fix find duplicates in qconfig #1155
Draft
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
elad-c
requested changes
Aug 8, 2024
| @@ -553,7 +553,7 @@ def get_qco(self, tpc: TargetPlatformCapabilities) -> QuantizationConfigOptions: | |||
| # Extract qco with is_match_type to overcome mismatch of function types in TF 2.15 | |||
| matching_qcos = [_qco for _type, _qco in tpc.layer2qco.items() if self.is_match_type(_type)] | |||
| if matching_qcos: | |||
| if len(matching_qcos) > 1: | |||
| if len(matching_qcos) > 1 and matching_qcos[0] != tpc.tp_model.default_qco: | |||
There was a problem hiding this comment.
- Must the deafult be at index 0?
- Add a test to simulate it.
ofirgo
reviewed
Aug 8, 2024
| @@ -553,7 +553,7 @@ def get_qco(self, tpc: TargetPlatformCapabilities) -> QuantizationConfigOptions: | |||
| # Extract qco with is_match_type to overcome mismatch of function types in TF 2.15 | |||
| matching_qcos = [_qco for _type, _qco in tpc.layer2qco.items() if self.is_match_type(_type)] | |||
| if matching_qcos: | |||
| if len(matching_qcos) > 1: | |||
| if len(matching_qcos) > 1 and matching_qcos[0] != tpc.tp_model.default_qco: | |||
There was a problem hiding this comment.
I don't think that this solves the issue correctly.
I suspect that the case generating this issue is when there is a pytorch function that has two aliases, and that one of them appears in a TPC operation set and the other is not, thus returning the default qco.
Please try to reproduce the issue, identify the pytorch layer that has duplicate configs and let's think of a more robust solution.
|
Stale pull request message |
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Outdated suggestions cannot be applied.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.
Pull Request Description:
Use names in is_match method only in tensorflow operations.
Checklist before requesting a review: