vwxyzjn / cleanrl

High-quality single file implementation of Deep Reinforcement Learning algorithms with research-friendly features (PPO, DQN, C51, DDPG, TD3, SAC, PPG)
http://docs.cleanrl.dev
Other
5.02k stars 575 forks source link

Is SAC exploration-noise used? #361

Closed StoneT2000 closed 1 year ago

StoneT2000 commented 1 year ago

Was looking at the SAC continuous action implementation and I noticed theres an used exploration-noise parameter. Is this used anywhere (or by any other libs?)

https://github.com/vwxyzjn/cleanrl/blob/master/cleanrl/sac_continuous_action.py#L52

Currently trying to compare my own code with OpenRL Benchmark results so was trying to align the hyperparameters.

vwxyzjn commented 1 year ago

Ah this is probably just copied and pasted from the ddpg base implementation. Let me make a PR to remove it.