Installation

All algorithms provided by this project are published on PyPi. As shown below, they can easily be installed via a Python package manager, such as pip. Unless you intend to modify the algorithms’ source code, in which case you should have a look at the section Building from Source, this is the recommended way for installing the software.

The following compatibility matrix provides an overview of the operating systems and architectures for which the packages mentioned below are currently available:

Operating System	x86-64	ARM64
Linux	✅	✅
macOS	✅	✅
Windows	✅	✅

Examples of how to use the algorithms in your own Python programs can be found in the section Using the Python API.

Installing the BOOMER Algorithm

The gradient boosting algorithm BOOMER is published as the Python package mlrl-boomer. It can be installed via the following command:

pip install mlrl-boomer

The description of the methodology used by the BOOMER algorithm, as well as examples of how to configure it, are given here.

Installing the SeCo Algorithm

In addition to the BOOMER algorithm, this project does also provide a Separate-and-Conquer (SeCo) algorithm based on traditional rule learning techniques that are particularly well-suited for learning interpretable models. It is published as the package mlrl-seco and can be installed as follows:

pip install mlrl-seco

In this section, we elaborate on the techniques utilized by the SeCo algorithm and discuss its parameters.

Installing the Command Line API

To ease the use of the algorithms that are developed by this project, we also provide a command line utility that allows configuring the algorithms and applying them to a given dataset without the need to write code. It is published as the Python package mlrl-testbed-sklearn and can optionally be installed via the following command:

pip install mlrl-testbed-sklearn

For more information about how to use the command line API, refer to the section Using the Command Line API.