Statistical modelling is a key element in many parts of physics, especially in High-Energy Physics (HEP). zfit is a Python library for unbinned, likelihood model fitting. Its main computational backend is TensorFlow, an easy-to-use, highly scalable computing library similar to Numpy. zfit provides a high level interface for advanced model building and fitting while also designed with a unified interface to be easily extendable, allowing the usage of custom and cutting-edge developments from the scientific Python ecosystem in a transparent way.
This talk presents zfit and its usability for data analyses in physics, especially in HEP, as well as recent developments and improvements to the library.