Implementation of Random Forests for ordinal regression as described in Ordinal Forests (2019)[1],[2] in Python. In addition, an ordinal-specific loss function, the All-Thresholds Loss, is implemented as the default metric for OOB selection of candidate z-intervals (motivated by the Loss Function (All-Threshold Loss) for ordinal response variables discussion[3]) as the z-scoring inherently being done by Ordinal Forests provide convenient thresholding for this loss.
Based on scikit-learn version 1.1.2[4].
Sources:
[1] https://link.springer.com/article/10.1007/s00357-018-9302-x
[2] https://cran.r-project.org/web/packages/ordinalForest/index.html