[Common] Unified Scales for SDPA #3205
Merged
+148
−4
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
github-actions
bot
added
NNCF PT
AlexanderDokuchaev
approved these changes
Jan 22, 2025
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Outdated suggestions cannot be applied.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.
Changes
Include SDPA metatype in scales unification map for all MinMax backends
Reason for changes
Scales were not being unified in quantizers inserted for such a subgraph:
Tests
Template test was created at
tests/cross_fw/test_templates/test_unified_scales.pyThe template tests uses a synthetic SDPA model with a concat operation. Then, it uses the method
_find_quantization_target_pointsof MinMaxQuantization algorithm to return the unified scale groups. These groups are then used for assertion.