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

Bug/challenger comparisons #45

Merged
merged 3 commits into from
Aug 28, 2024
Merged

Bug/challenger comparisons #45

merged 3 commits into from
Aug 28, 2024

Conversation

thorrester
Copy link
Member

Pull Request

Short Summary

The ModelChallenger does not currently handle failures well if a model, metric or run is missing. This PR adds better handling logic.

New Challenger Flow:

Scenario: No champion provided

  • In the case when no champion is provided, ModelChallenger will attempt to load the latest model
    • If no model is found --> challenger wins
    • If models are found, ModelChallenger will return the first version that is not the challenger version and has a runcard
    • if a model is found but there is no associated runcard and metrics --> challenger wins
    • If a model and runcard exists, but the metric does not --> challenger wins
    • If model, runcard and metric exists --> run battle report

Scenario: List of champions provided

  • If a champion model does not exist --> challenger wins
  • If a champion does not have a runcard --> challenger wins
  • if metric is missing --> challenger wins
  • If model, runcard and metric exist --> run battle report

Copy link

Dependency Review

✅ No vulnerabilities or license issues found.

Scanned Manifest Files

@thorrester thorrester marked this pull request as ready for review August 27, 2024 21:49
@thorrester thorrester merged commit f537388 into main Aug 28, 2024
14 checks passed
@thorrester thorrester deleted the bug/challenger-comparisons branch August 28, 2024 10:42
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.

1 participant