ZiyueHuang opened a new pull request #9303: fix wide&deep
URL: https://github.com/apache/incubator-mxnet/pull/9303
 
 
   ## Description ##
   fix wrong name in wide&deep example.
   Final val acc is around 85%
   
   ```
   python train.py
   2018-01-04 08:31:16,394 Namespace(batch_size=100, cuda=False, 
log_interval=100, lr=0.001, num_epoch=10, optimizer='adam')
   2018-01-04 08:31:17,483 Training started ...
   2018-01-04 08:31:18,045 Epoch[0] Batch [100] Speed: 18404.21 samples/sec     
accuracy=0.782500
   2018-01-04 08:31:18,602 Epoch[0] Batch [200] Speed: 17960.93 samples/sec     
accuracy=0.805900
   2018-01-04 08:31:19,136 Epoch[0] Batch [300] Speed: 18731.01 samples/sec     
accuracy=0.836500
   2018-01-04 08:31:19,983 epoch 0, accuracy = 0.843950617284
   2018-01-04 08:31:19,989 Saved checkpoint to "checkpoint-0000.params"
   2018-01-04 08:31:19,993 Saved optimizer state to "checkpoint-0000.states"
   2018-01-04 08:31:20,532 Epoch[1] Batch [100] Speed: 18666.05 samples/sec     
accuracy=0.835400
   2018-01-04 08:31:21,049 Epoch[1] Batch [200] Speed: 19363.37 samples/sec     
accuracy=0.844300
   2018-01-04 08:31:21,586 Epoch[1] Batch [300] Speed: 18625.41 samples/sec     
accuracy=0.847000
   2018-01-04 08:31:22,430 epoch 1, accuracy = 0.85037037037
   2018-01-04 08:31:22,437 Saved checkpoint to "checkpoint-0001.params"
   2018-01-04 08:31:22,441 Saved optimizer state to "checkpoint-0001.states"
   2018-01-04 08:31:22,993 Epoch[2] Batch [100] Speed: 18188.73 samples/sec     
accuracy=0.843000
   2018-01-04 08:31:23,542 Epoch[2] Batch [200] Speed: 18215.66 samples/sec     
accuracy=0.849500
   2018-01-04 08:31:24,107 Epoch[2] Batch [300] Speed: 17709.80 samples/sec     
accuracy=0.846200
   2018-01-04 08:31:24,954 epoch 2, accuracy = 0.846543209877
   2018-01-04 08:31:24,960 Saved checkpoint to "checkpoint-0002.params"
   2018-01-04 08:31:24,963 Saved optimizer state to "checkpoint-0002.states"
   2018-01-04 08:31:25,494 Epoch[3] Batch [100] Speed: 18926.82 samples/sec     
accuracy=0.843200
   2018-01-04 08:31:26,013 Epoch[3] Batch [200] Speed: 19257.54 samples/sec     
accuracy=0.848300
   2018-01-04 08:31:26,517 Epoch[3] Batch [300] Speed: 19850.32 samples/sec     
accuracy=0.854000
   2018-01-04 08:31:27,345 epoch 3, accuracy = 0.847160493827
   2018-01-04 08:31:27,351 Saved checkpoint to "checkpoint-0003.params"
   2018-01-04 08:31:27,354 Saved optimizer state to "checkpoint-0003.states"
   2018-01-04 08:31:27,868 Epoch[4] Batch [100] Speed: 19549.62 samples/sec     
accuracy=0.848300
   2018-01-04 08:31:28,411 Epoch[4] Batch [200] Speed: 18433.86 samples/sec     
accuracy=0.851000
   2018-01-04 08:31:28,928 Epoch[4] Batch [300] Speed: 19351.20 samples/sec     
accuracy=0.856300
   2018-01-04 08:31:29,786 epoch 4, accuracy = 0.847098765432
   2018-01-04 08:31:29,791 Saved checkpoint to "checkpoint-0004.params"
   2018-01-04 08:31:29,795 Saved optimizer state to "checkpoint-0004.states"
   2018-01-04 08:31:30,327 Epoch[5] Batch [100] Speed: 18892.66 samples/sec     
accuracy=0.848300
   2018-01-04 08:31:30,853 Epoch[5] Batch [200] Speed: 19013.75 samples/sec     
accuracy=0.849800
   2018-01-04 08:31:31,372 Epoch[5] Batch [300] Speed: 19268.64 samples/sec     
accuracy=0.859200
   2018-01-04 08:31:32,205 epoch 5, accuracy = 0.849382716049
   2018-01-04 08:31:32,211 Saved checkpoint to "checkpoint-0005.params"
   2018-01-04 08:31:32,215 Saved optimizer state to "checkpoint-0005.states"
   2018-01-04 08:31:32,723 Epoch[6] Batch [100] Speed: 19776.68 samples/sec     
accuracy=0.851000
   2018-01-04 08:31:33,242 Epoch[6] Batch [200] Speed: 19271.31 samples/sec     
accuracy=0.853300
   2018-01-04 08:31:33,777 Epoch[6] Batch [300] Speed: 18675.22 samples/sec     
accuracy=0.859400
   2018-01-04 08:31:34,587 epoch 6, accuracy = 0.84987654321
   2018-01-04 08:31:34,592 Saved checkpoint to "checkpoint-0006.params"
   2018-01-04 08:31:34,596 Saved optimizer state to "checkpoint-0006.states"
   2018-01-04 08:31:35,118 Epoch[7] Batch [100] Speed: 19269.87 samples/sec     
accuracy=0.851500
   2018-01-04 08:31:35,615 Epoch[7] Batch [200] Speed: 20107.65 samples/sec     
accuracy=0.852300
   2018-01-04 08:31:36,163 Epoch[7] Batch [300] Speed: 18262.51 samples/sec     
accuracy=0.859200
   2018-01-04 08:31:36,979 epoch 7, accuracy = 0.850679012346
   2018-01-04 08:31:36,984 Saved checkpoint to "checkpoint-0007.params"
   2018-01-04 08:31:36,987 Saved optimizer state to "checkpoint-0007.states"
   2018-01-04 08:31:37,529 Epoch[8] Batch [100] Speed: 18537.02 samples/sec     
accuracy=0.853200
   2018-01-04 08:31:38,090 Epoch[8] Batch [200] Speed: 17838.41 samples/sec     
accuracy=0.854300
   2018-01-04 08:31:38,619 Epoch[8] Batch [300] Speed: 18902.09 samples/sec     
accuracy=0.859900
   2018-01-04 08:31:39,445 epoch 8, accuracy = 0.849691358025
   2018-01-04 08:31:39,451 Saved checkpoint to "checkpoint-0008.params"
   2018-01-04 08:31:39,454 Saved optimizer state to "checkpoint-0008.states"
   2018-01-04 08:31:39,975 Epoch[9] Batch [100] Speed: 19294.07 samples/sec     
accuracy=0.853200
   2018-01-04 08:31:40,511 Epoch[9] Batch [200] Speed: 18670.72 samples/sec     
accuracy=0.855900
   2018-01-04 08:31:41,022 Epoch[9] Batch [300] Speed: 19576.13 samples/sec     
accuracy=0.860100
   2018-01-04 08:31:41,832 epoch 9, accuracy = 0.850185185185
   2018-01-04 08:31:41,837 Saved checkpoint to "checkpoint-0009.params"
   2018-01-04 08:31:41,841 Saved optimizer state to "checkpoint-0009.states"
   2018-01-04 08:31:41,841 Training completed.
   ```
   cc @eric-haibin-lin 
   
   ## Checklist ##
   ### Essentials ###
   - [ ] Passed code style checking (`make lint`)
   - [ ] Changes are complete (i.e. I finished coding on this PR)
   - [ ] All changes have test coverage:
   - Unit tests are added for small changes to verify correctness (e.g. adding 
a new operator)
   - Nightly tests are added for complicated/long-running ones (e.g. changing 
distributed kvstore)
   - Build tests will be added for build configuration changes (e.g. adding a 
new build option with NCCL)
   - [ ] Code is well-documented: 
   - For user-facing API changes, API doc string has been updated. 
   - For new C++ functions in header files, their functionalities and arguments 
are documented. 
   - For new examples, README.md is added to explain the what the example does, 
the source of the dataset, expected performance on test set and reference to 
the original paper if applicable
   - [ ] To the my best knowledge, examples are either not affected by this 
change, or have been fixed to be compatible with this change
   
   ### Changes ###
   - [ ] Feature1, tests, (and when applicable, API doc)
   - [ ] Feature2, tests, (and when applicable, API doc)
   
   ## Comments ##
   - If this change is a backward incompatible change, why must this change be 
made.
   - Interesting edge cases to note here
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

Reply via email to