`python-m5p` - M5 Prime regression trees in python, compliant with scikit-learn
summary
An actual implementation of M5 Prime exists as a scikit‑learn style estimator claiming scikit-learn compliant behavior and published as version 0.3.0 available on PyPI. It builds piecewise linear regression trees with linear models at leaves, implements pruning and Quinlan’s smoothing, reproduces the canonical two‑segment artificial example (one split, two linear models) after pruning, exposes open issues about exact benchmark numbers from the original papers, and omits some advanced options.
AI-written summary · about
chapters (6)
related talks
34mPrediction intervals
01h 26mProbabilistic Programming and Bayesian Computing with PyMC
40mCART: Not only Classification and Regression Trees
36mBuild prod ML systems with feature /training /inference pipelines
26mMLbase: A Distributed Machine Learning System
30mBoosting AI Reliability- Uncertainty Quantification with MAPIE
09mMultiSPPL: extending SPPL with multivariate leaf nodes
01h 17mPomegranate: fast and flexible probabilistic models in python