celerite2 is a re-write of celerite (
ascl:1709.008), an algorithm for fast and scalable Gaussian Process (GP) Regression in one dimension. celerite2 improves numerical stability and integration with various machine learning frameworks. The implementation includes interfaces in Python and C++, with full support for PyMC (
ascl:1610.016) and JAX (
ascl:2111.002).