API tools faq
paste
Guest User

Untitled

a guest
Jul 19th, 2018
83
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 10.65 KB | None | 0 0
raw download clone embed print report
  1. Docker entrypoint called with argument(s): train
  2. [07/19/2018 16:09:27 INFO 139838496249664] Reading default configuration from /opt/amazon/lib/python2.7/site-packages/image_classification/default-input.json: {u'beta_1': 0.9, u'gamma': 0.9, u'beta_2': 0.999, u'optimizer': u'sgd', u'training_precision': u'float32', u'use_pretrained_model': 0, u'eps': 1e-08, u'epochs': 30, u'lr_scheduler_factor': 0.1, u'num_layers': 152, u'image_shape': u'3,224,224', u'mini_batch_size': 32, u'weight_decay': 0.0001, u'learning_rate': 0.1, u'momentum': 0}
  3. [07/19/2018 16:09:27 INFO 139838496249664] Reading provided configuration from /opt/ml/input/config/hyperparameters.json: {u'learning_rate': u'0.01', u'top_k': u'5', u'use_pretrained_model': u'1', u'epochs': u'6', u'num_classes': u'257', u'num_layers': u'18', u'image_shape': u'3,224,224', u'mini_batch_size': u'128', u'checkpoint_frequency': u'2', u'resize': u'256', u'num_training_samples': u'15240'}
  4. [07/19/2018 16:09:27 INFO 139838496249664] Final configuration: {u'top_k': u'5', u'optimizer': u'sgd', u'learning_rate': u'0.01', u'epochs': u'6', u'resize': u'256', u'lr_scheduler_factor': 0.1, u'num_layers': u'18', u'mini_batch_size': u'128', u'num_classes': u'257', u'beta_1': 0.9, u'beta_2': 0.999, u'training_precision': u'float32', u'use_pretrained_model': u'1', u'eps': 1e-08, u'checkpoint_frequency': u'2', u'weight_decay': 0.0001, u'momentum': 0, u'image_shape': u'3,224,224', u'gamma': 0.9, u'num_training_samples': u'15240'}
  5. [07/19/2018 16:09:27 INFO 139838496249664] Creating record files for catdog-12500-2500-train.lst
  6. [07/19/2018 16:10:33 INFO 139838496249664] Done creating record files...
  7. [07/19/2018 16:10:33 INFO 139838496249664] Creating record files for catdog-12500-2500-val.lst
  8. [07/19/2018 16:10:49 INFO 139838496249664] Done creating record files...
  9. [07/19/2018 16:10:49 INFO 139838496249664] Using pretrained model for initalizing weights
  10. [07/19/2018 16:10:49 INFO 139838496249664] ---- Parameters ----
  11. [07/19/2018 16:10:49 INFO 139838496249664] num_layers: 18
  12. [07/19/2018 16:10:49 INFO 139838496249664] data type: <type 'numpy.float32'>
  13. [07/19/2018 16:10:49 INFO 139838496249664] epochs: 6
  14. [07/19/2018 16:10:49 INFO 139838496249664] image resize size: 256
  15. [07/19/2018 16:10:49 INFO 139838496249664] optimizer: sgd
  16. [07/19/2018 16:10:49 INFO 139838496249664] momentum: 0.900000
  17. [07/19/2018 16:10:49 INFO 139838496249664] weight_decay: 0.000100
  18. [07/19/2018 16:10:49 INFO 139838496249664] learning_rate: 0.010000
  19. [07/19/2018 16:10:49 INFO 139838496249664] lr_scheduler_step defined without lr_scheduler_factor, will be ignored...
  20. [07/19/2018 16:10:49 INFO 139838496249664] mini_batch_size: 128
  21. [07/19/2018 16:10:49 INFO 139838496249664] image_shape: 3,224,224
  22. [07/19/2018 16:10:49 INFO 139838496249664] num_classes: 257
  23. [07/19/2018 16:10:49 INFO 139838496249664] num_training_samples: 15240
  24. [07/19/2018 16:10:49 INFO 139838496249664] augmentation_type: None
  25. [07/19/2018 16:10:49 INFO 139838496249664] kv_store: device
  26. [07/19/2018 16:10:49 INFO 139838496249664] top_k: 5
  27. [07/19/2018 16:10:49 INFO 139838496249664] checkpoint_frequency: 2
  28. [07/19/2018 16:10:49 INFO 139838496249664] --------------------
  29. [16:10:49] /opt/brazil-pkg-cache/packages/AIAlgorithmsMXNet/AIAlgorithmsMXNet-1.1.x.200520.0/RHEL5_64/generic-flavor/src/src/nnvm/legacy_json_util.cc:190: Loading symbol saved by previous version v0.8.0. Attempting to upgrade...
  30. [16:10:49] /opt/brazil-pkg-cache/packages/AIAlgorithmsMXNet/AIAlgorithmsMXNet-1.1.x.200520.0/RHEL5_64/generic-flavor/src/src/nnvm/legacy_json_util.cc:198: Symbol successfully upgraded!
  31. [16:10:57] /opt/brazil-pkg-cache/packages/AIAlgorithmsMXNet/AIAlgorithmsMXNet-1.1.x.200520.0/RHEL5_64/generic-flavor/src/src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:107: Running performance tests to find the best convolution algorithm, this can take a while... (setting env variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
  32. [07/19/2018 16:11:17 INFO 139838496249664] Epoch[0] Batch [20]#011Speed: 161.72 samples/sec#011accuracy=0.851935#011top_k_accuracy_5=0.901042
  33. [07/19/2018 16:11:33 INFO 139838496249664] Epoch[0] Batch [40]#011Speed: 161.46 samples/sec#011accuracy=0.909680#011top_k_accuracy_5=0.949314
  34. [07/19/2018 16:11:48 INFO 139838496249664] Epoch[0] Batch [60]#011Speed: 160.94 samples/sec#011accuracy=0.929047#011top_k_accuracy_5=0.965932
  35. [07/19/2018 16:12:05 INFO 139838496249664] Epoch[0] Batch [80]#011Speed: 159.39 samples/sec#011accuracy=0.940490#011top_k_accuracy_5=0.974344
  36. [07/19/2018 16:12:21 INFO 139838496249664] Epoch[0] Batch [100]#011Speed: 159.25 samples/sec#011accuracy=0.946860#011top_k_accuracy_5=0.979425
  37. [07/19/2018 16:12:35 INFO 139838496249664] Epoch[0] Train-accuracy=0.950893
  38. [07/19/2018 16:12:35 INFO 139838496249664] Epoch[0] Train-top_k_accuracy_5=0.982537
  39. [07/19/2018 16:12:35 INFO 139838496249664] Epoch[0] Time cost=94.606
  40. [07/19/2018 16:12:45 INFO 139838496249664] Epoch[0] Validation-accuracy=0.981641
  41. [07/19/2018 16:12:45 INFO 139838496249664] Epoch[0] Validation-top_k_accuracy_5=1.000000
  42. [07/19/2018 16:13:02 INFO 139838496249664] Epoch[1] Batch [20]#011Speed: 161.09 samples/sec#011accuracy=0.983259#011top_k_accuracy_5=1.000000
  43. [07/19/2018 16:13:18 INFO 139838496249664] Epoch[1] Batch [40]#011Speed: 159.29 samples/sec#011accuracy=0.986662#011top_k_accuracy_5=1.000000
  44. [07/19/2018 16:13:34 INFO 139838496249664] Epoch[1] Batch [60]#011Speed: 159.31 samples/sec#011accuracy=0.987961#011top_k_accuracy_5=1.000000
  45. [07/19/2018 16:13:50 INFO 139838496249664] Epoch[1] Batch [80]#011Speed: 159.18 samples/sec#011accuracy=0.989294#011top_k_accuracy_5=1.000000
  46. [07/19/2018 16:14:06 INFO 139838496249664] Epoch[1] Batch [100]#011Speed: 159.32 samples/sec#011accuracy=0.990486#011top_k_accuracy_5=1.000000
  47. [07/19/2018 16:14:21 INFO 139838496249664] Epoch[1] Train-accuracy=0.990743
  48. [07/19/2018 16:14:21 INFO 139838496249664] Epoch[1] Train-top_k_accuracy_5=1.000000
  49. [07/19/2018 16:14:21 INFO 139838496249664] Epoch[1] Time cost=95.523
  50. [07/19/2018 16:14:21 INFO 139838496249664] Saved checkpoint to "/opt/ml/model/image-classification-0002.params"
  51. [07/19/2018 16:14:30 INFO 139838496249664] Epoch[1] Validation-accuracy=0.975962
  52. [07/19/2018 16:14:30 INFO 139838496249664] Epoch[1] Validation-top_k_accuracy_5=1.000000
  53. [07/19/2018 16:14:47 INFO 139838496249664] Epoch[2] Batch [20]#011Speed: 160.64 samples/sec#011accuracy=0.996652#011top_k_accuracy_5=1.000000
  54. [07/19/2018 16:15:03 INFO 139838496249664] Epoch[2] Batch [40]#011Speed: 158.53 samples/sec#011accuracy=0.997713#011top_k_accuracy_5=1.000000
  55. [07/19/2018 16:15:20 INFO 139838496249664] Epoch[2] Batch [60]#011Speed: 158.82 samples/sec#011accuracy=0.998207#011top_k_accuracy_5=1.000000
  56. [07/19/2018 16:15:36 INFO 139838496249664] Epoch[2] Batch [80]#011Speed: 158.45 samples/sec#011accuracy=0.998650#011top_k_accuracy_5=1.000000
  57. [07/19/2018 16:15:52 INFO 139838496249664] Epoch[2] Batch [100]#011Speed: 158.64 samples/sec#011accuracy=0.998685#011top_k_accuracy_5=1.000000
  58. [07/19/2018 16:16:06 INFO 139838496249664] Epoch[2] Train-accuracy=0.998687
  59. [07/19/2018 16:16:06 INFO 139838496249664] Epoch[2] Train-top_k_accuracy_5=1.000000
  60. [07/19/2018 16:16:06 INFO 139838496249664] Epoch[2] Time cost=95.844
  61. [07/19/2018 16:16:16 INFO 139838496249664] Epoch[2] Validation-accuracy=0.986779
  62. [07/19/2018 16:16:16 INFO 139838496249664] Epoch[2] Validation-top_k_accuracy_5=1.000000
  63. [07/19/2018 16:16:33 INFO 139838496249664] Epoch[3] Batch [20]#011Speed: 160.93 samples/sec#011accuracy=1.000000#011top_k_accuracy_5=1.000000
  64. [07/19/2018 16:16:49 INFO 139838496249664] Epoch[3] Batch [40]#011Speed: 158.54 samples/sec#011accuracy=1.000000#011top_k_accuracy_5=1.000000
  65. [07/19/2018 16:17:05 INFO 139838496249664] Epoch[3] Batch [60]#011Speed: 158.50 samples/sec#011accuracy=1.000000#011top_k_accuracy_5=1.000000
  66. [07/19/2018 16:17:21 INFO 139838496249664] Epoch[3] Batch [80]#011Speed: 158.66 samples/sec#011accuracy=1.000000#011top_k_accuracy_5=1.000000
  67. [07/19/2018 16:17:37 INFO 139838496249664] Epoch[3] Batch [100]#011Speed: 158.55 samples/sec#011accuracy=1.000000#011top_k_accuracy_5=1.000000
  68. [07/19/2018 16:17:52 INFO 139838496249664] Epoch[3] Train-accuracy=1.000000
  69. [07/19/2018 16:17:52 INFO 139838496249664] Epoch[3] Train-top_k_accuracy_5=1.000000
  70. [07/19/2018 16:17:52 INFO 139838496249664] Epoch[3] Time cost=95.881
  71. [07/19/2018 16:17:52 INFO 139838496249664] Saved checkpoint to "/opt/ml/model/image-classification-0004.params"
  72. [07/19/2018 16:18:02 INFO 139838496249664] Epoch[3] Validation-accuracy=0.985978
  73. [07/19/2018 16:18:02 INFO 139838496249664] Epoch[3] Validation-top_k_accuracy_5=1.000000
  74. [07/19/2018 16:18:18 INFO 139838496249664] Epoch[4] Batch [20]#011Speed: 160.53 samples/sec#011accuracy=0.999256#011top_k_accuracy_5=1.000000
  75. [07/19/2018 16:18:35 INFO 139838496249664] Epoch[4] Batch [40]#011Speed: 158.30 samples/sec#011accuracy=0.999619#011top_k_accuracy_5=1.000000
  76. [07/19/2018 16:18:51 INFO 139838496249664] Epoch[4] Batch [60]#011Speed: 158.19 samples/sec#011accuracy=0.999616#011top_k_accuracy_5=1.000000
  77. [07/19/2018 16:19:07 INFO 139838496249664] Epoch[4] Batch [80]#011Speed: 158.13 samples/sec#011accuracy=0.999711#011top_k_accuracy_5=1.000000
  78. [07/19/2018 16:19:23 INFO 139838496249664] Epoch[4] Batch [100]#011Speed: 157.71 samples/sec#011accuracy=0.999768#011top_k_accuracy_5=1.000000
  79. [07/19/2018 16:19:38 INFO 139838496249664] Epoch[4] Train-accuracy=0.999803
  80. [07/19/2018 16:19:38 INFO 139838496249664] Epoch[4] Train-top_k_accuracy_5=1.000000
  81. [07/19/2018 16:19:38 INFO 139838496249664] Epoch[4] Time cost=96.173
  82. [07/19/2018 16:19:47 INFO 139838496249664] Epoch[4] Validation-accuracy=0.986378
  83. [07/19/2018 16:19:47 INFO 139838496249664] Epoch[4] Validation-top_k_accuracy_5=1.000000
  84. [07/19/2018 16:20:04 INFO 139838496249664] Epoch[5] Batch [20]#011Speed: 160.12 samples/sec#011accuracy=1.000000#011top_k_accuracy_5=1.000000
  85. [07/19/2018 16:20:20 INFO 139838496249664] Epoch[5] Batch [40]#011Speed: 158.03 samples/sec#011accuracy=1.000000#011top_k_accuracy_5=1.000000
  86. [07/19/2018 16:20:37 INFO 139838496249664] Epoch[5] Batch [60]#011Speed: 158.01 samples/sec#011accuracy=1.000000#011top_k_accuracy_5=1.000000
  87. [07/19/2018 16:20:53 INFO 139838496249664] Epoch[5] Batch [80]#011Speed: 157.88 samples/sec#011accuracy=1.000000#011top_k_accuracy_5=1.000000
  88. [07/19/2018 16:21:09 INFO 139838496249664] Epoch[5] Batch [100]#011Speed: 158.15 samples/sec#011accuracy=1.000000#011top_k_accuracy_5=1.000000
  89. [07/19/2018 16:21:24 INFO 139838496249664] Epoch[5] Train-accuracy=1.000000
  90. [07/19/2018 16:21:24 INFO 139838496249664] Epoch[5] Train-top_k_accuracy_5=1.000000
  91. [07/19/2018 16:21:24 INFO 139838496249664] Epoch[5] Time cost=96.236
  92. [07/19/2018 16:21:24 INFO 139838496249664] Saved checkpoint to "/opt/ml/model/image-classification-0006.params"
  93. [07/19/2018 16:21:34 INFO 139838496249664] Epoch[5] Validation-accuracy=0.986178
  94. [07/19/2018 16:21:34 INFO 139838496249664] Epoch[5] Validation-top_k_accuracy_5=1.000000
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!