What is Apache Mahout?

The Apache Mahout™ project's goal is to build an environment for quickly creating scalable performant machine learning applications.

Latest release version 0.13.0 has

Apache Mahout Samsara Environment includes

• Distributed Algebraic optimizer
• R-Like DSL Scala API
• Linear algebra operations
• Ops are extensions to Scala
• IScala REPL based interactive shell
• Integrates with compatible libraries like MLLib
• Runs on distributed Spark, H2O, and Flink
• fastutil to speed up sparse matrix and vector computations
• Matrix to tsv conversions for integration with Apache Zeppelin
• Computation on GPU for speed improvements
• Algorithm development framework for simplicity and coherent workflow

Apache Mahout Samsara Algorithms included

• Stochastic Singular Value Decomposition (ssvd, dssvd)
• Stochastic Principal Component Analysis (spca, dspca)
• Distributed Cholesky QR (thinQR)
• Distributed regularized Alternating Least Squares (dals)
• Collaborative Filtering: Item and Row Similarity
• Naive Bayes Classification
• Distributed and in-core

**Apache Mahout software** provides three major features:

• A simple and extensible programming environment and framework for building scalable algorithms
• A wide variety of premade algorithms for Scala + Apache Spark, H2O, Apache Flink
• Samsara, a vector math experimentation environment with R-like syntax which works at scale
• On-GPU compute for performance improvements in large matrix multiplications

Read an [Overview of programming a Mahout Samsara Application][1], learn [How To Contribute to Mahout][2], [report an issue, bug, or suggestion][3] in our JIRA, see the [Samsara bindings for Scala and Spark][4], and [contact us on our mailing lists][5].

**17 April 2017 - Apache Mahout 0.13.0 released**

Mahout 0.13.0 is a major release that includes GPU-compute capabilities along with an algorithm development framework to simplify contributions.

**13 Jun 2016 - Apache Mahout 0.12.2 released**

Mahout 0.12.2 is a minor release following 0.12.1 that provides Matrix exports for Apache Zeppelin integration along with a few minor bug fixes and dependency upgrades.

**19 May 2016 - "Visualizing Apache Mahout in R via Apache Zeppelin" posted**

A tutorial on connecting Mahout, R, Spark, and Zeppelin by Trevor Grant showing how to plot results in Apache Zeppelin.

**18 May 2016 - Apache Mahout 0.12.1 released**

Mahout 0.12.1 is a minor release following 0.12.0 that fixes issues in the newly added Flink backend and supports Flink 1.0.3.

**11 Apr 2016 - Apache Mahout 0.12.0 released**

The Mahout 0.12.0 release marks a major milestone for the “Samsara” environment’s goal of providing an engine neutral math platform by now supporting Flink. While still experimental, the mahout Flink bindings now offer all of the R-Like semantics for linear algebra operations, matrix decompositions, and algorithms of the “Samsara” platform for execution on a Flink back-end.

**12 Mar 2016 - Apache Mahout 0.11.2 released**

**Apache Mahout introduces a new math environment called** [**Samsara**](http://mahout.apache.org/users/sparkbindings/home.html), for its theme of universal renewal. It reflects a fundamental rethinking of how scalable machine learning algorithms are built and customized. Mahout-Samsara is here to help people create their own math while providing some off-the-shelf algorithm implementations. At its core are general linear algebra and statistical operations along with the data structures to support them. You can use is as a library or customize it in Scala with Mahout-specific extensions that look something like R. Mahout-Samsara comes with an interactive shell that runs distributed operations on an Apache Spark cluster. This make prototyping or task submission much easier and allows users to customize algorithms with a whole new degree of freedom.

[**Mahout Algorithms**](http://mahout.apache.org/users/basics/algorithms.html) include many new implementations built for speed on Mahout-Samsara. They run on Spark 1.3+, Flink 1.0.1, and some on H2O, which means as much as a 10x speed increase. You’ll find robust matrix decomposition algorithms as well as a **[Naive Bayes][6]** classifier and collaborative filtering. The new spark-itemsimilarity enables the next generation of **[cooccurrence recommenders][7]** that can use entire user click streams and context in making recommendations.

Our [release notes](http://mahout.apache.org/general/release-notes.html) have details. Interested in helping? Join the Mailing lists.