singa-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From GitBox <...@apache.org>
Subject [GitHub] [singa] XJDKC commented on pull request #697: New Model Layer Operator API
Date Mon, 01 Jun 2020 14:54:19 GMT

XJDKC commented on pull request #697:
URL: https://github.com/apache/singa/pull/697#issuecomment-636905268


   ```bash
   $ singa/examples/cnn git:(py-api*) # python train.py --bs 32 --id 1 resnet cifar10
   Starting Epoch 0:
   Training loss = 2851.114990, training accuracy = 0.353993
   Evaluation accuracy = 0.100160, Elapsed Time = 279.040055s
   Starting Epoch 1:
   Training loss = 2029.516113, training accuracy = 0.532110
   Evaluation accuracy = 0.099860, Elapsed Time = 280.957387s
   Starting Epoch 2:
   Training loss = 1621.794800, training accuracy = 0.632702
   Evaluation accuracy = 0.100160, Elapsed Time = 280.972929s
   Starting Epoch 3:
   Training loss = 1348.243774, training accuracy = 0.696063
   Evaluation accuracy = 0.099860, Elapsed Time = 281.006120s
   Starting Epoch 4:
   Training loss = 1157.080811, training accuracy = 0.743718
   Evaluation accuracy = 0.099860, Elapsed Time = 281.074095s
   Starting Epoch 5:
   Training loss = 1004.825439, training accuracy = 0.779389
   Evaluation accuracy = 0.100160, Elapsed Time = 281.058311s
   Starting Epoch 6:
   Training loss = 897.393677, training accuracy = 0.802857
   Evaluation accuracy = 0.100160, Elapsed Time = 281.011532s
   Starting Epoch 7:
   Training loss = 813.738464, training accuracy = 0.820603
   Evaluation accuracy = 0.099860, Elapsed Time = 280.985507s
   Starting Epoch 8:
   Training loss = 749.492859, training accuracy = 0.835307
   Evaluation accuracy = 0.099860, Elapsed Time = 280.996533s
   Starting Epoch 9:
   Training loss = 691.353882, training accuracy = 0.848852
   Evaluation accuracy = 0.099860, Elapsed Time = 280.966963s
   ```


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



Mime
View raw message