seungeunrho/minimalRL

Issues

• Training speed is very slow！！！

  #62 opened 4 months ago by xuzhou666
  1

• TypeError: expected np.ndarray (got tuple)

  #59 opened a year ago by InguChoi
  1

• DQN why train iterate for 10 times

  #57 opened 3 years ago by FeynmanDNA
  0

• MuZero minimal implementation

  #56 opened 3 years ago by ipsec
  0

• Add minimal IMPALA？

  #51 opened 3 years ago by meadewaking
  2

• The ratio in ppo.py should be detach() ?

  #33 opened 3 years ago by dedekinds
  5

• Remove unused import

  #43 opened 3 years ago by neal2018
  1

• Minimal way to save / replay trained model?

  #52 opened 3 years ago by HanClinto
  0

• Cartpole environment with Multidiscrete action space

  #47 opened 3 years ago by mgazzin
  3

• Add new algorithms

  #11 opened 5 years ago by rahulptel
  7

• Query about LSTM

  #50 opened 3 years ago by npitsillos
  0

• Add meta RL algorithms?

  #49 opened 3 years ago
  0

• Cartpole environment with Multidiscrete action space

  #48 opened 3 years ago by mg64ve
  0

• A naive question about updating parameters in DDPG.

  #46 opened 3 years ago by HiddenBeginner
  0

• cartpole ppo train , reward drop

  #42 opened 3 years ago by SeungyounShin
  1

• Maybe a bug in SAC Implementation?

  #40 opened 3 years ago by arthur-x
  1

• Use pytorch-lightning for better readability and optimization

  #39 opened 3 years ago by EmmanuelMess
  0

• Please add 1 continuous env

  #6 opened 4 years ago by bionicles
  2

• PPO Continuous Action Space

  #12 opened 4 years ago by raunakdoesdev
  2

• Soft Actor Critic?

  #38 opened 4 years ago by EmmanuelMess
  1

• Missing done mask?

  #32 opened 4 years ago by Junyoungpark
  3

• PPO update mistake?

  #36 opened 4 years ago by zcaicaros
  1

• Questions about A3C

  #29 opened 4 years ago by LoveRL
  1

• RuntimeError while running DDPG.py

  #34 opened 4 years ago by rl-max
  2

• TD3: Twin Delayed DDPG

  #37 opened 4 years ago by zcaicaros
  2

• torch.gather in relevant to policy gradient

  #31 opened 4 years ago by migom6
  0

• PPO has no entropy factor

  #30 opened 4 years ago by CesMak
  0

• Termination of a CartPole episode in REINFORCE.py

  #28 opened 4 years ago by ansari1375
  1

• Problem of `train_net()` in REINFORCE algorithm.

  #26 opened 4 years ago by fuyw
  5

• Add SAC?

  #19 opened 5 years ago by banma12956
  0

• TF2 implementation for Policy Gradient Reinforce

  #18 opened 5 years ago by dragen1860
  0

• LSTM + PPO value fitting

  #17 opened 5 years ago by hnshahao
  1

• Wrong gradient flow in bias correction term of ACER?

  #15 opened 5 years ago by wwiiiii
  1

• Improper asynchronous update in a3c

  #9 opened 5 years ago by rahulptel
  1

• Wrong td_target and test() call in a3c implementation

  #8 opened 5 years ago by rahulptel
  1

• Typo of actor_critic.py?

  #7 opened 5 years ago by seungwonpark
  1

• Use maxlen in deque initializer

  #3 opened 5 years ago by jwergieluk
  1

• train() overwrites the base method of nn.Module

  #2 opened 5 years ago by NikEyX
  1

• Reinforce implementation looks to use old data without importance sampling

  #1 opened 5 years ago by sritee
  1