andrewherren / scikit-tree

Flexible library for implementing and experimenting with tree methods, based on the scikit-learn tree codebase

scikit-tree

scikit-tree is a project designed to make experimentation with tree-based machine learning methods straightforward. It relies on, and extends, the tree code in scikit-learn.

Installation

This project is currently brand new and not yet available on PyPI or conda-forge. The only way to install is to build the source package, as detailed below.

Installing from source

Building scikit-tree from the Github source largely follows the instructions given in scikit-learn's "advanced installation" instructions. For MacOS and Linux this can be accomplished by installing python dependencies, compilers, and C/C++ dependencies into a conda environment via conda-forge:

conda create -n partition_env -c conda-forge python=3.10 \
    numpy scipy cython pytest matplotlib pandas scikit-learn \
    joblib threadpoolctl pytest compilers llvm-openmp

conda activate partition_env

cd ~/[path to folder]/scikit-tree
python setup.py clean
pip install --no-build-isolation -e .

Getting Started

The easiest way to get up and running with scikit-tree, once installed as above, is to run a script provided in the examples subfolder of the project. For example:

conda activate partition_env
cd ~/[path to folder]/scikit-tree
python -m examples.tree.plot_regression_tree