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.

Sign up for GitHub

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

Jump to bottom

Stopped using the transformers (4.39.3) trainer compute_metrics to calculate custom metrics as large validation sets will crash #54

Merged
merged 3 commits into from
Sep 9, 2024
Merged

Stopped using the transformers (4.39.3) trainer compute_metrics to calculate custom metrics as large validation sets will crash #54

merged 3 commits into from
Sep 9, 2024

Commits on Sep 9, 2024

  1. stopped using the compute_metrics to calculate roc_auc/pr_auc/accurac… 

    …y for the evaluation step, the reason is that for large eval datasets, the evalulation steps can run out of the CPU memory as it keeps all predictions on CPU
    @ChaoPang
    ChaoPang committed Sep 9, 2024
    Configuration menu
    Copy the full SHA
    63db5d6 View commit details
    Browse the repository at this point in the history

  2. added the backward compatibility if the finetuning data is constructe… 

    …d from the OMOP data, where the age_at_index is labelled as age
    @ChaoPang
    ChaoPang committed Sep 9, 2024
    Configuration menu
    Copy the full SHA
    58b761c View commit details
    Browse the repository at this point in the history

  3. updated the docstring of compute_metrics

    @ChaoPang
    ChaoPang committed Sep 9, 2024
    Configuration menu
    Copy the full SHA
    1b1ccb8 View commit details
    Browse the repository at this point in the history