singa-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From GitBox <...@apache.org>
Subject [GitHub] [singa] chrishkchris commented on issue #651: Add new example APIs
Date Mon, 06 Apr 2020 06:58:12 GMT
chrishkchris commented on issue #651: Add new example APIs
URL: https://github.com/apache/singa/pull/651#issuecomment-609601411
 
 
   It is now able to support (cnn|resnet) + (cifar10|mnist)
   
   ```
   root@877a3759b148:~/dcsysh/singa/examples/autograd# python3 train.py cnn cifar10
   3
   Starting Epoch 0:
   Training loss = 1334.210571, training accuracy = 0.383543
   Evaluation accuracy = 0.497997, Elapsed Time = 17.269876s
   Starting Epoch 1:
   Training loss = 1080.154541, training accuracy = 0.507542
   Evaluation accuracy = 0.568810, Elapsed Time = 17.324370s
   Starting Epoch 2:
   Training loss = 974.208618, training accuracy = 0.555598
   Evaluation accuracy = 0.612680, Elapsed Time = 17.230355s
   Starting Epoch 3:
   Training loss = 907.781128, training accuracy = 0.585147
   Evaluation accuracy = 0.625901, Elapsed Time = 17.264113s
   Starting Epoch 4:
   Training loss = 858.995056, training accuracy = 0.609435
   Evaluation accuracy = 0.650841, Elapsed Time = 17.296929s
   Starting Epoch 5:
   Training loss = 819.569641, training accuracy = 0.630642
   Evaluation accuracy = 0.663862, Elapsed Time = 17.185191s
   Starting Epoch 6:
   Training loss = 791.844910, training accuracy = 0.643606
   Evaluation accuracy = 0.681991, Elapsed Time = 17.057992s
   Starting Epoch 7:
   Training loss = 763.051758, training accuracy = 0.657690
   Evaluation accuracy = 0.688301, Elapsed Time = 16.876048s
   Starting Epoch 8:
   Training loss = 735.888611, training accuracy = 0.671195
   Evaluation accuracy = 0.686999, Elapsed Time = 16.930443s
   Starting Epoch 9:
   Training loss = 715.715027, training accuracy = 0.680778
   Evaluation accuracy = 0.704928, Elapsed Time = 16.901450s
   
   root@877a3759b148:~/dcsysh/singa/examples/autograd# python3 train.py resnet mnist --bs
32
   Starting Epoch 0:
   Training loss = 682.914490, training accuracy = 0.892883
   Evaluation accuracy = 0.958233, Elapsed Time = 302.701322s
   
   ```

----------------------------------------------------------------
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


With regards,
Apache Git Services

Mime
View raw message