use graph transform to deal with more general cases for efficient_conv_bn_eval #1259
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Outdated suggestions cannot be applied.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.
Motivation
Prior to this PR,
efficient_conv_bn_eval
operates in module level. It modifies theforward
function ofconv
andbn
modules. Therefore, it cannot deal with multiple usage ofconv
.This PR improves the
efficient_conv_bn_eval
feature to operate in computation level. We fuse each computation ofconv
andbn
. No changes instate_dict
, no constraints for the network architecture. As long as there areconv
+bn
ineval
computation patterns, this PR works.See the updated test for example.
BC-breaking (Optional)
No. It just changes the implementation of
efficient_conv_bn_eval
.