nihui opened a new pull request #10098: implement 
DepthwiseConv2dBackwardFilterKernel from tensorflow codebase
   implement DepthwiseConv2dBackwardFilterKernel from tensorflow codebase, 
which is over 5x faster
   ## Description ##
   The current mxnet implementation calls __syncthreads() function too much, 
which is extemely slow.
   The new code comes from tensorflow, but the variable names are adjusted for 
   My model uses depthwise conv heavily, and now its training time per 
iteration is over 5x faster on single P40 gpu.  ( old 92s vs new 18s )
   ## Checklist ##
   ### Essentials ###
   - [ ] Passed code style checking (`make lint`)
   - [x] 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)
   - [x] Code is well-documented: 
   - [x] To the my best knowledge, examples are either not affected by this 
change, or have been fixed to be compatible with this change

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:

With regards,
Apache Git Services

Reply via email to