pyCSEP: Tools for Earthquake Forecast Developers
PyCSEP tools help earthquake forecast model developers evaluate their forecasts and provide the machinery to implement experiments within CSEP testing centers.
The Collaboratory for the Study of Earthquake Predictability (CSEP) supports an international effort to conduct earthquake forecasting experiments. CSEP supports these activities by developing the cyberinfrastructure necessary to run earthquake forecasting experiments including the statistical framework required to evaluate probabilistic earthquake forecasts.
PyCSEP is a python library that provides tools for (1) evaluating probabilistic earthquake forecasts, (2) working with earthquake catalogs in this context, and (3) creating visualizations. Official experiments that run in CSEP testing centers will be implemented using the code provided by this package.
Help modelers become familiar with formats, procedures, and evaluations used in CSEP Testing Centers.
Provide vetted software for model developers to use in their research.
Provide quantitative and visual tools to assess earthquake forecast quality.
Promote open-science ideas by ensuring transparency and availability of scientific code and results.
Curate benchmark models and data sets for modelers to conduct retrospective experiments of their forecasts.
We highly encourage users of this package to get involved in the development process. Any contribution is helpful, even suggestions on how to improve the package, or additions to the documentation (those are particularly welcome!). Check out the Contribution guidelines for a step by step on how to contribute to the project. If there are any questions, please contact us!
List of Contributors
(Lead) William Savran
wsavran [at] usc [dot] edu
Muhammad Asim Khawaja