Differentiable SDE solvers with GPU support and efficient sensitivity analysis.
Two-and-a-half-years since the last release. How time flies. This is primarily a maintenance release.
Note that Diffrax now comes highly recommended. This is a JAX-based suite of numerical ODE+SDE solvers. For most applications it is more advanced than torchsde. (In particular, it is often several times faster.) Whilst torchsde will continue to exist, new projects should consider this alternative.
Full Changelog: https://github.com/google-research/torchsde/compare/v0.2.4...v0.2.6
Efficiency improvements:
f_and_g
and f_and_g_prod
functions so that drift and diffusion can be computed togetherBug fixes:
solver.integrate
twice internallyNew features include
from torchsde.brownian_lib import BrownianPath, BrownianTree
.This codebase provides stochastic differential equation (SDE) solvers with GPU support and efficient sensitivity analysis.