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

Add Metrics #6

Merged
merged 10 commits into from
Sep 8, 2023
Merged

Add Metrics #6

merged 10 commits into from
Sep 8, 2023

Conversation

harrykeightley
Copy link
Contributor

Add support for metrics.

  • Default metrics are "wer" and "cer", but can be extended with parameters in the training job.
  • Increase logging around metrics and dataset processing.
  • Eval metrics saved under eval_results.json post training.

elpis/trainer/trainer.py Outdated Show resolved Hide resolved
num_proc=PROCESSOR_COUNT,
)

logger.debug(f"Dataset post prep: {dataset}")
logger.debug(f"Training labels: {dataset['train']['labels']}")
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Logging the labels here shows an array of vectors. Maybe remove this debug line after sorting out the tokenisation.

Copy link
Contributor

@benfoley benfoley left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@harrykeightley harrykeightley merged commit cff43a0 into main Sep 8, 2023
2 checks passed
@harrykeightley harrykeightley deleted the metrics branch September 8, 2023 04:41
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

Successfully merging this pull request may close these issues.

2 participants