Optimize max_pooling for inference for MKL-DNN/IDEEP device #10156
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
yinghai
reviewed
Aug 2, 2018
| mkldnn::prop_kind::forward_training : mkldnn::prop_kind::forward_inference; | ||
|
|
||
| ideep::pooling_forward::compute(X, Y_dims, *Y, | ||
| stride_, kernel_, pad_tl(), pad_br(), algo_); |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
Collaborator
Author
|
@yinghai OK to merge? |
Contributor
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
yinghai has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
goodlux
pushed a commit
to goodlux/pytorch
that referenced
this pull request
Aug 15, 2018
…10156) Summary: Optimize the max_pooling operation for inference path by setting the "inference" flag to the underlying MKL-DNN, saving the computation and store of max indices which is only needed for training. To make the API compatible, training mode is still the default and inference mode is set in the optimizeForIdeep path. Test shows the speed-up of a single max_pooling operation is up to 7X on BDW. Pull Request resolved: pytorch#10156 Differential Revision: D9276755 Pulled By: yinghai fbshipit-source-id: ad533d53aabb8ccb3b592da984d6269d9b794a8a
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Optimize the max_pooling operation for inference path by setting the "inference" flag to the underlying MKL-DNN, saving the computation and store of max indices which is only needed for training. To make the API compatible, training mode is still the default and inference mode is set in the optimizeForIdeep path.
Test shows the speed-up of a single max_pooling operation is up to 7X on BDW.