[ONNX] Resize op converter and add half_pixel mode to Resize op #4536
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
masahi
commented
Dec 17, 2019
|
LGTM, nice job! |
jwfromm
approved these changes
Dec 22, 2019
|
should we close this one? |
|
@yzhliu ok, I'll open another PR for onnx bits, when the CI is ready. |
|
@masahi I'm getting this error:
It's from a PyTorch 1.3 -> ONNX v4 opset version 11 model Does it have a different type of option in the Resize operator or isn't the error handling not updated yet? |
|
@Aeroxander if you are using the change in this PR correctly, you shouldn't get that error. ONNX change is not merged upstream, so you need to manually add them to your repo. |
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Outdated suggestions cannot be applied.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.
Implemented a working opset 11 Resize op converter with test cases validated against onnxruntime.
It needs #4538 to be merged.
I removed the "align_corners" option from resize op, and instead introduced "coordinate_transformation_mode" option, which can be
Refer to ONNX Resize op spec for the meaning of these options.
Related discussion:
https://discuss.tvm.ai/t/relay-onnx-frontend-implement-resize-operation/5131/6