Vilonge

Fudan University Shanghai, China

Results 2 comments of


                                            Vilonge

The coppeliasim never shuts down and the PID is constantly changing

Same. You can close it by running `env.close()` of the source RLbench environment.

Actor grad Norms are always NaN (GRPO Training)

> I am facing the same problem. The grad_norm is na after 50 steps. same