Parallelize AveragePool op #138
Merged
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
Generalize the existing parallel MaxPool implementation and use it to implement both MaxPool and AveragePool. The result is still far from optimal, but it serves as a starting point for implementing pooling ops for each reduction type (max, average, lp) and future data type.
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Outdated suggestions cannot be applied.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.
Generalize the existing parallel MaxPool implementation and use it to implement both MaxPool and AveragePool. This should generalize to LpPool in future too. The result is still far from optimal, but it serves as a starting point for implementing pooling ops for each reduction type (max, average, lp) and future data type.
On a Yolov9e ONNX model taken from HuggingFace this reduces time in AveragePool ops by about 2x on my system (30ms -> 14ms with a 256x256 input).