(pre-beta stage!) zfit is a complete NLL minimisation library based on TensorFlow (for now) with PDFs. The focus is on scalability, parallelization and ease-of-use (no cython, no C++ needed to extend)
Branch: develop
Clone or download



Documentation Status

Zfit is around it's alpha stage: core parts are established, but some parts are also missing and bugs as well as changes may be encountered. If you already want to use it for a specific project, feel free contact us.


zfit has been developed with support from the University of Zürich and the Swiss National Science Foundation (SNF) under contracs 168169 and 174182.

The zfit structure is based on the framework developed by Anton Poluektov using the TensorFlow (TF) open source library https://gitlab.cern.ch/poluekt/TensorFlowAnalysis

The authors would like to thank numerous invaluable inputs from the theory community, i.e., D. van Dyk, G. Isidori, members of LHCb collaboration, as well as from the Scikit-HEP community.