Drop QDQ around MaxPool at opset 22#27059
Open
mcollinswisc wants to merge 5 commits intomicrosoft:mainfrom
Open
Drop QDQ around MaxPool at opset 22#27059mcollinswisc wants to merge 5 commits intomicrosoft:mainfrom
mcollinswisc wants to merge 5 commits intomicrosoft:mainfrom
Conversation
This file contains hidden or 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
Includes MaxPool at opset 22 in the DropQDQNodesSelector, so that this gets optimized to a full-integer graph. Previously this was only done for MaxPool at opset 12.
Motivation and Context
Right now, at the most recent opsets, this optimization is no longer being applied.
I'd guess MaxPool was limited to only being selected by the DropQDQ action at opset 12 because at earlier opsets MaxPool did not support any integer types. For example at opset 11:
https://github.com/onnx/onnx/blob/main/docs/Changelog.md#MaxPool-11
but opset 12 added support int8/uint8:
https://github.com/onnx/onnx/blob/main/docs/Changelog.md#MaxPool-12
Now a new version has been introduced at opset 22, that still supports int8/uint8:
https://github.com/onnx/onnx/blob/main/docs/Changelog.md#MaxPool-22