API tools faq
paste
Guest User

TF Dump Training with distribution strategy

a guest
Nov 1st, 2019
126
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 42.34 KB | None | 0 0
raw download clone embed print report
  1. Output with MirroredStrategy
  2. ds_valid.shuffle(len(list_paths_valid), seed=42,
  3. reshuffle_each_iteration=True)
  4.  
  5.  
  6. 2019-11-01 10:25:31.215552: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1
  7. 2019-11-01 10:25:31.295080: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1546] Found device 0 with properties:
  8. pciBusID: 0000:86:00.0 name: GeForce RTX 2080 Ti computeCapability: 7.5
  9. coreClock: 1.545GHz coreCount: 68 deviceMemorySize: 10.76GiB deviceMemoryBandwidth: 573.69GiB/s
  10. 2019-11-01 10:25:31.295801: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1546] Found device 1 with properties:
  11. pciBusID: 0000:af:00.0 name: GeForce RTX 2080 Ti computeCapability: 7.5
  12. coreClock: 1.545GHz coreCount: 68 deviceMemorySize: 10.76GiB deviceMemoryBandwidth: 573.69GiB/s
  13. 2019-11-01 10:25:31.296119: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0
  14. 2019-11-01 10:25:31.298252: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10.0
  15. 2019-11-01 10:25:31.299728: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10.0
  16. 2019-11-01 10:25:31.300093: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10.0
  17. 2019-11-01 10:25:31.302095: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10.0
  18. 2019-11-01 10:25:31.303592: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10.0
  19. 2019-11-01 10:25:31.308127: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
  20. 2019-11-01 10:25:31.310488: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1688] Adding visible gpu devices: 0, 1
  21. 2019-11-01 10:25:31.310882: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 AVX512F FMA
  22. 2019-11-01 10:25:31.354089: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 2100000000 Hz
  23. 2019-11-01 10:25:31.365967: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x57ab6f0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:
  24. 2019-11-01 10:25:31.366024: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version
  25. 2019-11-01 10:25:31.584146: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x53772a0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:
  26. 2019-11-01 10:25:31.584201: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): GeForce RTX 2080 Ti, Compute Capability 7.5
  27. 2019-11-01 10:25:31.584216: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (1): GeForce RTX 2080 Ti, Compute Capability 7.5
  28. 2019-11-01 10:25:31.585984: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1546] Found device 0 with properties:
  29. pciBusID: 0000:86:00.0 name: GeForce RTX 2080 Ti computeCapability: 7.5
  30. coreClock: 1.545GHz coreCount: 68 deviceMemorySize: 10.76GiB deviceMemoryBandwidth: 573.69GiB/s
  31. 2019-11-01 10:25:31.587358: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1546] Found device 1 with properties:
  32. pciBusID: 0000:af:00.0 name: GeForce RTX 2080 Ti computeCapability: 7.5
  33. coreClock: 1.545GHz coreCount: 68 deviceMemorySize: 10.76GiB deviceMemoryBandwidth: 573.69GiB/s
  34. 2019-11-01 10:25:31.587435: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0
  35. 2019-11-01 10:25:31.587471: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10.0
  36. 2019-11-01 10:25:31.587503: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10.0
  37. 2019-11-01 10:25:31.587535: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10.0
  38. 2019-11-01 10:25:31.587567: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10.0
  39. 2019-11-01 10:25:31.587598: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10.0
  40. 2019-11-01 10:25:31.587631: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
  41. 2019-11-01 10:25:31.592892: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1688] Adding visible gpu devices: 0, 1
  42. 2019-11-01 10:25:31.592970: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0
  43. 2019-11-01 10:25:31.597198: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1087] Device interconnect StreamExecutor with strength 1 edge matrix:
  44. 2019-11-01 10:25:31.597230: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1093] 0 1
  45. 2019-11-01 10:25:31.597248: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1106] 0: N N
  46. 2019-11-01 10:25:31.597263: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1106] 1: N N
  47. 2019-11-01 10:25:31.599301: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1232] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 10312 MB memory) -> physical GPU (device: 0, name: GeForce RTX 2080 Ti, pci bus id: 0000:86:00.0, compute capability: 7.5)
  48. 2019-11-01 10:25:31.600473: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1232] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:1 with 10312 MB memory) -> physical GPU (device: 1, name: GeForce RTX 2080 Ti, pci bus id: 0000:af:00.0, compute capability: 7.5)
  49. 2019-11-01 09:25:31,621 | file_handler | INFO : found 24 datasets
  50. 2019-11-01 09:25:31,624 | file_handler | INFO : loaded evaluation_overview.csv
  51. 2019-11-01 09:25:31,627 | file_handler | INFO : found 272 trainings
  52. 2019-11-01 09:25:31,629 | deeplearning_tools | INFO : Train on trainUid : 8
  53. Model: "resnet50"
  54. __________________________________________________________________________________________________
  55. Layer (type) Output Shape Param # Connected to
  56. ==================================================================================================
  57. input_1 (InputLayer) [(None, 270, 270, 1) 0
  58. __________________________________________________________________________________________________
  59. conv1_pad (ZeroPadding2D) (None, 276, 276, 1) 0 input_1[0][0]
  60. __________________________________________________________________________________________________
  61. conv1_conv (Conv2D) (None, 135, 135, 64) 3200 conv1_pad[0][0]
  62. __________________________________________________________________________________________________
  63. conv1_bn (BatchNormalization) (None, 135, 135, 64) 256 conv1_conv[0][0]
  64. __________________________________________________________________________________________________
  65. conv1_relu (Activation) (None, 135, 135, 64) 0 conv1_bn[0][0]
  66. __________________________________________________________________________________________________
  67. pool1_pad (ZeroPadding2D) (None, 137, 137, 64) 0 conv1_relu[0][0]
  68. __________________________________________________________________________________________________
  69. pool1_pool (MaxPooling2D) (None, 68, 68, 64) 0 pool1_pad[0][0]
  70. __________________________________________________________________________________________________
  71. conv2_block1_1_conv (Conv2D) (None, 68, 68, 64) 4160 pool1_pool[0][0]
  72. __________________________________________________________________________________________________
  73. conv2_block1_1_bn (BatchNormali (None, 68, 68, 64) 256 conv2_block1_1_conv[0][0]
  74. __________________________________________________________________________________________________
  75. conv2_block1_1_relu (Activation (None, 68, 68, 64) 0 conv2_block1_1_bn[0][0]
  76. __________________________________________________________________________________________________
  77. conv2_block1_2_conv (Conv2D) (None, 68, 68, 64) 36928 conv2_block1_1_relu[0][0]
  78. __________________________________________________________________________________________________
  79. conv2_block1_2_bn (BatchNormali (None, 68, 68, 64) 256 conv2_block1_2_conv[0][0]
  80. __________________________________________________________________________________________________
  81. conv2_block1_2_relu (Activation (None, 68, 68, 64) 0 conv2_block1_2_bn[0][0]
  82. __________________________________________________________________________________________________
  83. conv2_block1_0_conv (Conv2D) (None, 68, 68, 256) 16640 pool1_pool[0][0]
  84. __________________________________________________________________________________________________
  85. conv2_block1_3_conv (Conv2D) (None, 68, 68, 256) 16640 conv2_block1_2_relu[0][0]
  86. __________________________________________________________________________________________________
  87. conv2_block1_0_bn (BatchNormali (None, 68, 68, 256) 1024 conv2_block1_0_conv[0][0]
  88. __________________________________________________________________________________________________
  89. conv2_block1_3_bn (BatchNormali (None, 68, 68, 256) 1024 conv2_block1_3_conv[0][0]
  90. __________________________________________________________________________________________________
  91. conv2_block1_add (Add) (None, 68, 68, 256) 0 conv2_block1_0_bn[0][0]
  92. conv2_block1_3_bn[0][0]
  93. __________________________________________________________________________________________________
  94. conv2_block1_out (Activation) (None, 68, 68, 256) 0 conv2_block1_add[0][0]
  95. __________________________________________________________________________________________________
  96. conv2_block2_1_conv (Conv2D) (None, 68, 68, 64) 16448 conv2_block1_out[0][0]
  97. __________________________________________________________________________________________________
  98. conv2_block2_1_bn (BatchNormali (None, 68, 68, 64) 256 conv2_block2_1_conv[0][0]
  99. __________________________________________________________________________________________________
  100. conv2_block2_1_relu (Activation (None, 68, 68, 64) 0 conv2_block2_1_bn[0][0]
  101. __________________________________________________________________________________________________
  102. conv2_block2_2_conv (Conv2D) (None, 68, 68, 64) 36928 conv2_block2_1_relu[0][0]
  103. __________________________________________________________________________________________________
  104. conv2_block2_2_bn (BatchNormali (None, 68, 68, 64) 256 conv2_block2_2_conv[0][0]
  105. __________________________________________________________________________________________________
  106. conv2_block2_2_relu (Activation (None, 68, 68, 64) 0 conv2_block2_2_bn[0][0]
  107. __________________________________________________________________________________________________
  108. conv2_block2_3_conv (Conv2D) (None, 68, 68, 256) 16640 conv2_block2_2_relu[0][0]
  109. __________________________________________________________________________________________________
  110. conv2_block2_3_bn (BatchNormali (None, 68, 68, 256) 1024 conv2_block2_3_conv[0][0]
  111. __________________________________________________________________________________________________
  112. conv2_block2_add (Add) (None, 68, 68, 256) 0 conv2_block1_out[0][0]
  113. conv2_block2_3_bn[0][0]
  114. __________________________________________________________________________________________________
  115. conv2_block2_out (Activation) (None, 68, 68, 256) 0 conv2_block2_add[0][0]
  116. __________________________________________________________________________________________________
  117. conv2_block3_1_conv (Conv2D) (None, 68, 68, 64) 16448 conv2_block2_out[0][0]
  118. __________________________________________________________________________________________________
  119. conv2_block3_1_bn (BatchNormali (None, 68, 68, 64) 256 conv2_block3_1_conv[0][0]
  120. __________________________________________________________________________________________________
  121. conv2_block3_1_relu (Activation (None, 68, 68, 64) 0 conv2_block3_1_bn[0][0]
  122. __________________________________________________________________________________________________
  123. conv2_block3_2_conv (Conv2D) (None, 68, 68, 64) 36928 conv2_block3_1_relu[0][0]
  124. __________________________________________________________________________________________________
  125. conv2_block3_2_bn (BatchNormali (None, 68, 68, 64) 256 conv2_block3_2_conv[0][0]
  126. __________________________________________________________________________________________________
  127. conv2_block3_2_relu (Activation (None, 68, 68, 64) 0 conv2_block3_2_bn[0][0]
  128. __________________________________________________________________________________________________
  129. conv2_block3_3_conv (Conv2D) (None, 68, 68, 256) 16640 conv2_block3_2_relu[0][0]
  130. __________________________________________________________________________________________________
  131. conv2_block3_3_bn (BatchNormali (None, 68, 68, 256) 1024 conv2_block3_3_conv[0][0]
  132. __________________________________________________________________________________________________
  133. conv2_block3_add (Add) (None, 68, 68, 256) 0 conv2_block2_out[0][0]
  134. conv2_block3_3_bn[0][0]
  135. __________________________________________________________________________________________________
  136. conv2_block3_out (Activation) (None, 68, 68, 256) 0 conv2_block3_add[0][0]
  137. __________________________________________________________________________________________________
  138. conv3_block1_1_conv (Conv2D) (None, 34, 34, 128) 32896 conv2_block3_out[0][0]
  139. __________________________________________________________________________________________________
  140. conv3_block1_1_bn (BatchNormali (None, 34, 34, 128) 512 conv3_block1_1_conv[0][0]
  141. __________________________________________________________________________________________________
  142. conv3_block1_1_relu (Activation (None, 34, 34, 128) 0 conv3_block1_1_bn[0][0]
  143. __________________________________________________________________________________________________
  144. conv3_block1_2_conv (Conv2D) (None, 34, 34, 128) 147584 conv3_block1_1_relu[0][0]
  145. __________________________________________________________________________________________________
  146. conv3_block1_2_bn (BatchNormali (None, 34, 34, 128) 512 conv3_block1_2_conv[0][0]
  147. __________________________________________________________________________________________________
  148. conv3_block1_2_relu (Activation (None, 34, 34, 128) 0 conv3_block1_2_bn[0][0]
  149. __________________________________________________________________________________________________
  150. conv3_block1_0_conv (Conv2D) (None, 34, 34, 512) 131584 conv2_block3_out[0][0]
  151. __________________________________________________________________________________________________
  152. conv3_block1_3_conv (Conv2D) (None, 34, 34, 512) 66048 conv3_block1_2_relu[0][0]
  153. __________________________________________________________________________________________________
  154. conv3_block1_0_bn (BatchNormali (None, 34, 34, 512) 2048 conv3_block1_0_conv[0][0]
  155. __________________________________________________________________________________________________
  156. conv3_block1_3_bn (BatchNormali (None, 34, 34, 512) 2048 conv3_block1_3_conv[0][0]
  157. __________________________________________________________________________________________________
  158. conv3_block1_add (Add) (None, 34, 34, 512) 0 conv3_block1_0_bn[0][0]
  159. conv3_block1_3_bn[0][0]
  160. __________________________________________________________________________________________________
  161. conv3_block1_out (Activation) (None, 34, 34, 512) 0 conv3_block1_add[0][0]
  162. __________________________________________________________________________________________________
  163. conv3_block2_1_conv (Conv2D) (None, 34, 34, 128) 65664 conv3_block1_out[0][0]
  164. __________________________________________________________________________________________________
  165. conv3_block2_1_bn (BatchNormali (None, 34, 34, 128) 512 conv3_block2_1_conv[0][0]
  166. __________________________________________________________________________________________________
  167. conv3_block2_1_relu (Activation (None, 34, 34, 128) 0 conv3_block2_1_bn[0][0]
  168. __________________________________________________________________________________________________
  169. conv3_block2_2_conv (Conv2D) (None, 34, 34, 128) 147584 conv3_block2_1_relu[0][0]
  170. __________________________________________________________________________________________________
  171. conv3_block2_2_bn (BatchNormali (None, 34, 34, 128) 512 conv3_block2_2_conv[0][0]
  172. __________________________________________________________________________________________________
  173. conv3_block2_2_relu (Activation (None, 34, 34, 128) 0 conv3_block2_2_bn[0][0]
  174. __________________________________________________________________________________________________
  175. conv3_block2_3_conv (Conv2D) (None, 34, 34, 512) 66048 conv3_block2_2_relu[0][0]
  176. __________________________________________________________________________________________________
  177. conv3_block2_3_bn (BatchNormali (None, 34, 34, 512) 2048 conv3_block2_3_conv[0][0]
  178. __________________________________________________________________________________________________
  179. conv3_block2_add (Add) (None, 34, 34, 512) 0 conv3_block1_out[0][0]
  180. conv3_block2_3_bn[0][0]
  181. __________________________________________________________________________________________________
  182. conv3_block2_out (Activation) (None, 34, 34, 512) 0 conv3_block2_add[0][0]
  183. __________________________________________________________________________________________________
  184. conv3_block3_1_conv (Conv2D) (None, 34, 34, 128) 65664 conv3_block2_out[0][0]
  185. __________________________________________________________________________________________________
  186. conv3_block3_1_bn (BatchNormali (None, 34, 34, 128) 512 conv3_block3_1_conv[0][0]
  187. __________________________________________________________________________________________________
  188. conv3_block3_1_relu (Activation (None, 34, 34, 128) 0 conv3_block3_1_bn[0][0]
  189. __________________________________________________________________________________________________
  190. conv3_block3_2_conv (Conv2D) (None, 34, 34, 128) 147584 conv3_block3_1_relu[0][0]
  191. __________________________________________________________________________________________________
  192. conv3_block3_2_bn (BatchNormali (None, 34, 34, 128) 512 conv3_block3_2_conv[0][0]
  193. __________________________________________________________________________________________________
  194. conv3_block3_2_relu (Activation (None, 34, 34, 128) 0 conv3_block3_2_bn[0][0]
  195. __________________________________________________________________________________________________
  196. conv3_block3_3_conv (Conv2D) (None, 34, 34, 512) 66048 conv3_block3_2_relu[0][0]
  197. __________________________________________________________________________________________________
  198. conv3_block3_3_bn (BatchNormali (None, 34, 34, 512) 2048 conv3_block3_3_conv[0][0]
  199. __________________________________________________________________________________________________
  200. conv3_block3_add (Add) (None, 34, 34, 512) 0 conv3_block2_out[0][0]
  201. conv3_block3_3_bn[0][0]
  202. __________________________________________________________________________________________________
  203. conv3_block3_out (Activation) (None, 34, 34, 512) 0 conv3_block3_add[0][0]
  204. __________________________________________________________________________________________________
  205. conv3_block4_1_conv (Conv2D) (None, 34, 34, 128) 65664 conv3_block3_out[0][0]
  206. __________________________________________________________________________________________________
  207. conv3_block4_1_bn (BatchNormali (None, 34, 34, 128) 512 conv3_block4_1_conv[0][0]
  208. __________________________________________________________________________________________________
  209. conv3_block4_1_relu (Activation (None, 34, 34, 128) 0 conv3_block4_1_bn[0][0]
  210. __________________________________________________________________________________________________
  211. conv3_block4_2_conv (Conv2D) (None, 34, 34, 128) 147584 conv3_block4_1_relu[0][0]
  212. __________________________________________________________________________________________________
  213. conv3_block4_2_bn (BatchNormali (None, 34, 34, 128) 512 conv3_block4_2_conv[0][0]
  214. __________________________________________________________________________________________________
  215. conv3_block4_2_relu (Activation (None, 34, 34, 128) 0 conv3_block4_2_bn[0][0]
  216. __________________________________________________________________________________________________
  217. conv3_block4_3_conv (Conv2D) (None, 34, 34, 512) 66048 conv3_block4_2_relu[0][0]
  218. __________________________________________________________________________________________________
  219. conv3_block4_3_bn (BatchNormali (None, 34, 34, 512) 2048 conv3_block4_3_conv[0][0]
  220. __________________________________________________________________________________________________
  221. conv3_block4_add (Add) (None, 34, 34, 512) 0 conv3_block3_out[0][0]
  222. conv3_block4_3_bn[0][0]
  223. __________________________________________________________________________________________________
  224. conv3_block4_out (Activation) (None, 34, 34, 512) 0 conv3_block4_add[0][0]
  225. __________________________________________________________________________________________________
  226. conv4_block1_1_conv (Conv2D) (None, 17, 17, 256) 131328 conv3_block4_out[0][0]
  227. __________________________________________________________________________________________________
  228. conv4_block1_1_bn (BatchNormali (None, 17, 17, 256) 1024 conv4_block1_1_conv[0][0]
  229. __________________________________________________________________________________________________
  230. conv4_block1_1_relu (Activation (None, 17, 17, 256) 0 conv4_block1_1_bn[0][0]
  231. __________________________________________________________________________________________________
  232. conv4_block1_2_conv (Conv2D) (None, 17, 17, 256) 590080 conv4_block1_1_relu[0][0]
  233. __________________________________________________________________________________________________
  234. conv4_block1_2_bn (BatchNormali (None, 17, 17, 256) 1024 conv4_block1_2_conv[0][0]
  235. __________________________________________________________________________________________________
  236. conv4_block1_2_relu (Activation (None, 17, 17, 256) 0 conv4_block1_2_bn[0][0]
  237. __________________________________________________________________________________________________
  238. conv4_block1_0_conv (Conv2D) (None, 17, 17, 1024) 525312 conv3_block4_out[0][0]
  239. __________________________________________________________________________________________________
  240. conv4_block1_3_conv (Conv2D) (None, 17, 17, 1024) 263168 conv4_block1_2_relu[0][0]
  241. __________________________________________________________________________________________________
  242. conv4_block1_0_bn (BatchNormali (None, 17, 17, 1024) 4096 conv4_block1_0_conv[0][0]
  243. __________________________________________________________________________________________________
  244. conv4_block1_3_bn (BatchNormali (None, 17, 17, 1024) 4096 conv4_block1_3_conv[0][0]
  245. __________________________________________________________________________________________________
  246. conv4_block1_add (Add) (None, 17, 17, 1024) 0 conv4_block1_0_bn[0][0]
  247. conv4_block1_3_bn[0][0]
  248. __________________________________________________________________________________________________
  249. conv4_block1_out (Activation) (None, 17, 17, 1024) 0 conv4_block1_add[0][0]
  250. __________________________________________________________________________________________________
  251. conv4_block2_1_conv (Conv2D) (None, 17, 17, 256) 262400 conv4_block1_out[0][0]
  252. __________________________________________________________________________________________________
  253. conv4_block2_1_bn (BatchNormali (None, 17, 17, 256) 1024 conv4_block2_1_conv[0][0]
  254. __________________________________________________________________________________________________
  255. conv4_block2_1_relu (Activation (None, 17, 17, 256) 0 conv4_block2_1_bn[0][0]
  256. __________________________________________________________________________________________________
  257. conv4_block2_2_conv (Conv2D) (None, 17, 17, 256) 590080 conv4_block2_1_relu[0][0]
  258. __________________________________________________________________________________________________
  259. conv4_block2_2_bn (BatchNormali (None, 17, 17, 256) 1024 conv4_block2_2_conv[0][0]
  260. __________________________________________________________________________________________________
  261. conv4_block2_2_relu (Activation (None, 17, 17, 256) 0 conv4_block2_2_bn[0][0]
  262. __________________________________________________________________________________________________
  263. conv4_block2_3_conv (Conv2D) (None, 17, 17, 1024) 263168 conv4_block2_2_relu[0][0]
  264. __________________________________________________________________________________________________
  265. conv4_block2_3_bn (BatchNormali (None, 17, 17, 1024) 4096 conv4_block2_3_conv[0][0]
  266. __________________________________________________________________________________________________
  267. conv4_block2_add (Add) (None, 17, 17, 1024) 0 conv4_block1_out[0][0]
  268. conv4_block2_3_bn[0][0]
  269. __________________________________________________________________________________________________
  270. conv4_block2_out (Activation) (None, 17, 17, 1024) 0 conv4_block2_add[0][0]
  271. __________________________________________________________________________________________________
  272. conv4_block3_1_conv (Conv2D) (None, 17, 17, 256) 262400 conv4_block2_out[0][0]
  273. __________________________________________________________________________________________________
  274. conv4_block3_1_bn (BatchNormali (None, 17, 17, 256) 1024 conv4_block3_1_conv[0][0]
  275. __________________________________________________________________________________________________
  276. conv4_block3_1_relu (Activation (None, 17, 17, 256) 0 conv4_block3_1_bn[0][0]
  277. __________________________________________________________________________________________________
  278. conv4_block3_2_conv (Conv2D) (None, 17, 17, 256) 590080 conv4_block3_1_relu[0][0]
  279. __________________________________________________________________________________________________
  280. conv4_block3_2_bn (BatchNormali (None, 17, 17, 256) 1024 conv4_block3_2_conv[0][0]
  281. __________________________________________________________________________________________________
  282. conv4_block3_2_relu (Activation (None, 17, 17, 256) 0 conv4_block3_2_bn[0][0]
  283. __________________________________________________________________________________________________
  284. conv4_block3_3_conv (Conv2D) (None, 17, 17, 1024) 263168 conv4_block3_2_relu[0][0]
  285. __________________________________________________________________________________________________
  286. conv4_block3_3_bn (BatchNormali (None, 17, 17, 1024) 4096 conv4_block3_3_conv[0][0]
  287. __________________________________________________________________________________________________
  288. conv4_block3_add (Add) (None, 17, 17, 1024) 0 conv4_block2_out[0][0]
  289. conv4_block3_3_bn[0][0]
  290. __________________________________________________________________________________________________
  291. conv4_block3_out (Activation) (None, 17, 17, 1024) 0 conv4_block3_add[0][0]
  292. __________________________________________________________________________________________________
  293. conv4_block4_1_conv (Conv2D) (None, 17, 17, 256) 262400 conv4_block3_out[0][0]
  294. __________________________________________________________________________________________________
  295. conv4_block4_1_bn (BatchNormali (None, 17, 17, 256) 1024 conv4_block4_1_conv[0][0]
  296. __________________________________________________________________________________________________
  297. conv4_block4_1_relu (Activation (None, 17, 17, 256) 0 conv4_block4_1_bn[0][0]
  298. __________________________________________________________________________________________________
  299. conv4_block4_2_conv (Conv2D) (None, 17, 17, 256) 590080 conv4_block4_1_relu[0][0]
  300. __________________________________________________________________________________________________
  301. conv4_block4_2_bn (BatchNormali (None, 17, 17, 256) 1024 conv4_block4_2_conv[0][0]
  302. __________________________________________________________________________________________________
  303. conv4_block4_2_relu (Activation (None, 17, 17, 256) 0 conv4_block4_2_bn[0][0]
  304. __________________________________________________________________________________________________
  305. conv4_block4_3_conv (Conv2D) (None, 17, 17, 1024) 263168 conv4_block4_2_relu[0][0]
  306. __________________________________________________________________________________________________
  307. conv4_block4_3_bn (BatchNormali (None, 17, 17, 1024) 4096 conv4_block4_3_conv[0][0]
  308. __________________________________________________________________________________________________
  309. conv4_block4_add (Add) (None, 17, 17, 1024) 0 conv4_block3_out[0][0]
  310. conv4_block4_3_bn[0][0]
  311. __________________________________________________________________________________________________
  312. conv4_block4_out (Activation) (None, 17, 17, 1024) 0 conv4_block4_add[0][0]
  313. __________________________________________________________________________________________________
  314. conv4_block5_1_conv (Conv2D) (None, 17, 17, 256) 262400 conv4_block4_out[0][0]
  315. __________________________________________________________________________________________________
  316. conv4_block5_1_bn (BatchNormali (None, 17, 17, 256) 1024 conv4_block5_1_conv[0][0]
  317. __________________________________________________________________________________________________
  318. conv4_block5_1_relu (Activation (None, 17, 17, 256) 0 conv4_block5_1_bn[0][0]
  319. __________________________________________________________________________________________________
  320. conv4_block5_2_conv (Conv2D) (None, 17, 17, 256) 590080 conv4_block5_1_relu[0][0]
  321. __________________________________________________________________________________________________
  322. conv4_block5_2_bn (BatchNormali (None, 17, 17, 256) 1024 conv4_block5_2_conv[0][0]
  323. __________________________________________________________________________________________________
  324. conv4_block5_2_relu (Activation (None, 17, 17, 256) 0 conv4_block5_2_bn[0][0]
  325. __________________________________________________________________________________________________
  326. conv4_block5_3_conv (Conv2D) (None, 17, 17, 1024) 263168 conv4_block5_2_relu[0][0]
  327. __________________________________________________________________________________________________
  328. conv4_block5_3_bn (BatchNormali (None, 17, 17, 1024) 4096 conv4_block5_3_conv[0][0]
  329. __________________________________________________________________________________________________
  330. conv4_block5_add (Add) (None, 17, 17, 1024) 0 conv4_block4_out[0][0]
  331. conv4_block5_3_bn[0][0]
  332. __________________________________________________________________________________________________
  333. conv4_block5_out (Activation) (None, 17, 17, 1024) 0 conv4_block5_add[0][0]
  334. __________________________________________________________________________________________________
  335. conv4_block6_1_conv (Conv2D) (None, 17, 17, 256) 262400 conv4_block5_out[0][0]
  336. __________________________________________________________________________________________________
  337. conv4_block6_1_bn (BatchNormali (None, 17, 17, 256) 1024 conv4_block6_1_conv[0][0]
  338. __________________________________________________________________________________________________
  339. conv4_block6_1_relu (Activation (None, 17, 17, 256) 0 conv4_block6_1_bn[0][0]
  340. __________________________________________________________________________________________________
  341. conv4_block6_2_conv (Conv2D) (None, 17, 17, 256) 590080 conv4_block6_1_relu[0][0]
  342. __________________________________________________________________________________________________
  343. conv4_block6_2_bn (BatchNormali (None, 17, 17, 256) 1024 conv4_block6_2_conv[0][0]
  344. __________________________________________________________________________________________________
  345. conv4_block6_2_relu (Activation (None, 17, 17, 256) 0 conv4_block6_2_bn[0][0]
  346. __________________________________________________________________________________________________
  347. conv4_block6_3_conv (Conv2D) (None, 17, 17, 1024) 263168 conv4_block6_2_relu[0][0]
  348. __________________________________________________________________________________________________
  349. conv4_block6_3_bn (BatchNormali (None, 17, 17, 1024) 4096 conv4_block6_3_conv[0][0]
  350. __________________________________________________________________________________________________
  351. conv4_block6_add (Add) (None, 17, 17, 1024) 0 conv4_block5_out[0][0]
  352. conv4_block6_3_bn[0][0]
  353. __________________________________________________________________________________________________
  354. conv4_block6_out (Activation) (None, 17, 17, 1024) 0 conv4_block6_add[0][0]
  355. __________________________________________________________________________________________________
  356. conv5_block1_1_conv (Conv2D) (None, 9, 9, 512) 524800 conv4_block6_out[0][0]
  357. __________________________________________________________________________________________________
  358. conv5_block1_1_bn (BatchNormali (None, 9, 9, 512) 2048 conv5_block1_1_conv[0][0]
  359. __________________________________________________________________________________________________
  360. conv5_block1_1_relu (Activation (None, 9, 9, 512) 0 conv5_block1_1_bn[0][0]
  361. __________________________________________________________________________________________________
  362. conv5_block1_2_conv (Conv2D) (None, 9, 9, 512) 2359808 conv5_block1_1_relu[0][0]
  363. __________________________________________________________________________________________________
  364. conv5_block1_2_bn (BatchNormali (None, 9, 9, 512) 2048 conv5_block1_2_conv[0][0]
  365. __________________________________________________________________________________________________
  366. conv5_block1_2_relu (Activation (None, 9, 9, 512) 0 conv5_block1_2_bn[0][0]
  367. __________________________________________________________________________________________________
  368. conv5_block1_0_conv (Conv2D) (None, 9, 9, 2048) 2099200 conv4_block6_out[0][0]
  369. __________________________________________________________________________________________________
  370. conv5_block1_3_conv (Conv2D) (None, 9, 9, 2048) 1050624 conv5_block1_2_relu[0][0]
  371. __________________________________________________________________________________________________
  372. conv5_block1_0_bn (BatchNormali (None, 9, 9, 2048) 8192 conv5_block1_0_conv[0][0]
  373. __________________________________________________________________________________________________
  374. conv5_block1_3_bn (BatchNormali (None, 9, 9, 2048) 8192 conv5_block1_3_conv[0][0]
  375. __________________________________________________________________________________________________
  376. conv5_block1_add (Add) (None, 9, 9, 2048) 0 conv5_block1_0_bn[0][0]
  377. conv5_block1_3_bn[0][0]
  378. __________________________________________________________________________________________________
  379. conv5_block1_out (Activation) (None, 9, 9, 2048) 0 conv5_block1_add[0][0]
  380. __________________________________________________________________________________________________
  381. conv5_block2_1_conv (Conv2D) (None, 9, 9, 512) 1049088 conv5_block1_out[0][0]
  382. __________________________________________________________________________________________________
  383. conv5_block2_1_bn (BatchNormali (None, 9, 9, 512) 2048 conv5_block2_1_conv[0][0]
  384. __________________________________________________________________________________________________
  385. conv5_block2_1_relu (Activation (None, 9, 9, 512) 0 conv5_block2_1_bn[0][0]
  386. __________________________________________________________________________________________________
  387. conv5_block2_2_conv (Conv2D) (None, 9, 9, 512) 2359808 conv5_block2_1_relu[0][0]
  388. __________________________________________________________________________________________________
  389. conv5_block2_2_bn (BatchNormali (None, 9, 9, 512) 2048 conv5_block2_2_conv[0][0]
  390. __________________________________________________________________________________________________
  391. conv5_block2_2_relu (Activation (None, 9, 9, 512) 0 conv5_block2_2_bn[0][0]
  392. __________________________________________________________________________________________________
  393. conv5_block2_3_conv (Conv2D) (None, 9, 9, 2048) 1050624 conv5_block2_2_relu[0][0]
  394. __________________________________________________________________________________________________
  395. conv5_block2_3_bn (BatchNormali (None, 9, 9, 2048) 8192 conv5_block2_3_conv[0][0]
  396. __________________________________________________________________________________________________
  397. conv5_block2_add (Add) (None, 9, 9, 2048) 0 conv5_block1_out[0][0]
  398. conv5_block2_3_bn[0][0]
  399. __________________________________________________________________________________________________
  400. conv5_block2_out (Activation) (None, 9, 9, 2048) 0 conv5_block2_add[0][0]
  401. __________________________________________________________________________________________________
  402. conv5_block3_1_conv (Conv2D) (None, 9, 9, 512) 1049088 conv5_block2_out[0][0]
  403. __________________________________________________________________________________________________
  404. conv5_block3_1_bn (BatchNormali (None, 9, 9, 512) 2048 conv5_block3_1_conv[0][0]
  405. __________________________________________________________________________________________________
  406. conv5_block3_1_relu (Activation (None, 9, 9, 512) 0 conv5_block3_1_bn[0][0]
  407. __________________________________________________________________________________________________
  408. conv5_block3_2_conv (Conv2D) (None, 9, 9, 512) 2359808 conv5_block3_1_relu[0][0]
  409. __________________________________________________________________________________________________
  410. conv5_block3_2_bn (BatchNormali (None, 9, 9, 512) 2048 conv5_block3_2_conv[0][0]
  411. __________________________________________________________________________________________________
  412. conv5_block3_2_relu (Activation (None, 9, 9, 512) 0 conv5_block3_2_bn[0][0]
  413. __________________________________________________________________________________________________
  414. conv5_block3_3_conv (Conv2D) (None, 9, 9, 2048) 1050624 conv5_block3_2_relu[0][0]
  415. __________________________________________________________________________________________________
  416. conv5_block3_3_bn (BatchNormali (None, 9, 9, 2048) 8192 conv5_block3_3_conv[0][0]
  417. __________________________________________________________________________________________________
  418. conv5_block3_add (Add) (None, 9, 9, 2048) 0 conv5_block2_out[0][0]
  419. conv5_block3_3_bn[0][0]
  420. __________________________________________________________________________________________________
  421. conv5_block3_out (Activation) (None, 9, 9, 2048) 0 conv5_block3_add[0][0]
  422. __________________________________________________________________________________________________
  423. avg_pool (GlobalAveragePooling2 (None, 2048) 0 conv5_block3_out[0][0]
  424. __________________________________________________________________________________________________
  425. probs (Dense) (None, 3) 6147 avg_pool[0][0]
  426. ==================================================================================================
  427. Total params: 23,587,587
  428. Trainable params: 23,534,467
  429. Non-trainable params: 53,120
  430. __________________________________________________________________________________________________
  431. Train for 563.0 steps, validate for 179.0 steps
  432. Epoch 1/10
  433. 2019-11-01 10:25:52.942924: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10.0
  434. 2019-11-01 10:25:54.882248: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
  435. 563/563 [==============================] - 119s 211ms/step - loss: 1.0535 - accuracy: 0.5099 - val_loss: 1.0735 - val_accuracy: 0.6682
  436. Epoch 2/10
  437. 563/563 [==============================] - 95s 169ms/step - loss: 1.0123 - accuracy: 0.5277 - val_loss: 1.0721 - val_accuracy: 0.6682
  438. Epoch 3/10
  439. 563/563 [==============================] - 95s 169ms/step - loss: 1.0121 - accuracy: 0.5277 - val_loss: 1.0709 - val_accuracy: 0.6682
  440. Epoch 4/10
  441. 563/563 [==============================] - 95s 169ms/step - loss: 1.0124 - accuracy: 0.5277 - val_loss: 1.0667 - val_accuracy: 0.6682
  442. Epoch 5/10
  443. 563/563 [==============================] - 95s 169ms/step - loss: 1.0121 - accuracy: 0.5277 - val_loss: 1.0687 - val_accuracy: 0.6682
  444. Epoch 6/10
  445. 563/563 [==============================] - 95s 168ms/step - loss: 1.0125 - accuracy: 0.5277 - val_loss: 1.0638 - val_accuracy: 0.6682
  446. Epoch 7/10
  447. 563/563 [==============================] - 94s 167ms/step - loss: 1.0125 - accuracy: 0.5277 - val_loss: 1.0639 - val_accuracy: 0.6682
  448. Epoch 8/10
  449. 400/563 [====================>.........] - ETA: 24s - loss: 1.0135 - accuracy: 0.5268
Add Comment
Please, Sign In to add comment
Public Pastes
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!