Closed nabenabe0928 closed 6 months ago
Attention: 2 lines
in your changes are missing coverage. Please review.
Comparison is base (
8f39413
) 69.53% compared to head (3efee74
) 69.53%. Report is 6 commits behind head on main.
Files | Patch % | Lines |
---|---|---|
optuna_dashboard/_importance.py | 71.42% | 2 Missing :warning: |
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
@c-bata Thank you for your review, I addressed your suggestion and removed the error in CI.
Contributor License Agreement
This repository (
optuna-dashboard
) and Goptuna share common code. This pull request may therefore be ported to Goptuna. Make sure that you understand the consequences concerning licenses and check the box below if you accept the term before creating this pull request.Reference Issues/PRs
This PR aims to add PED-ANOVA, a quick f-ANOVA calculation algorithm, to the first choice of the importance calculation of Optuna Dashboard to enhance UX.
For PED-ANOVA, please check the following PR:
What does this implement/fix? Explain your changes.
This PR introduces PED-ANOVA to Optuna Dashboard if the Optuna version in the environment is 3.6 or later. By using PED-ANOVA, the visualization becomes very quick, so users will not experience very lagged dashboard behavior anymore due to the importance evaluation.
The current Optuna Dashboard would not give anything after several minutes when
n_trials
is huge (e.g.,n_trails > 10000
).Meanwhile, if we use PED-ANOVA, Optuna Dashboard gives the importance plot in a few seconds even with
n_trails > 10000
.