Pyspark-ds-toolbox

Changed

* Module `pyspark_ds_toolbox.ml.classification.baseline_binary_classfiers` now algo return features scores.

Added

* Module `pyspark_ds_toolbox.ml.feature_importance` with the functions:
* `extract_features_score()`

Changed

* Module `pyspark_ds_toolbox.ml.shap_values` became `pyspark_ds_toolbox.ml.feature_importance.shap_values`

Added

* Module pyspark_ds_toolbox.ml.classification

changed

* Module pyspark_ds_toolbox.ml.eval became pyspark_ds_toolbox.ml.classification.eval

Changed

* [fix] Class pyspark_ds_toolbox.stats.association.Association now can properly receive only numerical or only categorical features.

Added

* CHANGELOG.md file

Changed

* pyspark dependency is now >=3.2
* Class pyspark_ds_toolbox.stats.association.Association now uses pyspark.pandas.frame.DataFrame instead of databricks.koalas.frame.DataFrame.