29 July 2019 to 2 August 2019
Northeastern University
US/Eastern timezone

pyhf: a pure Python statistical fitting library from the high energy physics community with tensors and autograd

30 Jul 2019, 14:40
20m
Shillman 425 (Northeastern University)

Shillman 425

Northeastern University

Oral Presentation Computing, Analysis Tools, & Data Handling Computing, Analysis Tools, & Data Handling

Speaker

Giordon Holtsberg Stark (University of California,Santa Cruz (US))

Description

The HistFactory p.d.f. template $\href{https://cds.cern.ch/record/1456844}{\text{[CERN-OPEN-2012-016]}}$ is per-se independent of its implementation in ROOT and it is useful to be able to run statistical analysis outside of the ROOT, RooFit, RooStats framework. pyhf is a pure-python implementation of that statistical model for multi-bin histogram-based analysis and its interval estimation is based on the asymptotic formulas of "Asymptotic formulae for likelihood-based tests of new physics" $\href{https://arxiv.org/abs/1007.1727}{\text{[arxiv:1007.1727]}}$. pyhf supports modern computational graph libraries such as TensorFlow and PyTorch in order to make use of features such as auto-differentiation and GPU acceleration.

Primary authors

Giordon Holtsberg Stark (University of California,Santa Cruz (US)) Matthew Feickert (Southern Methodist University (US)) Lukas Alexander Heinrich (CERN)

Presentation Materials