tune-sklearn v0.5.0

⚠️ tune-sklearn is no longer being maintained

This release (0.5.0) is the last version of the library that will be released by the Ray team. It has been updated to be compatible
with ray>=2.7.x, ray<=2.9.x. The library will not be guaranteed to work with future Ray versions.

The recommended alternative to keep up with the latest features of Ray is to migrate tune-sklearn usage
to the Ray Tune APIs to accomplish the same thing.

Feel free to post an issue on the Ray Github if you run into any issues in migrating.

What's Changed

• Fix ExperimentAnalysis usage for Ray 2.7 by @justinvyu in #272
• [doc] Add auto-generated docs by @justinvyu in #279
• Bump version to 0.5.0 and update Ray Tune lower bound to 2.7.1 by @justinvyu in #281
• Update README with archival message by @justinvyu in #282

Full Changelog: v0.4.6...v0.5.0

tune-sklearn 0.4.6

What's Changed

• Fix error_score argument being ignored by @Yard1 in #259
• Import mlflow integration from new air.integrations module by @justinvyu in #264
• [release] Bump version number to 0.4.6 by @justinvyu in #268

New Contributors

• @justinvyu made their first contribution in #264

Full Changelog: v0.4.5...v0.4.6

tune-sklearn 0.4.5

What's Changed

• Don't store fit_params, use with_parameters for it by @Yard1 in #256
• Bump version to 0.4.5 by @Yard1 in #258

Full Changelog: v0.4.4...v0.4.5

tune-sklearn 0.4.4

What's Changed

• add license metadata to package setup by @jimthompson5802 in #250
• Import correct MLFlow logger by @Yard1 in #251
• Bump version to 0.4.4 by @Yard1 in #252
• Move tune imports to new locations by @Yard1 in #253
• Fix early_stopping=True with BOHB not setting the correct scheduler by @Yard1 in #254

The minimum supported Ray version is 2.0.0 now.

New Contributors

• @jimthompson5802 made their first contribution in #250

Full Changelog: v0.4.3...v0.4.4

tune-sklearn 0.4.3

Changelog

• Move from loggers to logger callbacks by @krfricke in #241
• Disable build_doc by @Yard1 in #242
• Bump version to 0.4.3 by @Yard1 in #243

Full Changelog: v0.4.2...v0.4.3

tune-sklearn 0.4.2

Changelog

• Fix detection of BOHB scheduler by @Yard1 in #219
• Remove max_concurrency from BOHB kwargs by @Yard1 in #220
• Save analysis object by @Yard1 in #224
• Allow 1 as sk_n_jobs, fix cloning issues by @Yard1 in #225
• Fix cloning issues with early_stopping by @Yard1 in #229
• Remove os.path.expanduser (defer to Ray Tune) by @krfricke in #231
• Remove ray_context by @Yard1 in #236
• Use tune.with_parameters by @Yard1 in #237
• Bump version to 0.4.2 by @Yard1 in #238

Full Changelog: v0.4.1...v0.4.2

tune-sklearn 0.4.1 release notes

Releasing a new version of tune-sklearn! This version should be compatible with both Ray master and Ray 1.6.

This is a maintenance release without any public-facing changes.

Changelog:

• Refactor TuneBaseSearchCV.fit (#215)
• Use Tune search spaces for Optuna (#216)

tune-sklearn 0.4.0 release notes

Releasing a new version of tune-sklearn! This version should be compatible with both Ray master and Ray 1.4.

Changelog:

• You can now pass any tune.run params in fit using the tune_params argument (#212)
• Fixed an exception in _is_param_distributions_all_tune_domains (#209)

tune-sklearn 0.3.0 release notes

Releasing a new version of tune-sklearn! This version should be compatible with both Ray master and Ray 1.3.

Changelog:

• Allow any Ray Tune searcher to be passed in (#198) (thanks @Yard1 !)
• You can now specify a name to your run args (#200) (thanks, @rspeare!)
• Optimization: Estimators can be passed via object store IDs (#196)
• Change BaseEstimator to BaseSearchCV (#192) (thanks @timvink!)
• Fix groups for cross validation (#191)
• Random search with tune search spaces previously failed to produce multiple trials (#180)
• Update code snippets in README to showcase ray.tune SearchSpaces (#176) (thanks @mkretsch327 !)
• Positional arg to keyword in pipeline partial fit (#173)

tune-sklearn 0.2.1 -- patch release

• Adds support for scikit-learn 0.24
• Fixes an issue with gradient boosted models and pipelines