Younggyo Seo c354ead107 Optimized codebase to speed up training (#20) ...

- Modified codes to be compatible with torch.compile
- Modified empirical normalizer to use in-place operator to avoid costly __setattr__
- Parallel soft Q-update
- As a default option, we disabled gradient norm clipping as it is quite expensive

2025-07-02 19:39:02 -07:00

..

environments

Bug fix -- MTBench evaluation and missing code (#18)

2025-06-25 09:21:04 -07:00

__init__.py

Initial Public Release

2025-05-29 01:49:23 +00:00

fast_td3_deploy.py

Support FastTD3 + SimbaV2 (#13)

2025-06-15 12:49:59 -07:00

fast_td3_simbav2.py

Optimized codebase to speed up training (#20)

2025-07-02 19:39:02 -07:00

fast_td3_utils.py

Optimized codebase to speed up training (#20)

2025-07-02 19:39:02 -07:00

fast_td3.py

Optimized codebase to speed up training (#20)

2025-07-02 19:39:02 -07:00

hyperparams.py

Optimized codebase to speed up training (#20)

2025-07-02 19:39:02 -07:00

train.py

Optimized codebase to speed up training (#20)

2025-07-02 19:39:02 -07:00

training_notebook.ipynb

Bug fix -- MTBench evaluation and missing code (#18)

2025-06-25 09:21:04 -07:00