dynamicslab/hydrogym
A Gymnasium-compatible reinforcement learning environment platform for data-driven modeling and control of fluid dynamics.
Collecting fresh signals — velocity needs a few days of history.
collecting data…
star history
HydroGym provides 61+ pre-configured reinforcement learning environments for computational fluid dynamics, spanning canonical benchmarks to turbulent flows. It offers a standardized Gymnasium-compatible interface that works with RL libraries such as Stable-Baselines3 and RLlib. The platform supports multiple solver backends including finite element, lattice Boltzmann, finite volume, spectral element, and fully differentiable JAX-based solvers, with MPI-parallelized execution for HPC deployment.
Frequently asked
- What is dynamicslab/hydrogym?
- A Gymnasium-compatible reinforcement learning environment platform for data-driven modeling and control of fluid dynamics.
- Is hydrogym open source?
- Yes — dynamicslab/hydrogym is open source, released under the MIT license.
- What language is hydrogym written in?
- dynamicslab/hydrogym is primarily written in Python.
- How popular is hydrogym?
- dynamicslab/hydrogym has 500 stars on GitHub.
- Where can I find hydrogym?
- dynamicslab/hydrogym is on GitHub at https://github.com/dynamicslab/hydrogym.