Yoshihiko Ozaki comments

Results 10 comments of


                                            Yoshihiko Ozaki

Add line to Pareto front for multi-objective optimisation

Hi. Thank you for your useful proposal. I leave a brief comment that may be helpful. This is not mandatory but can be our future work. In this case, I...

Support multi-objective CMA-ES (MOCMAES) sampling algorithm

@hnanacc @tenzen-y MO-CMA-ES is now available on [OptunaHub](https://hub.optuna.org/)! - [Multi-objective CMA-ES (MO-CMA-ES) Sampler](https://hub.optuna.org/samplers/mocma/) You can use `MoCmaSampler` as follows. ```python import optunahub sampler = optunahub.load_module("samplers/mocma").MoCmaSampler(popsize=100, seed=42) ```

Introduce `UPSERT` in `set_trial_user_attr`

@not522 @nabenabe0928 Could you review this PR?

Run tests with Python 3.13

@HideakiImamura Could you review this PR?

Adding optunahub

👍

Reduce `SELECT` statements by passing `study_id` to `check_and_add` in `TrialParamModel`

@nabenabe0928 Could you review this PR?

Add a list of supported storage for artifact store

@c-bata, @porink0424 Could you review this PR?

show trend in plot_optimization_history

Thank you for your proposal. I'd like to point out that introducing this feature into Optuna's plot_optimization_history requires discussion. There are several approaches to calculating the trend, and it is...

show trend in plot_optimization_history

@Param1304 Thank you for your contribution. Would you please make a pull request introducing a package that implements the feature to [optunahub-registry](https://github.com/optuna/optunahub-registry) instead of optuna following https://github.com/optuna/optuna/issues/5531#issuecomment-2242165309?

Save used sampler for each parameter in Trial

Regarding AutoSampler, the sampler used for each trial is tracked. https://github.com/optuna/optunahub-registry/blob/main/package/samplers/auto_sampler/_sampler.py#L245-L246