Run LCE optimize pass before and after TFLite optimize pass #525
Conversation
This file contains 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
What do these changes do?
This PR changes the converter to also run the LCE optimize pass before the TFLite optimiyations happen. This allows for transforming
bconv -> mul/add -> relutobconv + mul/add -> reluwhich would've been prevented by TFLite optimizations that already fusemul/add -> relu.To make sure that all possible optimizations are applied, I kept the LCE optimize pass after all TFLite optimizations have been applied since they might enable further optimizations from our side (e.g. clipping might be converted to a relu op which could be fused by our transformations.
How Has This Been Tested?
Manually verified that fusion is correctly handled for quicknet and some literature models