Skip to content

CMS Machine Learning Group

Welcome to the CMS ML documentation hub

A central place for CMS analyzers to find machine learning information relevant to their work, from training best practices to running models in CMSSW.

Explore the guides Skills marketplace

What you'll find here

ML best practices General advice and common pitfalls for CMS analyzers, before, during and after training

Optimization How to optimize a NN: model optimization, feature importance and data augmentation

Inference Direct and indirect inferencing using a variety of ML packages

CMSSW integration How to get a model integrated into CMSSW, with a checklist and successful integrations

Resources Cloud, dataset, FPGA and GPU resources for machine learning in CMS

Newsletters The ML Knowledge Newsletter, featuring machine learning across CMS analyses

New

CMS skills marketplace

Plugins that teach AI coding agents such as Claude Code how CMS analysis is done: CMSSW release areas, NanoAOD with Awkward Array, vector and hist, condor jobs on the LPC, trigger efficiencies and HEPData records. Each skill is grounded in commands tested on cmslpc against real CMS releases and data.

Get started Open on GitHub Contribute a skill

Install in Claude Code

/plugin marketplace add cms-ml/marketplace
/plugin install cms@cms-marketplace

We recently launched v3 of the ML Knowledge Newsletter, located here, and we're also looking for new writers and contributors. If you would like to be featured in this newsletter or would like to nominate someone to be featured, please let us know!

The goal of this site is to give CMS analyzers a centralized place to gather machine learning information relevant to their work. We are not seeking to rewrite external documentation: whenever applicable, we link to it, such as the iML group's HEP Living Review or their ML Resources repository.

If you think we are missing some important information, please contact the ML Knowledge Subgroup!