google-research/torchsde
PyTorch library providing GPU-accelerated stochastic differential equation solvers with efficient gradient backpropagation.
Not currently ranked — collecting fresh signals.
star history
This library implements differentiable SDE solvers that support GPU computation and automatic differentiation for training neural stochastic differential equation models. It provides the sdeint function to numerically integrate SDEs with customizable drift and diffusion functions, enabling applications like learning latent stochastic processes from data. The library is part of the neural differential equations research area within deep learning.
Frequently asked
- What is google-research/torchsde?
- PyTorch library providing GPU-accelerated stochastic differential equation solvers with efficient gradient backpropagation.
- Is torchsde open source?
- Yes — google-research/torchsde is open source, released under the Apache-2.0 license.
- What language is torchsde written in?
- google-research/torchsde is primarily written in Python.
- How popular is torchsde?
- google-research/torchsde has 1.7k stars on GitHub.
- Where can I find torchsde?
- google-research/torchsde is on GitHub at https://github.com/google-research/torchsde.