PyCC.id: A package for hypothesis-driven equation discovery with structural identifiability
Researchers have developed PyCC.id, a Python library designed to aid in the discovery of ordinary differential equations (ODEs) from time-series data. This tool addresses the challenge of ill-conditioned inverse problems in equation discovery by allowing users to incorporate hypotheses and constraints, such as characteristic curve skeletons, into the model refinement process. The library supports various equation discovery paradigms, including neural networks and symbolic regression, and leverages structural identifiability properties to validate candidate models. AI
IMPACT Provides a new tool for researchers to discover governing equations from data, potentially accelerating scientific modeling and simulation.