Reinforcement-learning-with-tensorflow
Reinforcement-learning-with-tensorflow copied to clipboard
bug_issue: A3C环境交互step() 后返回的done 被下面一行判断覆盖了.
https://github.com/MorvanZhou/Reinforcement-learning-with-tensorflow/blob/97dba9bafce7fb5203d395ba77a770fad80931b3/contents/10_A3C/A3C_continuous_action.py#L131
130行环境返回回来的done(游戏是否结束). 被131行 (该episode是否到达最后一步)强行覆盖了. 也就是说,环境里面游戏结束, 这一轮episode也不会结束.