0.2638, 0.4196, 0.2368, 0.5704, 0.5261],
  30499. [0.5931, 0.4030, 0.9394, 0.4244, 0.3533, 0.3649, 0.5951, 0.5182],
  30500. [0.6915, 0.4722, 0.7127, 0.3191, 0.3733, 0.2816, 0.5586, 0.5610],
  30501. [0.5933, 0.4062, 0.7941, 0.2518, 0.3930, 0.2678, 0.5787, 0.5510],
  30502. [0.5903, 0.4067, 0.7309, 0.2871, 0.3600, 0.3128, 0.5212, 0.5276]],
  30503. device='cuda:0', grad_fn=<AddmmBackward>)
  30504. landmarks are: tensor([[[0.6202, 0.4064, 0.7879, 0.2179, 0.4567, 0.1725, 0.5955, 0.5478],
  30505. [0.6282, 0.4029, 0.8988, 0.3000, 0.4250, 0.4183, 0.7042, 0.5575],
  30506. [0.6140, 0.4034, 0.8850, 0.4317, 0.4288, 0.5067, 0.5825, 0.5533],
  30507. [0.6186, 0.4154, 0.6825, 0.2633, 0.4150, 0.2300, 0.5713, 0.5517],
  30508. [0.6203, 0.4056, 0.8942, 0.4086, 0.3643, 0.3617, 0.5917, 0.5482],
  30509. [0.6164, 0.4066, 0.6625, 0.3033, 0.3775, 0.2967, 0.5725, 0.5833],
  30510. [0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717],
  30511. [0.6042, 0.3990, 0.6831, 0.2875, 0.3500, 0.3133, 0.5143, 0.5510]]],
  30512. device='cuda:0')
  30513. loss_train_step before backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  30514. loss_train_step after backward: tensor(0.0009, device='cuda:0', grad_fn=<MseLossBackward>)
  30515. loss_train: 0.03596113619278185
  30516. step: 79
  30517. running loss: 0.00045520425560483354
  30518. Train Steps: 79/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30519. torch.Size([8, 8])
  30520. tensor([[ nan, nan, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  30521. [0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  30522. [0.6064, 0.4019, 0.8650, 0.4517, 0.4037, 0.5367, 0.5703, 0.5609],
  30523. [0.6190, 0.4135, 0.8000, 0.4883, 0.3566, 0.3647, 0.5613, 0.5900],
  30524. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  30525. [0.6276, 0.4002, 0.8800, 0.5533, 0.3575, 0.4400, 0.6132, 0.4672],
  30526. [0.6097, 0.3988, 0.8650, 0.5250, 0.4213, 0.5200, 0.5675, 0.5050],
  30527. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317]],
  30528. device='cuda:0', dtype=torch.float64)
  30529. predictions are: tensor([[-0.0010, 0.0065, 0.7765, 0.2641, 0.4116, 0.2566, 0.5234, 0.5575],
  30530. [ 0.6268, 0.4027, 0.8630, 0.4974, 0.4012, 0.5111, 0.7375, 0.5369],
  30531. [ 0.6158, 0.4085, 0.8687, 0.4403, 0.4261, 0.5473, 0.5952, 0.5478],
  30532. [ 0.6377, 0.4170, 0.8245, 0.4601, 0.3776, 0.3475, 0.5361, 0.5637],
  30533. [ 0.6432, 0.4277, 0.8856, 0.2673, 0.4200, 0.2645, 0.6101, 0.5431],
  30534. [ 0.6187, 0.3922, 0.8841, 0.5153, 0.3864, 0.4361, 0.6395, 0.4873],
  30535. [ 0.6301, 0.4144, 0.8737, 0.4948, 0.4597, 0.5032, 0.5711, 0.5104],
  30536. [ 0.6202, 0.4080, 0.8791, 0.5169, 0.3921, 0.3731, 0.5952, 0.5370]],
  30537. device='cuda:0', grad_fn=<AddmmBackward>)
  30538. landmarks are: tensor([[[0.0000, 0.0000, 0.7553, 0.2722, 0.3875, 0.2550, 0.5125, 0.5633],
  30539. [0.6307, 0.3998, 0.8500, 0.5233, 0.3850, 0.5050, 0.7446, 0.5320],
  30540. [0.6064, 0.4019, 0.8650, 0.4517, 0.4038, 0.5367, 0.5703, 0.5609],
  30541. [0.6190, 0.4135, 0.8000, 0.4883, 0.3566, 0.3647, 0.5612, 0.5900],
  30542. [0.6203, 0.4076, 0.8611, 0.2878, 0.4050, 0.2554, 0.5907, 0.5496],
  30543. [0.6276, 0.4002, 0.8800, 0.5533, 0.3575, 0.4400, 0.6132, 0.4672],
  30544. [0.6097, 0.3988, 0.8650, 0.5250, 0.4212, 0.5200, 0.5675, 0.5050],
  30545. [0.6227, 0.4049, 0.8750, 0.5367, 0.3775, 0.3667, 0.5725, 0.5317]]],
  30546. device='cuda:0')
  30547. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  30548. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  30549. loss_train: 0.036256922205211595
  30550. step: 80
  30551. running loss: 0.0004532115275651449
  30552. Train Steps: 80/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30553. torch.Size([8, 8])
  30554. tensor([[0.6179, 0.3993, 0.8925, 0.4789, 0.3879, 0.4900, 0.6041, 0.5279],
  30555. [0.6140, 0.4070, 0.8700, 0.5000, 0.4612, 0.4900, 0.5260, 0.5852],
  30556. [0.6257, 0.4167, 0.8775, 0.3433, 0.3563, 0.4133, 0.6200, 0.5667],
  30557. [0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297],
  30558. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  30559. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  30560. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  30561. [0.6284, 0.4093, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167]],
  30562. device='cuda:0', dtype=torch.float64)
  30563. predictions are: tensor([[0.6646, 0.4309, 0.8815, 0.4630, 0.3912, 0.5001, 0.6282, 0.5300],
  30564. [0.6050, 0.4034, 0.8733, 0.5086, 0.4680, 0.4959, 0.5481, 0.5852],
  30565. [0.6617, 0.4315, 0.8732, 0.3614, 0.3578, 0.4102, 0.6455, 0.5599],
  30566. [0.6401, 0.4181, 0.8733, 0.5595, 0.3670, 0.4489, 0.6568, 0.5374],
  30567. [0.6406, 0.4236, 0.8144, 0.3135, 0.4189, 0.2087, 0.5796, 0.5340],
  30568. [0.6158, 0.4083, 0.8343, 0.4385, 0.3714, 0.4626, 0.5720, 0.5615],
  30569. [0.6516, 0.4322, 0.7191, 0.2065, 0.4086, 0.2620, 0.6223, 0.5469],
  30570. [0.6441, 0.4177, 0.8827, 0.4689, 0.3794, 0.3867, 0.6357, 0.5137]],
  30571. device='cuda:0', grad_fn=<AddmmBackward>)
  30572. landmarks are: tensor([[[0.6179, 0.3993, 0.8925, 0.4789, 0.3879, 0.4900, 0.6041, 0.5279],
  30573. [0.6140, 0.4070, 0.8700, 0.5000, 0.4613, 0.4900, 0.5260, 0.5852],
  30574. [0.6257, 0.4167, 0.8775, 0.3433, 0.3562, 0.4133, 0.6200, 0.5667],
  30575. [0.6357, 0.4159, 0.8788, 0.5583, 0.3638, 0.4433, 0.6488, 0.5297],
  30576. [0.6186, 0.4013, 0.8191, 0.3188, 0.4279, 0.2060, 0.5767, 0.5411],
  30577. [0.6093, 0.3990, 0.8400, 0.4333, 0.3688, 0.4633, 0.5560, 0.5656],
  30578. [0.6201, 0.4065, 0.7300, 0.1933, 0.4075, 0.2533, 0.6162, 0.5483],
  30579. [0.6284, 0.4092, 0.8900, 0.4700, 0.3650, 0.3850, 0.6212, 0.5167]]],
  30580. device='cuda:0')
  30581. loss_train_step before backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  30582. loss_train_step after backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  30583. loss_train: 0.03644900632207282
  30584. step: 81
  30585. running loss: 0.00044998773237126937
  30586. Train Steps: 81/90 Loss: 0.0004 torch.Size([8, 600, 800])
  30587. torch.Size([8, 8])
  30588. tensor([[0.6305, 0.3983, 0.8950, 0.4833, 0.3688, 0.4683, 0.6375, 0.5117],
  30589. [0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5988, 0.5700],
  30590. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  30591. [0.6179, 0.4118, 0.7278, 0.4237, 0.3588, 0.3400, 0.5675, 0.5917],
  30592. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323],
  30593. [0.6193, 0.4050, 0.7313, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  30594. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  30595. [0.6186, 0.4060, 0.8750, 0.5050, 0.3538, 0.4367, 0.5813, 0.6083]],
  30596. device='cuda:0', dtype=torch.float64)
  30597. predictions are: tensor([[0.6277, 0.3878, 0.9010, 0.5158, 0.3749, 0.4779, 0.6465, 0.5020],
  30598. [0.6288, 0.4082, 0.7888, 0.2897, 0.3721, 0.3737, 0.5910, 0.5788],
  30599. [0.5866, 0.3920, 0.6858, 0.2162, 0.4286, 0.2067, 0.5621, 0.5650],
  30600. [0.6369, 0.4128, 0.7630, 0.4370, 0.3602, 0.3508, 0.5537, 0.5989],
  30601. [0.6528, 0.4111, 0.7752, 0.2209, 0.4564, 0.2176, 0.6471, 0.5461],
  30602. [0.6481, 0.4258, 0.7230, 0.2635, 0.4205, 0.2282, 0.5779, 0.5720],
  30603. [0.6336, 0.3955, 0.8779, 0.5571, 0.3745, 0.4803, 0.6392, 0.4987],
  30604. [0.6238, 0.4054, 0.8646, 0.5141, 0.3662, 0.4475, 0.5779, 0.6036]],
  30605. device='cuda:0', grad_fn=<AddmmBackward>)
  30606. landmarks are: tensor([[[0.6305, 0.3983, 0.8950, 0.4833, 0.3688, 0.4683, 0.6375, 0.5117],
  30607. [0.6239, 0.4107, 0.8162, 0.2763, 0.3625, 0.3600, 0.5987, 0.5700],
  30608. [0.6082, 0.4042, 0.6975, 0.1917, 0.4100, 0.1983, 0.5502, 0.5578],
  30609. [0.6179, 0.4118, 0.7278, 0.4237, 0.3587, 0.3400, 0.5675, 0.5917],
  30610. [0.6282, 0.4034, 0.7830, 0.2080, 0.4532, 0.2080, 0.6404, 0.5323],
  30611. [0.6193, 0.4050, 0.7312, 0.2433, 0.4075, 0.2117, 0.5649, 0.5656],
  30612. [0.6300, 0.4013, 0.8938, 0.5350, 0.3675, 0.4600, 0.6456, 0.4973],
  30613. [0.6186, 0.4060, 0.8750, 0.5050, 0.3537, 0.4367, 0.5813, 0.6083]]],
  30614. device='cuda:0')
  30615. loss_train_step before backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  30616. loss_train_step after backward: tensor(0.0002, device='cuda:0', grad_fn=<MseLossBackward>)
  30617. loss_train: 0.036643293395172805
  30618. step: 82
  30619. running loss: 0.00044686943164844883
  30620.  
  30621. Train Steps: 82/90 Loss: 0.0004 torch.Size([8, 600, 800])
  30622. torch.Size([8, 8])
  30623. tensor([[0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  30624. [ nan, nan, 0.7097, 0.2346, 0.4250, 0.1850, 0.5175, 0.5583],
  30625. [ nan, nan, 0.7425, 0.2117, 0.3937, 0.2433, 0.5438, 0.5567],
  30626. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5363, 0.5550],
  30627. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  30628. [0.6189, 0.4029, 0.8375, 0.5767, 0.4745, 0.4829, 0.5551, 0.5598],
  30629. [0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717],
  30630. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267]],
  30631. device='cuda:0', dtype=torch.float64)
  30632. predictions are: tensor([[0.6645, 0.4274, 0.8826, 0.4182, 0.3632, 0.4493, 0.6396, 0.5114],
  30633. [0.1255, 0.0742, 0.7000, 0.2189, 0.4249, 0.2004, 0.5262, 0.5696],
  30634. [0.1267, 0.0676, 0.7274, 0.2114, 0.3882, 0.2491, 0.5524, 0.5611],
  30635. [0.6629, 0.4283, 0.6798, 0.2681, 0.3986, 0.2032, 0.5701, 0.5781],
  30636. [0.6747, 0.4213, 0.8377, 0.5059, 0.4841, 0.4928, 0.5337, 0.5692],
  30637. [0.6728, 0.4323, 0.8164, 0.5713, 0.4575, 0.4565, 0.5950, 0.5876],
  30638. [0.6986, 0.4415, 0.7335, 0.2450, 0.3789, 0.2552, 0.6003, 0.5945],
  30639. [0.7199, 0.4423, 0.8084, 0.5859, 0.3913, 0.4404, 0.6138, 0.5356]],
  30640. device='cuda:0', grad_fn=<AddmmBackward>)
  30641. landmarks are: tensor([[[0.6264, 0.4067, 0.9050, 0.4183, 0.3775, 0.4600, 0.6308, 0.4862],
  30642. [0.0000, 0.0000, 0.7097, 0.2346, 0.4250, 0.1850, 0.5175, 0.5583],
  30643. [0.0000, 0.0000, 0.7425, 0.2117, 0.3938, 0.2433, 0.5437, 0.5567],
  30644. [0.6129, 0.4114, 0.6950, 0.2467, 0.4050, 0.2133, 0.5362, 0.5550],
  30645. [0.6124, 0.4030, 0.8650, 0.4867, 0.4999, 0.5106, 0.5137, 0.5773],
  30646. [0.6189, 0.4029, 0.8375, 0.5767, 0.4745, 0.4829, 0.5551, 0.5598],
  30647. [0.6193, 0.4108, 0.7425, 0.2350, 0.3887, 0.2750, 0.5900, 0.5717],
  30648. [0.6224, 0.3964, 0.8225, 0.5717, 0.4150, 0.4617, 0.5775, 0.5267]]],
  30649. device='cuda:0')
  30650. loss_train_step before backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  30651. loss_train_step after backward: tensor(0.0014, device='cuda:0', grad_fn=<MseLossBackward>)
  30652. loss_train: 0.03802865924080834
  30653. step: 83
  30654. running loss: 0.0004581766173591366
  30655. Train Steps: 83/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30656. torch.Size([8, 8])
  30657. tensor([[ nan, nan, 0.8363, 0.3317, 0.3563, 0.3367, 0.5329, 0.5142],
  30658. [0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533],
  30659. [0.6332, 0.4165, 0.9100, 0.3350, 0.4188, 0.3683, 0.7438, 0.5528],
  30660. [0.6229, 0.4066, 0.7612, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  30661. [0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  30662. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  30663. [0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  30664. [0.6260, 0.4214, 0.8538, 0.5500, 0.3663, 0.3767, 0.5587, 0.5983]],
  30665. device='cuda:0', dtype=torch.float64)
  30666. predictions are: tensor([[0.0603, 0.0229, 0.7907, 0.3199, 0.3460, 0.3630, 0.5440, 0.5371],
  30667. [0.6344, 0.3954, 0.6876, 0.2187, 0.3711, 0.3368, 0.6029, 0.5666],
  30668. [0.6674, 0.4167, 0.8742, 0.3484, 0.4033, 0.3879, 0.7346, 0.5647],
  30669. [0.6683, 0.3995, 0.7279, 0.2874, 0.4217, 0.2310, 0.5822, 0.5594],
  30670. [0.6703, 0.4092, 0.8422, 0.4734, 0.4231, 0.4850, 0.5436, 0.5723],
  30671. [0.6740, 0.4162, 0.8302, 0.2882, 0.4772, 0.2208, 0.6648, 0.5577],
  30672. [0.6306, 0.3858, 0.8653, 0.4550, 0.3553, 0.4258, 0.5858, 0.5191],
  30673. [0.6742, 0.4291, 0.8340, 0.5555, 0.3548, 0.3969, 0.5518, 0.6012]],
  30674. device='cuda:0', grad_fn=<AddmmBackward>)
  30675. landmarks are: tensor([[[0.0000, 0.0000, 0.8363, 0.3317, 0.3562, 0.3367, 0.5329, 0.5142],
  30676. [0.6222, 0.4072, 0.7164, 0.2166, 0.3738, 0.3167, 0.6100, 0.5533],
  30677. [0.6332, 0.4165, 0.9100, 0.3350, 0.4187, 0.3683, 0.7438, 0.5528],
  30678. [0.6229, 0.4066, 0.7613, 0.2967, 0.4250, 0.2167, 0.5925, 0.5350],
  30679. [0.6138, 0.4020, 0.8800, 0.4717, 0.4375, 0.4717, 0.5502, 0.5611],
  30680. [0.6361, 0.4102, 0.8650, 0.2983, 0.4888, 0.2000, 0.6702, 0.5459],
  30681. [0.6214, 0.3982, 0.8938, 0.4517, 0.3663, 0.4083, 0.5863, 0.5050],
  30682. [0.6260, 0.4214, 0.8537, 0.5500, 0.3663, 0.3767, 0.5587, 0.5983]]],
  30683. device='cuda:0')
  30684. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  30685. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  30686. loss_train: 0.03851493066758849
  30687. step: 84
  30688. running loss: 0.0004585110793760534
  30689. Train Steps: 84/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30690. torch.Size([8, 8])
  30691. tensor([[0.6203, 0.4096, 0.8862, 0.4267, 0.3538, 0.4117, 0.6025, 0.5650],
  30692. [0.6240, 0.4217, 0.8150, 0.3133, 0.4425, 0.2650, 0.5650, 0.5817],
  30693. [0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  30694. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  30695. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896],
  30696. [0.6185, 0.4129, 0.8900, 0.4567, 0.3937, 0.5417, 0.5734, 0.5110],
  30697. [0.6186, 0.4060, 0.8750, 0.5050, 0.3538, 0.4367, 0.5813, 0.6083],
  30698. [0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332]],
  30699. device='cuda:0', dtype=torch.float64)
  30700. predictions are: tensor([[0.6436, 0.4039, 0.8519, 0.4282, 0.3554, 0.4118, 0.5926, 0.5785],
  30701. [0.6478, 0.4293, 0.7909, 0.3250, 0.4391, 0.2788, 0.5641, 0.5899],
  30702. [0.6030, 0.3804, 0.8897, 0.3661, 0.4324, 0.3936, 0.7197, 0.5556],
  30703. [0.5850, 0.3690, 0.7338, 0.2239, 0.3816, 0.2882, 0.5983, 0.5582],
  30704. [0.5742, 0.3576, 0.7863, 0.3901, 0.3561, 0.3192, 0.5261, 0.6010],
  30705. [0.5740, 0.3685, 0.8627, 0.4725, 0.3765, 0.5421, 0.5540, 0.5164],
  30706. [0.6090, 0.3848, 0.8427, 0.4962, 0.3518, 0.4459, 0.5653, 0.6092],
  30707. [0.6362, 0.3829, 0.8636, 0.4940, 0.3785, 0.4176, 0.6509, 0.5184]],
  30708. device='cuda:0', grad_fn=<AddmmBackward>)
  30709. landmarks are: tensor([[[0.6203, 0.4096, 0.8863, 0.4267, 0.3537, 0.4117, 0.6025, 0.5650],
  30710. [0.6240, 0.4217, 0.8150, 0.3133, 0.4425, 0.2650, 0.5650, 0.5817],
  30711. [0.6332, 0.4128, 0.9200, 0.3517, 0.4400, 0.3833, 0.7461, 0.5494],
  30712. [0.6182, 0.3982, 0.7541, 0.2379, 0.3959, 0.2792, 0.6079, 0.5389],
  30713. [0.6137, 0.4084, 0.8076, 0.3889, 0.3650, 0.3150, 0.5356, 0.5896],
  30714. [0.6186, 0.4129, 0.8900, 0.4567, 0.3938, 0.5417, 0.5734, 0.5110],
  30715. [0.6186, 0.4060, 0.8750, 0.5050, 0.3537, 0.4367, 0.5813, 0.6083],
  30716. [0.6361, 0.4071, 0.9100, 0.4783, 0.3738, 0.3967, 0.6670, 0.5332]]],
  30717. device='cuda:0')
  30718. loss_train_step before backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  30719. loss_train_step after backward: tensor(0.0004, device='cuda:0', grad_fn=<MseLossBackward>)
  30720. loss_train: 0.03894413818488829
  30721. step: 85
  30722. running loss: 0.00045816633158692104
  30723. Train Steps: 85/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30724. torch.Size([8, 8])
  30725. tensor([[0.6204, 0.4049, 0.7975, 0.2700, 0.3937, 0.2567, 0.5700, 0.5183],
  30726. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  30727. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  30728. [0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767],
  30729. [0.6200, 0.3961, 0.8461, 0.5497, 0.4142, 0.4577, 0.5892, 0.5402],
  30730. [0.6272, 0.4071, 0.8738, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  30731. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050],
  30732. [0.6274, 0.4003, 0.8638, 0.5967, 0.3688, 0.4900, 0.6108, 0.4661]],
  30733. device='cuda:0', dtype=torch.float64)
  30734. predictions are: tensor([[0.6141, 0.3935, 0.7946, 0.2674, 0.3842, 0.2769, 0.5639, 0.5398],
  30735. [0.6237, 0.3929, 0.8780, 0.4692, 0.4058, 0.5315, 0.6351, 0.5098],
  30736. [0.6030, 0.3903, 0.8513, 0.2258, 0.5386, 0.2168, 0.7136, 0.5800],
  30737. [0.6055, 0.3781, 0.8742, 0.4740, 0.3947, 0.4553, 0.5317, 0.6069],
  30738. [0.5899, 0.3682, 0.8513, 0.5185, 0.4040, 0.4640, 0.5885, 0.5614],
  30739. [0.6357, 0.3988, 0.8772, 0.5516, 0.3704, 0.3951, 0.5980, 0.4916],
  30740. [0.5939, 0.3816, 0.8711, 0.5063, 0.4372, 0.5173, 0.5934, 0.5332],
  30741. [0.6167, 0.3730, 0.8572, 0.5795, 0.3756, 0.4856, 0.6109, 0.4972]],
  30742. device='cuda:0', grad_fn=<AddmmBackward>)
  30743. landmarks are: tensor([[[0.6204, 0.4049, 0.7975, 0.2700, 0.3938, 0.2567, 0.5700, 0.5183],
  30744. [0.6263, 0.4029, 0.8900, 0.4933, 0.4075, 0.5183, 0.6406, 0.4758],
  30745. [0.6329, 0.4175, 0.8550, 0.2333, 0.5425, 0.2250, 0.7398, 0.5609],
  30746. [0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767],
  30747. [0.6200, 0.3961, 0.8461, 0.5497, 0.4142, 0.4577, 0.5892, 0.5402],
  30748. [0.6272, 0.4071, 0.8737, 0.5600, 0.3675, 0.3783, 0.5926, 0.4742],
  30749. [0.6189, 0.4033, 0.8650, 0.5267, 0.4487, 0.5150, 0.5925, 0.5050],
  30750. [0.6274, 0.4003, 0.8637, 0.5967, 0.3688, 0.4900, 0.6108, 0.4661]]],
  30751. device='cuda:0')
  30752. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  30753. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  30754. loss_train: 0.03921863157302141
  30755. step: 86
  30756. running loss: 0.0004560305996862955
  30757.  
  30758. Train Steps: 86/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30759. torch.Size([8, 8])
  30760. tensor([[0.6270, 0.4267, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  30761. [0.6307, 0.4045, 0.8025, 0.5833, 0.3775, 0.4867, 0.6892, 0.5459],
  30762. [ nan, nan, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  30763. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  30764. [ nan, nan, 0.7725, 0.2611, 0.3675, 0.2733, 0.5413, 0.5167],
  30765. [0.6115, 0.4081, 0.6725, 0.2433, 0.4088, 0.1933, 0.5167, 0.5544],
  30766. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  30767. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092]],
  30768. device='cuda:0', dtype=torch.float64)
  30769. predictions are: tensor([[ 0.6535, 0.4477, 0.7380, 0.3077, 0.4887, 0.1751, 0.5390, 0.6052],
  30770. [ 0.6257, 0.4103, 0.8346, 0.5530, 0.4003, 0.4779, 0.6911, 0.5352],
  30771. [-0.0232, -0.0088, 0.8852, 0.2373, 0.5437, 0.2008, 0.7311, 0.5667],
  30772. [ 0.6336, 0.4117, 0.8687, 0.4738, 0.4279, 0.4695, 0.5236, 0.5331],
  30773. [ 0.0291, 0.0120, 0.7872, 0.2631, 0.3804, 0.2830, 0.5299, 0.5368],
  30774. [ 0.5745, 0.3887, 0.7025, 0.2337, 0.4180, 0.1965, 0.5124, 0.5520],
  30775. [ 0.6495, 0.4207, 0.8767, 0.5533, 0.3920, 0.4965, 0.6463, 0.5284],
  30776. [ 0.6101, 0.3976, 0.8975, 0.4903, 0.3915, 0.5094, 0.5763, 0.4975]],
  30777. device='cuda:0', grad_fn=<AddmmBackward>)
  30778. landmarks are: tensor([[[0.6270, 0.4266, 0.7150, 0.3317, 0.4900, 0.1817, 0.5560, 0.6183],
  30779. [0.6307, 0.4045, 0.8025, 0.5833, 0.3775, 0.4867, 0.6892, 0.5459],
  30780. [0.0000, 0.0000, 0.8625, 0.2550, 0.5487, 0.2200, 0.7335, 0.5737],
  30781. [0.6109, 0.3943, 0.8525, 0.4950, 0.4338, 0.4800, 0.5449, 0.5383],
  30782. [0.0000, 0.0000, 0.7725, 0.2611, 0.3675, 0.2733, 0.5412, 0.5167],
  30783. [0.6115, 0.4081, 0.6725, 0.2433, 0.4087, 0.1933, 0.5167, 0.5544],
  30784. [0.6364, 0.4092, 0.8525, 0.5633, 0.3875, 0.4950, 0.6599, 0.5285],
  30785. [0.6201, 0.4004, 0.8786, 0.5043, 0.3833, 0.5138, 0.5997, 0.5092]]],
  30786. device='cuda:0')
  30787. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  30788. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  30789. loss_train: 0.039484524459112436
  30790. step: 87
  30791. running loss: 0.00045384510872543027
  30792. Train Steps: 87/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30793. torch.Size([8, 8])
  30794. tensor([[0.6267, 0.4080, 0.8438, 0.2633, 0.4763, 0.1800, 0.6259, 0.5240],
  30795. [0.6239, 0.4123, 0.8313, 0.2550, 0.4500, 0.2050, 0.6175, 0.5400],
  30796. [0.6254, 0.3993, 0.8988, 0.4767, 0.3987, 0.5517, 0.6955, 0.5285],
  30797. [0.6182, 0.3972, 0.8552, 0.5914, 0.3683, 0.4181, 0.5688, 0.5378],
  30798. [0.6223, 0.4028, 0.8988, 0.4200, 0.3763, 0.5733, 0.6375, 0.5167],
  30799. [0.6185, 0.4079, 0.8838, 0.4617, 0.4838, 0.5650, 0.6175, 0.5850],
  30800. [0.6224, 0.4097, 0.7438, 0.2267, 0.3850, 0.2850, 0.5988, 0.5250],
  30801. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297]],
  30802. device='cuda:0', dtype=torch.float64)
  30803. predictions are: tensor([[0.5603, 0.3784, 0.8676, 0.2693, 0.4874, 0.1654, 0.6212, 0.5118],
  30804. [0.5558, 0.3819, 0.8536, 0.2485, 0.4586, 0.1737, 0.6244, 0.5170],
  30805. [0.5744, 0.3882, 0.9210, 0.4832, 0.4121, 0.5297, 0.6812, 0.5251],
  30806. [0.6119, 0.4007, 0.8821, 0.6022, 0.3731, 0.4060, 0.5844, 0.5124],
  30807. [0.6405, 0.4299, 0.9178, 0.4135, 0.3962, 0.5504, 0.6143, 0.5061],
  30808. [0.5793, 0.4005, 0.9085, 0.4624, 0.4908, 0.5308, 0.6161, 0.5828],
  30809. [0.5847, 0.4053, 0.7593, 0.2255, 0.4041, 0.2817, 0.5970, 0.5267],
  30810. [0.5840, 0.4049, 0.8577, 0.5849, 0.3925, 0.4816, 0.6681, 0.5257]],
  30811. device='cuda:0', grad_fn=<AddmmBackward>)
  30812. landmarks are: tensor([[[0.6267, 0.4080, 0.8438, 0.2633, 0.4762, 0.1800, 0.6259, 0.5240],
  30813. [0.6239, 0.4123, 0.8313, 0.2550, 0.4500, 0.2050, 0.6175, 0.5400],
  30814. [0.6254, 0.3993, 0.8988, 0.4767, 0.3988, 0.5517, 0.6955, 0.5285],
  30815. [0.6182, 0.3972, 0.8552, 0.5914, 0.3683, 0.4181, 0.5688, 0.5378],
  30816. [0.6223, 0.4028, 0.8988, 0.4200, 0.3762, 0.5733, 0.6375, 0.5167],
  30817. [0.6184, 0.4079, 0.8838, 0.4617, 0.4837, 0.5650, 0.6175, 0.5850],
  30818. [0.6224, 0.4097, 0.7437, 0.2267, 0.3850, 0.2850, 0.5987, 0.5250],
  30819. [0.6357, 0.4139, 0.8450, 0.5883, 0.3775, 0.4950, 0.6488, 0.5297]]],
  30820. device='cuda:0')
  30821. loss_train_step before backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  30822. loss_train_step after backward: tensor(0.0005, device='cuda:0', grad_fn=<MseLossBackward>)
  30823. loss_train: 0.03999189910246059
  30824. step: 88
  30825. running loss: 0.00045445339889159766
  30826. Train Steps: 88/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30827. torch.Size([8, 8])
  30828. tensor([[0.6163, 0.4001, 0.8788, 0.5033, 0.4012, 0.4633, 0.5338, 0.5767],
  30829. [0.6117, 0.4019, 0.8538, 0.4067, 0.3513, 0.3583, 0.5663, 0.5133],
  30830. [0.6245, 0.4100, 0.7762, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  30831. [0.6289, 0.4032, 0.8419, 0.5446, 0.4075, 0.5017, 0.6312, 0.5117],
  30832. [0.6131, 0.4037, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  30833. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5787, 0.5550],
  30834. [0.6216, 0.4099, 0.7225, 0.2033, 0.4188, 0.2217, 0.5975, 0.5283],
  30835. [0.6055, 0.4015, 0.7425, 0.2033, 0.4113, 0.1883, 0.5217, 0.4823]],
  30836. device='cuda:0', dtype=torch.float64)
  30837. predictions are: tensor([[0.6066, 0.3985, 0.9077, 0.5174, 0.4141, 0.4509, 0.5534, 0.5614],
  30838. [0.5354, 0.3544, 0.8979, 0.4296, 0.3646, 0.3531, 0.5639, 0.4929],
  30839. [0.5940, 0.4112, 0.8060, 0.2632, 0.5045, 0.1279, 0.6124, 0.5162],
  30840. [0.6000, 0.3899, 0.8762, 0.5717, 0.4203, 0.5012, 0.6693, 0.4945],
  30841. [0.5872, 0.3957, 0.7123, 0.2898, 0.3786, 0.2703, 0.5455, 0.5416],
  30842. [0.5807, 0.3968, 0.8483, 0.2766, 0.3685, 0.3557, 0.5858, 0.5332],
  30843. [0.5939, 0.4277, 0.7381, 0.2260, 0.4264, 0.2097, 0.5972, 0.5062],
  30844. [0.5100, 0.3544, 0.7657, 0.2191, 0.4093, 0.1690, 0.5424, 0.4659]],
  30845. device='cuda:0', grad_fn=<AddmmBackward>)
  30846. landmarks are: tensor([[[0.6163, 0.4001, 0.8788, 0.5033, 0.4013, 0.4633, 0.5337, 0.5767],
  30847. [0.6116, 0.4019, 0.8537, 0.4067, 0.3512, 0.3583, 0.5663, 0.5133],
  30848. [0.6245, 0.4100, 0.7763, 0.2583, 0.4963, 0.1517, 0.5875, 0.5417],
  30849. [0.6289, 0.4031, 0.8419, 0.5446, 0.4075, 0.5017, 0.6313, 0.5117],
  30850. [0.6131, 0.4036, 0.6907, 0.2819, 0.3688, 0.2700, 0.5217, 0.5680],
  30851. [0.6172, 0.4055, 0.8175, 0.2650, 0.3550, 0.3683, 0.5788, 0.5550],
  30852. [0.6216, 0.4099, 0.7225, 0.2033, 0.4187, 0.2217, 0.5975, 0.5283],
  30853. [0.6055, 0.4015, 0.7425, 0.2033, 0.4112, 0.1883, 0.5217, 0.4823]]],
  30854. device='cuda:0')
  30855. loss_train_step before backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  30856. loss_train_step after backward: tensor(0.0007, device='cuda:0', grad_fn=<MseLossBackward>)
  30857. loss_train: 0.040671683149412274
  30858. step: 89
  30859. running loss: 0.0004569852039259806
  30860. Train Steps: 89/90 Loss: 0.0005 torch.Size([8, 600, 800])
  30861. torch.Size([8, 8])
  30862. tensor([[0.6110, 0.4047, 0.8700, 0.4483, 0.3713, 0.3967, 0.5088, 0.5517],
  30863. [0.6138, 0.4101, 0.8800, 0.5083, 0.4637, 0.5950, 0.5587, 0.5077],
  30864. [0.6272, 0.4045, 0.8538, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  30865. [0.6224, 0.4179, 0.8700, 0.5683, 0.4037, 0.4683, 0.5650, 0.5600],
  30866. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  30867. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250],
  30868. [0.6069, 0.3975, 0.8625, 0.5083, 0.4388, 0.5483, 0.5650, 0.4967],
  30869. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6038, 0.4833]],
  30870. device='cuda:0', dtype=torch.float64)
  30871. predictions are: tensor([[0.5591, 0.3871, 0.8722, 0.4501, 0.3691, 0.3893, 0.5137, 0.5304],
  30872. [0.6191, 0.4178, 0.8732, 0.5310, 0.4599, 0.5640, 0.5903, 0.5062],
  30873. [0.6055, 0.4156, 0.8485, 0.5866, 0.3716, 0.4301, 0.6193, 0.4586],
  30874. [0.5629, 0.3912, 0.8800, 0.5770, 0.4015, 0.4430, 0.5927, 0.5503],
  30875. [0.5660, 0.3942, 0.8862, 0.4269, 0.3550, 0.3593, 0.5990, 0.5308],
  30876. [0.5981, 0.4102, 0.9046, 0.4074, 0.3817, 0.3150, 0.6375, 0.5045],
  30877. [0.6029, 0.4083, 0.8549, 0.5220, 0.4479, 0.5373, 0.5806, 0.5060],
  30878. [0.6232, 0.4102, 0.8882, 0.4807, 0.3777, 0.4603, 0.6204, 0.4733]],
  30879. device='cuda:0', grad_fn=<AddmmBackward>)
  30880. landmarks are: tensor([[[0.6110, 0.4047, 0.8700, 0.4483, 0.3713, 0.3967, 0.5088, 0.5517],
  30881. [0.6138, 0.4101, 0.8800, 0.5083, 0.4638, 0.5950, 0.5587, 0.5077],
  30882. [0.6271, 0.4045, 0.8537, 0.5900, 0.3750, 0.4417, 0.5989, 0.4649],
  30883. [0.6224, 0.4179, 0.8700, 0.5683, 0.4038, 0.4683, 0.5650, 0.5600],
  30884. [0.6222, 0.4108, 0.8938, 0.4233, 0.3600, 0.3817, 0.5825, 0.5283],
  30885. [0.6272, 0.4120, 0.9038, 0.4117, 0.3725, 0.3200, 0.6175, 0.5250],
  30886. [0.6069, 0.3975, 0.8625, 0.5083, 0.4387, 0.5483, 0.5650, 0.4967],
  30887. [0.6189, 0.3911, 0.8800, 0.4917, 0.3713, 0.4800, 0.6037, 0.4833]]],
  30888. device='cuda:0')
  30889. loss_train_step before backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  30890. loss_train_step after backward: tensor(0.0003, device='cuda:0', grad_fn=<MseLossBackward>)
  30891. loss_train: 0.04101139504928142
  30892. step: 90
  30893. running loss: 0.00045568216721423796
  30894.  
  30895. Valid Steps: 10/10 Loss: nan 05
  30896. --------------------------------------------------
  30897. Epoch: 10 Train Loss: 0.0005 Valid Loss: nan
  30898. --------------------------------------------------
  30899. Training Complete
  30900. Total Elapsed Time : 452.1295247077942 s
  30901.  
  30902.  
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement
Public Pastes
Advertisement
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!