dgriff777/rl_a3c_pytorch

Issues

• Hyperparameters for training

  #45 opened a year ago by tiencapham
  0

• run a3c on 8 cpus, it still slow.

  #44 opened 2 years ago by TIANRENK
  1

• The links to the Gym environment evaluations is 404.

  #43 opened 2 years ago by Dear-Joe
  0

• Need a model, thank you

  #42 opened 3 years ago by abolaer
  1

• UserWarning: This overload of add_ is deprecated

  #41 opened 3 years ago by josh-tegus
  0

• question about trained models

  #39 opened 4 years ago by enochkan
  1

• Clarification needed regarding num_workers

  #38 opened 4 years ago by enochkan
  2

• Need for trained models

  #33 opened 5 years ago by pulkitso
  2

• How can I let the training automatically stop after a given number of episodes or after a given period of time?

  #37 opened 4 years ago by bokveizen
  1

• Is there any necessary to lock when update params?

  #36 opened 4 years ago by Ranglage
  1

• Question about Test function

  #35 opened 4 years ago by AlexTo
  1

• Cannot import test

  #34 opened 4 years ago by gitjeet
  1

• Reward Smoothing

  #32 opened 6 years ago by WangChen100
  2

• Stuck when training in MsPacman-v0

  #31 opened 6 years ago by KarlXing
  7

• eps for Adam

  #29 opened 6 years ago by jsuit
  2

• plot rewards as a function of number of timesteps

  #30 opened 6 years ago by rachit-dubey
  1

• Quick question on batch processing

  #28 opened 7 years ago by hohoCode
  4

• Pretrained models

  #27 opened 7 years ago by lweitkamp
  6

• Why ensure_shared_grads

  #25 opened 7 years ago by yiwan-rl
  1

• NotImplementedError

  #26 opened 7 years ago by rbunn80110
  1

• multi gpu support

  #24 opened 7 years ago by yhcao6
  6

• Normalize

  #23 opened 7 years ago by yhcao6
  10

• watching game visualization during training

  #22 opened 7 years ago by rbunn80110
  3

• Why one process run on 2 gpus?

  #21 opened 7 years ago by Jiankai-Sun
  4

• Question on training function

  #19 opened 7 years ago by yiwan-rl
  5

• Question on max-length

  #20 opened 7 years ago by yiwan-rl
  3

• Performance of Breakout

  #18 opened 7 years ago by yhcao6
  6

• synchronization among workers

  #17 opened 7 years ago by joe-redstone
  2

• Visualization not appearing

  #16 opened 7 years ago by riccitensor
  1

• How long have you trained the model

  #15 opened 7 years ago by 1601214542
  3

• Train a new game

  #14 opened 7 years ago by jason718
  1

• Move trained_models into release

  #13 opened 7 years ago by inoryy
  1

• Why would you end training episode if any lives are lost?

  #1 opened 8 years ago by ethancaballero
  7

• Seaquest-v0 not training as well as announced

  #11 opened 7 years ago by ThomasLecat
  5

• Reward is always 0 when training Breakout-v0

  #10 opened 7 years ago by NeymarL
  8

• cnn layers

  #9 opened 7 years ago by 404akhan
  4

• what is player.flag and player.starter

  #7 opened 8 years ago by 404akhan
  1

• Not an issue just a question

  #8 opened 8 years ago by sahilzubair
  1

• I just want to say your trained model has no effect

  #5 opened 8 years ago by lucasjinreal
  20

• Solving time

  #4 opened 8 years ago by hugemicrobe
  6

• Can you please list out the difference between your code and ikostrikov/pytorch-a3c

  #2 opened 8 years ago by slowbull
  14

• Is the max pool and activation in the right order?

  #3 opened 8 years ago by ethancaballero
  1