Workflow Feature

A task is added to the workflow. It checks the correctness/existence of a bump in the version when created a new PR.

Bump package version given changes from #251

What's Changed

• Version bump v4.2.1 by @vitorsrg in #253

Full Changelog: 4.2.0...4.2.1

4.2.0

• Enhancement
  • Support numpy>=1.26,<3 (adds numpy 2.x support).
  • Bump scikit-learn range to >=1.5,<1.8 (numpy 2.x support starts in 1.5; this
    raises the minimum supported version, which is a consumer-visible change).
  • Bump shap range to >=0.43,<0.49.
  • Bump lightgbm range to >=4,<5.
• Bug Fix
  • xgb_classification_learner / lgbm_classification_learner: handle SHAP >=0.45
    returning a single 3D ndarray for multiclass TreeExplainer (instead of the legacy
    list of per-class 2D arrays).
  • lgbm_classification_learner: handle SHAP >=0.45 returning a single 2D ndarray
    with a scalar expected_value for binary TreeExplainer (instead of the legacy
    [neg_class, pos_class] pair).
  • lgbm_classification_learner: pass feature_name and categorical_feature to
    lightgbm.Dataset instead of lightgbm.train (required by lightgbm >=4). Note:
    a user-supplied feature_name is now honored at the Dataset level — previously the
    Dataset always used string-cast features regardless of the feature_name argument.
  • cumulative_effect_curve (and the gain / relative-gain / effect_curves functions
    that build on it): use a stable sort when ordering rows by prediction score. Previous
    default (kind="quicksort") produced non-deterministic tie-breaking across numpy
    versions, which changed the rows included in each cumulative prefix when prediction
    had ties.

4.1.0

Migrate fklearn to UV + pyproject.toml

4.1.0rc0

Release candidate for v4.1.0 (UV migration)

4.0.1 - Restricts the maximum version of lightgbm to 4.5

What's Changed

• Bugfix
  • Restricts the maximum version of lightgbm to 4.5
  • Fix build_pipeline to correctly check the default value of the parameters

Full Changelog: 4.0.0...4.0.1

4.0.0 - Add support for python 3.10 and 3.11

This release is just a major bump of libraries. Its main goal was to add support for Python 3.10 and 3.11. We bumped some libraries necessary to achieve that. There wasn't any breaking change in the fklearn modules but their expected behavior can change given that some of the dependent packages were bumped a few major versions.

• Library version changes:
  • Increase the minimum and allow greater versions Sphinx | Before: Sphinx>=1.7.1,<2, Now: Sphinx>=5,<6
  • Reduce the maximum allowed version of shap | Before: shap>=0.43,<1, Now: shap>=0.43,<0.45

4.0.0rc1 - Add support for Python 3.10

Update changelog

4.0.0rc0 - Bump new fklearn version to support Python 3.11

This release is just a major bump of libraries. Its main goal was to add support for Python 3.11.

3.0.0 - Dropping support for python 3.6 and 3.7

This release is just a major bump of libraries. Its main goal was to drop the support for Python 3.6 and 3.7. We bumped all libraries necessary to achieve that. There wasn't any breaking change in the fklearn modules but their expected behavior can change given that many of the dependent packages were bumped a few major versions.