Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

GPTJ ran on CUDAExecutionProvider after O4 optimization yields Shape mismatch attempting to re-use buffer #800

Closed
2 of 4 tasks
fxmarty opened this issue Feb 21, 2023 · 2 comments
Labels
bug Something isn't working onnxruntime Related to ONNX Runtime

Comments

@fxmarty
Copy link
Contributor

fxmarty commented Feb 21, 2023

System Info

optimum dev
onnxruntime-gpu 1.14.0

Who can help?

No response

Information

  • The official example scripts
  • My own modified scripts

Tasks

  • An officially supported task in the examples folder (such as GLUE/SQuAD, ...)
  • My own task or dataset (give details below)

Reproduction

it is the only one to fail so kind of weird.

Remove the skipped test

if model_arch == "gptj" and use_cache and optimization_level == "O4":
and run

pytest tests/onnxruntime/test_optimization.py -k "test_optimization_levels_gpu and gptj" -s

Expected behavior

No error.

@fxmarty
Copy link
Contributor Author

fxmarty commented Feb 24, 2023

Probably related: microsoft/onnxruntime#14582

@fxmarty
Copy link
Contributor Author

fxmarty commented Mar 15, 2023

Fixed with #871

@fxmarty fxmarty closed this as completed Mar 15, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working onnxruntime Related to ONNX Runtime
Projects
None yet
Development

No branches or pull requests

1 participant