Offlinerllib

Latest version: v0.1.5

Safety actively analyzes 688578 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

2301.02328

2110.06169

2106.01345

+ [Advantage-Weighted Actor Critic (AWAC)](https://arxiv.org/abs/2006.09359)
+ [TD3-BC](https://arxiv.org/pdf/2106.06860.pdf)
all with consistent implementations and comparable performance to the reported numbers in their white paper.

Check [benchmark_result](https://github.com/typoverflow/OfflineRL-Lib/blob/master/reproduce/benchmark_result.md) for details.

0.0.8

Changes
+ New algorithm: [TD7](https://arxiv.org/abs/2306.02451).
+ Upload benchmark (Gym MuJoCo, DMC) results for online algorithms: SAC, TD3, XSAC, TD7.
+ Benchmark results are available at [lamda-rl](https://wandb.ai/lamda-rl).

0.0.4

Implemented and benchmarked algorithms:
+ [In-Sample Actor Critic (InAC)](https://arxiv.org/abs/2302.14372)

Links

Releases

© 2024 Safety CLI Cybersecurity Inc. All Rights Reserved.