About the ML-SPL API

The Splunk Machine Learning Toolkit (MLTK) ships with more than 40 algorithms including Birch, Lasso, DensityFunction, and RandomForestClassifier. If you want to use a custom algorithm in MLTK, do so using the machine learning extensibility API (ML-SPL API). To learn more about the algorithms packaged with MLTK, see Algorithms in the Machine Learning Toolkit in the User Guide.

You can also extend MLTK with over 300 open source Python algorithms from scikit-learn, pandas, statsmodel, numpy, and scipy libraries. These open source algorithms are available for MLTK through the Python for Scientific Computing (PSC) add-on.

The PSC add-on is a requirement for the MLTK app.

To add a custom algorithm to MLTK, you must write a Python class and register it to the MLTK app. Coding knowledge, advanced Python experience, or development experience is an asset when adding custom algorithms. You can also choose to package your custom algorithm as a separate app and share it on Splunkbase so it can be used by other Splunk platform users.

Add algorithms using GitHub

On-prem users can use GitHub to add more algorithms to MLTK. See the MLTK GitHub repo.

The Machine Learning Toolkit and Python for Scientific computing add-on must be installed in order for GitHub to work in your Splunk platform environment.

You can share and reuse algorithms in the Splunk Community for MLTK on GitHub. In the community you can also learn about new machine learning algorithms from the Splunk open source community, and help fellow users of MLTK.

Splunk Cloud Platform customers can also use GitHub to add more algorithms. The Splunk GitHub for Machine learning app provides access to custom algorithms and is based on the Machine Learning Toolkit open source repo.

Splunk Cloud Platform customers must create a support ticket to have the app installed.