Pinned issues
Issues
- 5
Poetry update the torch versioned from cuda (2.0.1+cu118) to cpu (2.1.1) defaultly on Windows
#1145 opened by coolermzb3 - 1
- 1
ImportError: cannot import name 'Self' from 'typing' (/root/miniconda3/lib/python3.10/typing.py)
#1148 opened by luweiagi - 1
[question] Why does Tianshou use a replay buffer in on-policy RL algorithms?
#1147 opened by maguro27 - 0
Document effects of the relations between buffer size, num workers and episode length
#1143 opened by MischaPanch - 6
How can I make action sampling within the range specified by my environment when using onpolicy_trainer?
#1142 opened by lidaken - 1
Revisit `Launcher` for starting multiple experiments
#1121 opened by MischaPanch - 0
Extend benchmark with mujoco v4 envs
#1140 opened by MischaPanch - 1
Does Tianshou truly supports MARL out of the box?
#1137 opened by Legendorik - 2
Change log is chaotic and partly uninformative
#1129 opened by opcode81 - 5
Some issues regarding configuration parameters
#1119 opened by yshichseu - 4
Potential confusion about where start timesteps are collected in HL interfaces
#1135 opened by MischaPanch - 0
- 1
how to run RL using multi-nodes in cluster
#1133 opened by HYB777 - 1
- 24
Batch: remove `is_empty`
#1108 opened by MischaPanch - 7
Buffer: fix discrepancy in slicing order
#1090 opened by MischaPanch - 0
Glad you agree with me on this ^^. I'm not sure whether anywhere in the code the retrieval of the slice with empty values is used. For me it's fine to completely remove it, however, many tests will need to be adjusted, as now many of them rely on this somehow weird retrieval mechanism.
#1120 opened by MischaPanch - 5
Chinese document pages return 404
#1078 opened by H-xie - 3
- 2
- 0
Provide a devcontainer, base GH actions off it
#1118 opened by MischaPanch - 0
Add the non-in-place counterpart of `Batch.to_torch`
#1116 opened by dantp-ai - 9
Batch: don't create new objects on getitem
#1086 opened by MischaPanch - 5
- 8
- 1
UnboundLocalError: cannot access local variable 'obs_space_dtype' in atari_wrapper.py
#1111 opened by zhuyuanyang - 1
Use Atari-5 for future benchmarking of discrete RL
#1110 opened by nuance1979 - 2
Should we use torch.compile?
#1114 opened by MischaPanch - 1
Should we use the new schedule-free optimizer?
#1115 opened by MischaPanch - 0
Revisit "warm-up" phase in examples
#1112 opened by MischaPanch - 3
- 1
Batch: deprecate setattr
#1085 opened by MischaPanch - 3
Batch: only allow entries with the same length
#1087 opened by MischaPanch - 5
Missing Link
#1099 opened by DarkTechPirate - 0
Don't pass envpool envs where vectorenvs are needed
#1096 opened by MischaPanch - 1
Reduce duplication between examples/atari/atari_network and examples/vizdoom/network
#1092 opened by MischaPanch - 7
Support Dict observation spaces
#1065 opened by MischaPanch - 0
Re-examine the whole state story for RNNs
#1095 opened by MischaPanch - 0
- 0
Fix docstring in BranchingNet
#1093 opened by MischaPanch - 0
- 1
- 4
data recording and saving method
#1079 opened by Xiong5Heng - 0
Typing annotations of step from MyTestEnv is incompatible with its current subclass gym.Env because it can generate non-scalar rewards.
#1080 opened by dantp-ai - 5
how to convert Batch into ndarray/tensor
#1064 opened by qmpzzpmq - 0
Revisit and maybe optimize Collectors
#1069 opened by MischaPanch - 0
Question: Is Recurrent net supported for FQF
#1075 opened by edoust - 2
Inquiry version 0.5.1 and version recommendation
#1073 opened by H-xie - 3
two dimensional input action in DDPG
#1070 opened by chenyi8920