Optbinning

Improvements:

- Adjust plot size 244.
- Save optimal binning object in JSON format 96.
- Plot IV/WoE metric in binning table plot for binary and continuous target.

Bugfixes:

- Keep pandas.DataFrame index in transform method 286.
- Fix BinningProcess's binning_transform_params="bins" 266 .

Bugfixes:

- Fix numpy array object 229
- Fix ``show_bin_labels`` 262
- Fix ``special_codes_y`` 263

---------------------------

Improvements:

- Implement ``sample_weight`` check in Scorecard class 228.

Bugfixes:

- Fix ``metric_missing`` ignored in Scorecard class 226.

Dependencies:

- Update RoPWR required version.

Improvements:

- Modify max-pvalue and min_diff constraints for CP and MIP formulation to avoid suboptimal solutions.

Bugfixes:

- Use keyword arguments in ``compute_class_weight`` 222.
- Remove preprocessing step when monotonic trend in (ascending, descending) for scenario-based binning 216 .

Dependencies:

- Update scikit-learn and ortools required versions.

New features

- Add parameter ``cat_unknown`` to assign values to the unobserved categories during training.

Improvements

- Add method ``decision_function`` to ``Scorecard`` 198.

New features:

- Optimize formulation of minimum difference constraints for all optimal binning classes and support these constraints regardless of the monotonic trend 201.

- Implementation of sample weight for ``ContinuousOptimalBinning`` 131.


Bugfixes:

- Fix ``ContinuousOptimalBinning`` prebinning step when no prebinning splits were generated 205.