Vivarium Fuzzy Checker

This package provides the FuzzyChecker, a tool for statistical “fuzzy” checks of values that are subject to stochastic variation. It uses statistical hypothesis testing to determine whether an observed value in a Vivarium simulation is extreme enough to reject the null hypothesis that the simulation is behaving correctly against a supplied verification or validation target.

It is one of the libraries in the vivarium-suite monorepo, maintained by the Institute for Health Metrics and Evaluation’s Simulation Science team.

See Fuzzy checking for the statistical methodology. Broader background on the team’s verification and validation process lives in the Vivarium Research documentation.