astooke/rlpyt

Issues

• How to load replaybuffer in rlpyt/ul module？

  #216 opened 6 months ago by Gechenseu
  1

• Clarifications on how to use ul directory?

  #212 opened 3 years ago by snailrowen1337
  1

• How to get replays for training UL?

  #202 opened 4 years ago by kevinghst
  2

• Replay data - 15M_VecEps_B78

  #196 opened 4 years ago by indrasweb
  1

• need help : KeyError: 'info' KeyError: 'action'

  #215 opened 2 years ago by lalala215
  0

• How to use on Windows？

  #214 opened 2 years ago by maaalf
  0

• Why DQN puts predicted q value and target q value on cpu?

  #213 opened 3 years ago by yueyang130
  0

• EnvSteps vs CumSteps for ATC

  #199 opened 4 years ago by bmazoure
  1

• Code for Responsive Safety in RL by PID Lagrangian Methods

  #211 opened 3 years ago by manuel-delverme
  1

• [Bug report] Parallel sampler hanging on shutdown

  #210 opened 3 years ago by joel99
  0

• KeyError: 'action'

  #205 opened 3 years ago by csingh27
  4

• Clarification about prior_action in SAC

  #209 opened 3 years ago by 0xJchen
  0

• KeyError: 'action'

  #188 opened 4 years ago by hnekoeiq
  4

• Hanging Manager or worker

  #208 opened 3 years ago by deepdad
  0

• Where are the rewards and plots saved, if at all by default ?

  #206 opened 4 years ago by csingh27
  1

• ReturnAverage and NonzeroRewardsAverage in logs become nan after a period of time.

  #207 opened 4 years ago by bryanyuan1
  0

• Error on running GpuSampler/CpuSampler

  #179 opened 4 years ago by nazarblch
  4

• example_1 DQN won't learn Pong

  #204 opened 4 years ago by TheExGenesis
  0

• ctrl.barrier_in.wait() waiting issue

  #203 opened 4 years ago by sungreong
  0

• Safe RL using PID: langrangian ppo objective issue

  #201 opened 4 years ago by akjayant
  1

• r2d1 pong training time

  #200 opened 4 years ago by lvarga37
  0

• A question about CategoricalPgAgent

  #198 opened 4 years ago by LTEnjoy
  0

• Reproducing figure 4 results from Decoupling Representation Learning from RL paper for atari games

  #197 opened 4 years ago by harmanpreet93
  0

• Asynchronous runners with CPU only?

  #195 opened 4 years ago by definitelyuncertain
  2

• Why is .item() not called on grad norm like on other opt info fields?

  #163 opened 5 years ago by neighthan
  1

• Question on Decoupling Representation Learning from Reinforcement Learning: Why use MoCo instead of BYOL and Why use W in loss instead of architecture

  #190 opened 4 years ago by SohamTamba
  0

• [LSTM PPO] how to increase batch_B without creating multiple environment objects?

  #193 opened 4 years ago by andrewsonga
  0

• Error in cpu affinity assignment

  #192 opened 4 years ago by backpropper
  0

• Incorrect URL in the UL arXiv paper

  #191 opened 4 years ago by qbilius
  0

• Does `CpuWaitResetCollector` collect only one episode in each batch?

  #186 opened 4 years ago by yd-yin
  2

• Testing code examples

  #183 opened 4 years ago by yd-yin
  2

• Diagnostics/NewCompletedTrajs 0 on some iterations

  #185 opened 4 years ago by jordan-schneider
  1

• Passing shared changing parameters to parallel envs during training.

  #181 opened 4 years ago by yannidd
  4

• Environment with object's causing exception in 'build_array'

  #187 opened 4 years ago by vanavreddy
  0

• custom environment

  #184 opened 4 years ago by vanavreddy
  0

• Sequence Buffer Sampling Performance

  #180 opened 4 years ago by AlexanderKoch-Koch
  3

• Doesn't work with (non-atari) env

  #164 opened 5 years ago by drozzy
  22

• codecov tests failing?

  #177 opened 4 years ago by astooke
  2

• Breakout benchmarks

  #174 opened 5 years ago by RexGLiu
  2

• Continue training

  #165 opened 5 years ago by OlgaMaslova
  3

• "Name clash in Globals: info"

  #168 opened 5 years ago by jordan-schneider
  2

• PPO RNN rnn_state not propagated

  #166 opened 5 years ago by petrikvladimir
  2

• How can I run atari_dqn_async_gpu.py with n_gpu=1?

  #162 opened 5 years ago by wkentaro
  4

• Why DQN related agents/algos default to cpu for loss computation ?

  #171 opened 5 years ago by afansi
  1

• Handling Early Resets in Procgen Envs

  #169 opened 5 years ago by jakegrigsby
  1

• Optimizing the game simulation

  #167 opened 5 years ago by maxwellreuter
  2

• Buffer cannot be saved while using 1-D batch_T

  #161 opened 5 years ago by LecJackS
  3

• [question] OpenAI Gym Env?

  #170 opened 5 years ago by jarlva
  1

• NoneType is not callable with simple wrapped environment

  #159 opened 5 years ago by drozzy
  3

• colab install (conda package?)

  #160 opened 5 years ago by drozzy
  1