Hideaki Imamura comments

Results 67 comments of


                                            Hideaki Imamura

Refactor `_tell.py`

@knshnb @contramundum53 Could you review this PR?

Change the default independent_sampler of CmaEsSampler

https://github.com/optuna/optuna/issues/2964#issuecomment-931131963 As above, currently, we have several benchmark task candidates. For those tasks, we should conduct a experiment to compare each `independent_sampler`.

Change the default independent_sampler of CmaEsSampler

We have decided to postpone this task after V3.

Unify tests for `plot_param_importances`

@contramundum53 Could you review this PR?

Unify tests for `plot_param_importances`

Note that https://github.com/optuna/optuna/pull/3765 should be merged at first.

Unify tests for `plot_param_importances`

@nzw0301 Could you update the PR to resolve the conflict?

Gotchas with the MaxTrialsCallback

Thanks for the raising this issue. It would be great if you give us a PR to fix this.

Simplify the after_trial arguments

Since there was no particular community input on this item, we will not be working on this item, at least not in V3 as well as #3251.

Extend the type annotation of `catch` argument of `study.optimize`

This issue is still an open. Please feel free to submit a PR. Let me remove the good-first-issue label since this issue has a slightly larger context.

Add mypy's `--strict` option

@29Takuya Thanks for the investigation. I think we should skip `--disallow-untyped-decorators` too.