README License DRL-code-pytorch Concise pytorch implementations of DRL algorithms, including REINFORCE, A2C, Rainbow DQN, PPO(discrete and continuous), DDPG, TD3, SAC, PPO-discrete-RNN(LSTM/GRU). Dependencies p
Concise pytorch implements of DRL algorithms, including REINFORCE, A2C, DQN, PPO(discrete and continuous), DDPG, TD3, SAC. - Initial commit · Lizhi-sjtu/DRL-code-pytorch@487738f
Lizhi-sjtu/DRL-code-pytorchPublic NotificationsYou must be signed in to change notification settings Fork199 Star1.3k Code Issues14 Pull requests1 Actions Projects Security Insights Additional navigation options New issue Open lgzidopened this issueJul 5, 2024· 0 comments ...
使用此处的PPO代码,训练时总是发现critic loss不断增长,甚至会增长到1e18的数量级; 经比较其他地方的PPO代码,怀疑是此处的PPO代码在计算target_value时使用了当前的critic网络来计算batch中state的value, 因此导致值估计越推越高; 将代码改为在replay buffer中存入记录的同时存入state的值估计,而不是在计算target_val...
Lizhi-sjtu / DRL-code-pytorch Public Notifications Fork 177 Star 1.1k Code Issues 14 Pull requests 1 Actions Projects Security Insights New issue 没有测试代码 #4 Open ligvxi opened this issue Feb 15, 2023· 0 comments
Lizhi-sjtu / DRL-code-pytorch Public Notifications Fork 174 Star 1.1k Code Issues 14 Pull requests 1 Actions Projects Security Insights New issue SAC-continuous.py中的log_std #8 Open jsdd25 opened this issue Apr 7, 2023· 0 comments Comments jsdd25 commented Apr 7, 2023 ...