Add nhwc support for NNAPI EP, add concat op, handle concurrent calls to NNAPI model#4356
Merged
guoyu-wang merged 4 commits intomasterfrom Jun 29, 2020
Merged
Add nhwc support for NNAPI EP, add concat op, handle concurrent calls to NNAPI model#4356guoyu-wang merged 4 commits intomasterfrom
guoyu-wang merged 4 commits intomasterfrom
Conversation
snnn
approved these changes
Jun 27, 2020
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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:
NNAPI natively runs more efficiently on nhwc inputs, while onnx uses nchw inputs, although it is supported by NNAPI but all the conv/pool/... operations will fall back to CPU which is much slower
Motivation and Context:
Add internal nchw->nhwc transpose of the inputs when it is necessary (such as the input of conv/pool/...), and only transpose back (nhwc->nchw) only if it is necessary.
Other changes includes in this PR: