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

predict.py expects tags by default #19

Closed
kai-car opened this issue Aug 6, 2024 · 1 comment
Closed

predict.py expects tags by default #19

kai-car opened this issue Aug 6, 2024 · 1 comment

Comments

@kai-car
Copy link
Collaborator

kai-car commented Aug 6, 2024

I tried starting inference on a trained model on the cluster (using Pegasus Bridle wrapper) with the following command:

pega_run python src/predict.py dataset=drugprot_prepared +pipeline.device=0 model_name_or_path=./models/drugprot/re_text_classification_with_indices/2024-08-01_11-21-41 dataset_split=test_background

I received the following warning and the process stopped:

[2024-08-06 09:54:28,173][src.utils.task_utils][INFO] - Enforcing tags! <cfg.extras.enforce_tags=True>
[2024-08-06 09:54:28,197][src.utils.rich_utils][WARNING] - No tags provided in config. Prompting user to input tags...

I solved this through adding the argument extras.enforce_tags=False to the initial command, as the default configuration in extras/default.yaml is True. In the end the command looks like this:

pega_run python src/predict.py dataset=drugprot_prepared +pipeline.device=0 model_name_or_path=./models/drugprot/re_text_classification_with_indices/2024-08-01_11-21-41 dataset_split=test_background extras.enforce_tags=False
@kai-car
Copy link
Collaborator Author

kai-car commented Aug 7, 2024

This issue is not directly related to Pegasus Bridle, but rather to the PyTorch-IE-Hydra-Template. Therefore, a new issue #175 was created in the respective repo.

@kai-car kai-car closed this as not planned Won't fix, can't repro, duplicate, stale Aug 7, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant