MADDPG-torch sample/predict action

PaddlePaddle / PARL

A high-performance distributed training framework for Reinforcement Learning

https://parl.readthedocs.io/

Apache License 2.0

3.24k stars 819 forks source link

Closed ShuaibinLi closed 2 years ago

ShuaibinLi commented 2 years ago

add predict func, rewrite sample add run_eval_episode