AnchorGenerator forces torch.arange in fp32 to allow export to onnx #113
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
torch.arange cannot export to onnx with fp16 precision so it is necessary to force the operation into fp32 and then cast the result to the desired dtype.
|
Hmm I'm not sure why this is failing tests. |
zhiqwang
reviewed
Jun 9, 2021
Codecov Report
@@ Coverage Diff @@
## master #113 +/- ##
=======================================
Coverage 83.54% 83.54%
=======================================
Files 8 8
Lines 784 784
=======================================
Hits 655 655
Misses 129 129
Flags with carried forward coverage won't be shown. Click here to find out more. Continue to review full report at Codecov.
|
|
@zhiqwang thanks for the suggestion, it looks like that fixed it and this does meet our needs. |
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Outdated suggestions cannot be applied.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.
torch.arange cannot export to onnx with fp16 precision so it is necessary to force the operation into fp32 and then cast the result to the desired dtype.
Fixing #107