IRIS-HEP Fellow: Peter Ridolfi
Fellowship dates: May – Aug, 2022
Home Institution: University of Pittsburgh
Project: Designing and implementing a converting tool for statistical models between pyhf and CMS combine
The main goal of this project is to develop a tool that actively identifies and accommodates similarities and nuances between pyhf and CMS Combine models. The tool must be able to convert the given parameters form one type of model to the other type, while maintaining a high degree of accuracy between the actual functionality and predictions that are made by the models. Additionally, the tool should be able to give feedback concerning expected differences in these outputs based on the parameters that are given, and the limitations of the tool itself. Ultimately, the tool will serve as another method for better understanding the models that are created and the predictions that they make.GitHub url: https://github.com/peterridolfi/Pyhf-to-Combine-converter.git
More information: My project proposal
Mentors:
-
Matthew Feickert (University of Wisconsin-Madison)
-
Alexander Held (University of Wisconsin-Madison)
-
Kyle Cranmer (NYU / University of Wisconsin-Madison)
- 15 Jun 2022 - "Pyhf to Combine Converter", Peter Ridolfi, IRIS-HEP Fellows Projects Lightning Talks 2022
- 21 Sep 2022 - "Designing and implementing a converting tool for statistical models between pyhf and CMS combine", Peter Ridolfi, IRIS-HEP Fellows Presentations 2022 Recording: Designing and implementing a converting tool for statistical models between pyhf and CMS combine
Current Status
August 2022 - Undergraduate Teaching Assistant at University of Pittsburgh
Contact me: