Scikit-psl

Added

- PSL classifier
- probability predictions can return confidence intervals
- probability calibration using BetaCalibration
- stages can now be sliced and iterated (\__getitem\__(), \__iter\__())
- Metrics
- Weighted loss metric
- Rank loss metric

Fixed

- PSL more robust against non-standard class labels like "True"/"False" instead of boolean values

Added

- PSL supports Dataframes as inputs

Added

- PSL supports instance weights

Added

- Extended precision at recall function

Added

- Significantly extended the configuration capabilities with predefined features to limit the PSLs searchspace

Changed

- PSL global loss defaults to sum(cascade)
- rewrote/extracted expected entropy calculation

Fixed

- PSL inspect is now more robust

Fixed

- PSL classifier optimization regarding global loss was incorrect