singa-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From GitBox <...@apache.org>
Subject [GitHub] [singa] chrishkchris opened a new pull request #694: [WIP] SINGA-510 Add Time Profiling
Date Sun, 10 May 2020 16:09:24 GMT

chrishkchris opened a new pull request #694:
URL: https://github.com/apache/singa/pull/694


   Work in Progress, a trial of test:
   
   ```
   root@71ac539cda77:~/dcsysh/singa/examples/cnn# python3 train.py cnn mnist
   Starting Epoch 0:
   Training loss = 578.907959, training accuracy = 0.796141
   Evaluation accuracy = 0.937400, Elapsed Time = 3.780531s
   Starting Epoch 1:
   Training loss = 232.124695, training accuracy = 0.922609
   Evaluation accuracy = 0.962841, Elapsed Time = 4.049076s
   Starting Epoch 2:
   Training loss = 167.437912, training accuracy = 0.944220
   Evaluation accuracy = 0.971855, Elapsed Time = 3.651554s
   Starting Epoch 3:
   Training loss = 138.634125, training accuracy = 0.953392
   Evaluation accuracy = 0.966747, Elapsed Time = 3.441150s
   Starting Epoch 4:
   Training loss = 117.458504, training accuracy = 0.961096
   Evaluation accuracy = 0.973057, Elapsed Time = 3.560500s
   Starting Epoch 5:
   Training loss = 104.992790, training accuracy = 0.965198
   Evaluation accuracy = 0.979267, Elapsed Time = 3.463262s
   Starting Epoch 6:
   Training loss = 96.263885, training accuracy = 0.967249
   Evaluation accuracy = 0.980369, Elapsed Time = 3.448452s
   Starting Epoch 7:
   Training loss = 89.073364, training accuracy = 0.970051
   Evaluation accuracy = 0.975561, Elapsed Time = 4.008111s
   Starting Epoch 8:
   Training loss = 82.311523, training accuracy = 0.972385
   Evaluation accuracy = 0.980369, Elapsed Time = 4.038986s
   Starting Epoch 9:
   Training loss = 78.408806, training accuracy = 0.974270
   Evaluation accuracy = 0.979968, Elapsed Time = 3.464626s
   cudnnConvForward : 5.03911e-05
   cudnnAddTensor : 1.75784e-05
   ReLU : 1.76721e-05
   GpuPoolingForward : 1.79932e-05
   cudnnConvForward : 8.69407e-05
   cudnnAddTensor : 1.26756e-05
   ReLU : 1.21277e-05
   GpuPoolingForward : 1.33688e-05
   GEMM : 3.86448e-05
   SetValue : 1.50594e-05
   GEMM : 2.04786e-05
   ReLU : 1.08824e-05
   GEMM : 2.76666e-05
   SetValue : 1.17772e-05
   GEMM : 1.52262e-05
   SoftMax : 1.74641e-05
   ComputeCrossEntropy : 1.15164e-05
   SetValue : 1.14181e-05
   SumAll : 2.52074e-05
   Div : 9.46707e-06
   CopyDataToFrom : 1.36201e-05
   SoftmaxCrossEntropyBackward : 1.03136e-05
   Div : 9.12541e-06
   SetValue : 1.14424e-05
   GEMV : 1.30881e-05
   Axpy : 1.0157e-05
   EltwiseMult : 1.07198e-05
   Axpy : 9.32879e-06
   Axpy : 9.56493e-06
   GEMM : 2.18145e-05
   GEMM : 1.80916e-05
   Axpy : 9.61272e-06
   EltwiseMult : 9.28487e-06
   Axpy : 1.01875e-05
   Axpy : 9.85858e-06
   ReLUBackward : 1.11622e-05
   SetValue : 1.15455e-05
   GEMV : 1.23761e-05
   Axpy : 9.40337e-06
   EltwiseMult : 9.53304e-06
   Axpy : 9.34529e-06
   Axpy : 9.59296e-06
   GEMM : 3.12802e-05
   GEMM : 2.18544e-05
   Axpy : 1.56967e-05
   EltwiseMult : 1.50189e-05
   Axpy : 1.54643e-05
   Axpy : 1.82553e-05
   GpuPoolingBackward : 2.87581e-05
   ReLUBackward : 1.45246e-05
   cudnnConvolutionBackwardData : 8.92449e-05
   cudnnConvolutionBackwardFilter : 8.9869e-05
   cudnnConvolutionBackwardBias : 3.91475e-05
   Axpy : 1.09387e-05
   EltwiseMult : 9.88989e-06
   Axpy : 1.10044e-05
   Axpy : 1.1091e-05
   Axpy : 9.35806e-06
   EltwiseMult : 8.96863e-06
   Axpy : 9.43876e-06
   Axpy : 9.37369e-06
   GpuPoolingBackward : 2.98794e-05
   ReLUBackward : 2.63507e-05
   cudnnConvolutionBackwardData : 7.08184e-05
   cudnnConvolutionBackwardFilter : 0.000111166
   cudnnConvolutionBackwardBias : 0.000126415
   Axpy : 9.51455e-06
   EltwiseMult : 9.19408e-06
   Axpy : 9.24186e-06
   Axpy : 9.65924e-06
   Axpy : 9.2399e-06
   EltwiseMult : 8.93243e-06
   Axpy : 9.3894e-06
   Axpy : 9.24875e-06
   ```
   


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