This repository has been archived by the owner on Oct 13, 2021. It is now read-only.
Fix GPT2 UT for transformers==2.8.0 #478
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
…nnx#394) * Efficient-net test cases. * Add DepthwiseConv2d to subclassed model. * disable efficient net test cases.
wenbingl
approved these changes
May 7, 2020
…to gpt2_test
…to gpt2_test
jiafatom
commented
May 7, 2020
…to gpt2_test
…to gpt2_test
…to gpt2_test
…to gpt2_test
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
GPT2 UT works for transformers==2.5.0 on my local dev. Nightly build uses transformers 2.8.0 and causes output shape mismatch for keras prediction
model.predict. This PR fixes it (and it also works for 2.5.0).The onnx inference result matches
model.predict, i.e., a list. But need postprocessing formodel(inputs)because keras outputs one Eagertensor as final result with internal (a tuple of Eagertensors).