开放
Use Altair inside a notebook to display benchmark results
documentationgood first issue
3 条评论0 个反应0 负责人
议题
开放
Don't pass envpool envs where vectorenvs are needed
buggood first issuerefactoring
0 条评论0 个反应0 负责人
开放
Reduce duplication between examples/atari/atari_network and examples/vizdoom/network
good first issuerefactoring
1 条评论0 个反应0 负责人
开放
Support Dict observation spaces
documentationenhancementgood first issuetentative
8 条评论0 个反应0 负责人
开放
Code duplication between ReplayBuffer and ReplayBufferManager
good first issuerefactoring
0 条评论0 个反应0 负责人
开放
Improve and extend Documentation Content
documentationgood first issue
0 条评论1 个反应0 负责人
开放
Include parts of atari/mujoco helpers in package code
good first issueminorrefactoring
1 条评论0 个反应0 负责人
开放
RNN related issues
RNNbughelp wanted
12 条评论0 个反应0 负责人
开放
Extend and fix reward/return normalizations
enhancementgood first issue
0 条评论0 个反应0 负责人
开放
Need a flexible method to record training data in TensorBoard
enhancementgood first issue
5 条评论0 个反应1 负责人
开放
Allow finding the corresponding episode from a sample in reply buffer
RNNenhancementgood first issue
1 条评论1 个反应0 负责人
开放
How to support multi-agent reinforcement learning
MARLdiscussiongood first issue
34 条评论1 个反应1 负责人