layout: global displayTitle: SystemML Documentation title: SystemML Documentation description: SystemML Documentation
SystemML is now an Apache Incubator project! Please see the Apache SystemML (incubating) website for more information.
SystemML is a flexible, scalable machine learning system. SystemML's distinguishing characteristics are:
- Algorithm customizability via R-like and Python-like languages.
- Multiple execution modes, including Spark MLContext, Spark Batch, Hadoop Batch, Standalone, and JMLC.
- Automatic optimization based on data and cluster characteristics to ensure both efficiency and scalability.
The SystemML GitHub README describes building, testing, and running SystemML. Please read Contributing to SystemML to find out how to help make SystemML even better!
To download SystemML, visit the downloads page.
- Spark MLContext - Spark MLContext is a programmatic API for running SystemML from Spark via Scala, Python, or Java.
- Spark Batch - Algorithms are automatically optimized to run across Spark clusters.
- Hadoop Batch - Algorithms are automatically optimized when distributed across Hadoop clusters.
- Standalone - Standalone mode allows data scientists to rapidly prototype algorithms on a single machine in R-like and Python-like declarative languages.
- The Standalone Guide provides examples of algorithm execution in Standalone Mode.
- JMLC - Java Machine Learning Connector.
- Algorithms Reference - The Algorithms Reference describes the machine learning algorithms included with SystemML in detail.
- Debugger Guide - SystemML supports DML script-level debugging through a command-line interface.
- IDE Guide - Useful IDE Guide for Developing SystemML.