https://github.com/ikostrikov/jaxrl2

https://github.com/ikostrikov/jaxrl2

Last synced: about 1 month ago
JSON representation

• Host: GitHub
• URL: https://github.com/ikostrikov/jaxrl2
• Owner: ikostrikov
• License: mit
• Created: 2022-10-26T04:07:40.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-01-20T18:37:08.000Z (over 1 year ago)
• Last Synced: 2024-02-10T05:39:34.958Z (5 months ago)
• Language: Jupyter Notebook
• Size: 888 KB
• Stars: 38
• Watchers: 5
• Forks: 8
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Lists

• awesome-deep-reinforcement-learning - ikostrikov/jaxrl2 - commit/ikostrikov/jaxrl2?label=last%20update) (Reinforcement Learning (RL) and Deep Reinforcement Learning (DRL) / RL/DRL Algorithm Implementations and Software Frameworks)

README

        
# jaxrl2

If you use JAXRL2 in your work, please cite this repository in publications:

```

@misc{jaxrl,

  author = {Kostrikov, Ilya},

  doi = {10.5281/zenodo.5535154},

  month = {10},

  title = {{JAXRL: Implementations of Reinforcement Learning algorithms in JAX}},

  url = {https://github.com/ikostrikov/jaxrl2},

  year = {2022},

  note = {v2}

}

```

## Installation

Run

```bash

pip install --upgrade pip

pip install -e .

pip install --upgrade "jax[cuda]" -f https://storage.googleapis.com/jax-releases/jax_releases.html  # Note: wheels only available on linux.

```

See instructions for other versions of CUDA [here](https://github.com/google/jax#pip-installation-gpu-cuda).

## Examples

[Here.](examples/)

## Tests

```bash

MUJOCO_GL=egl CUDA_VISIBLE_DEVICES= pytest tests

```

# Acknowledgements 

Thanks to [@evgenii-nikishin](https://github.com/evgenii-nikishin) for helping with JAX. And [@dibyaghosh](https://github.com/dibyaghosh) for helping with vmapped ensembles.