https://github.com/arogozhnikov/hep_ml

Machine Learning for High Energy Physics.
https://github.com/arogozhnikov/hep_ml

boosting-algorithms high-energy-physics machine-learning neural-networks python reweighting-algorithms scikit-learn splot

Last synced: about 19 hours ago
JSON representation

Machine Learning for High Energy Physics.

• Host: GitHub
• URL: https://github.com/arogozhnikov/hep_ml
• Owner: arogozhnikov
• License: other
• Created: 2015-05-05T14:17:36.000Z (over 9 years ago)
• Default Branch: master
• Last Pushed: 2024-10-16T21:51:37.000Z (3 months ago)
• Last Synced: 2025-01-18T10:02:53.398Z (8 days ago)
• Topics: boosting-algorithms, high-energy-physics, machine-learning, neural-networks, python, reweighting-algorithms, scikit-learn, splot
• Language: Jupyter Notebook
• Homepage: https://arogozhnikov.github.io/hep_ml/
• Size: 92.1 MB
• Stars: 180
• Watchers: 17
• Forks: 64
• Open Issues: 23
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# hep_ml

**hep_ml** provides specific machine learning tools for purposes of high energy physics.

[![Run tests](https://github.com/arogozhnikov/hep_ml/actions/workflows/run_tests.yml/badge.svg)](https://github.com/arogozhnikov/hep_ml/actions/workflows/run_tests.yml)

[![PyPI version](https://badge.fury.io/py/hep-ml.svg)](https://badge.fury.io/py/hep-ml)

[![Documentation](https://img.shields.io/badge/documentation-link-blue.svg)](https://arogozhnikov.github.io/hep_ml/)

[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.1247379.svg)](https://doi.org/10.5281/zenodo.1247379)

![hep_ml, python library for high energy physics](https://github.com/arogozhnikov/hep_ml/blob/data/data_to_download/hep_ml_image.png)

### Main features

* uniform classifiers - the classifiers with low correlation of predictions and mass (or some other variable, or even set of variables)

  * __uBoost__ optimized implementation inside

  * __UGradientBoosting__ (with different losses, specially __FlatnessLoss__ is of high interest)

* measures of uniformity (see **hep_ml.metrics**)

* advanced losses for classification, regression and ranking for __UGradientBoosting__ (see **hep_ml.losses**).  

* **hep_ml.nnet** - theano-based flexible neural networks 

* **hep_ml.reweight** - reweighting multidimensional distributions 


  (_multi_ here means 2, 3, 5 and more dimensions - see GBReweighter!)

* **hep_ml.splot** - minimalistic sPlot-ting 

* **hep_ml.speedup** - building models for fast classification (Bonsai BDT)

* **sklearn**-compatibility of estimators.

### Installation

Plain and simple:

```bash

pip install hep_ml

```

If you're new to python and never used `pip`, first install scikit-learn [with these instructions](http://scikit-learn.org/stable/install.html).

### Links

* [documentation](https://arogozhnikov.github.io/hep_ml/)

* [notebooks, code examples](https://github.com/arogozhnikov/hep_ml/tree/master/notebooks)

    - you may need to install `ROOT` and `root_numpy` to run those 

* [repository](https://github.com/arogozhnikov/hep_ml)

* [issue tracker](https://github.com/arogozhnikov/hep_ml/issues)

### Related projects 

Libraries you'll require to make your life easier and HEPpier.

* [IPython Notebook](http://ipython.org/notebook.html) — web-shell for python

* [scikit-learn](http://scikit-learn.org/)  — general-purpose library for machine learning in python

* [numpy](http://www.numpy.org/)  — 'MATLAB in python', vector operation in python. 

    Use it you need to perform any number crunching. 

* [theano](http://deeplearning.net/software/theano/)  — optimized vector analytical math engine in python

* [ROOT](https://root.cern.ch/)  — main data format in high energy physics 

* [root_numpy](http://rootpy.github.io/root_numpy/)  — python library to deal with ROOT files (without pain)

### License

Apache 2.0, `hep_ml` is an open-source library.

### Platforms 

Linux, Mac OS X and Windows are supported.

**hep_ml** supports both python 2 and python 3.