🧵 CORL is an Offline Reinforcement Learning library that provides high-quality and easy-to-follow single-file implementations of SOTA ORL algorithms. Each implementation is backed by a research-friendly codebase, allowing you to run or tune thousands of experiments. Heavily inspired by cleanrl for online RL, check them out too!

• 📜 Single-file implementation
• 📈 Benchmarked Implementation for N algorithms
• 🖼 Weights and Biases integration

git clone https://github.com/tinkoff-ai/CORL.git && cd CORL
pip install -r requirements/requirements_dev.txt

# alternatively, you could use docker
docker build -t <image_name> .
docker run gpus=all -it --rm --name <container_name> <image_name>

For learning curves and all the details, you can check the links above. Here, we report reproduced final and best scores. Note that thay differ by a big margin, and some papers may use different approaches not making it always explicit which one reporting methodology they chose.

If you use CORL in your work, please use the following bibtex

@inproceedings{
tarasov2022corl,
  title={{CORL}: Research-oriented Deep Offline Reinforcement Learning Library},
  author={Denis Tarasov and Alexander Nikulin and Dmitry Akimov and Vladislav Kurenkov and Sergey Kolesnikov},
  booktitle={3rd Offline RL Workshop: Offline RL as a ''Launchpad''},
  year={2022},
  url={https://openreview.net/forum?id=SyAS49bBcv}
}