Ignoring Gemm -> Q/Dq -> Softmax in GemmToQLinearMatMul transform #1343
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
rahul-tuli
approved these changes
Jan 26, 2023
corey-nm
commented
Jan 26, 2023
bfineran
approved these changes
Jan 26, 2023
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Outdated suggestions cannot be applied.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.
When there is a Softmax after a MatMul, this pass improperly assumes the Q/Dq is for itself, rather than for the softmax.
Original quantized graph
Current deepsparse graph from exporters
Note that this is invalid!!
deepsparse graph after this PR