-
Notifications
You must be signed in to change notification settings - Fork 1.1k
Issues: thu-ml/tianshou
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Author
Label
Projects
Milestones
Assignee
Sort
Issues list
Need a flexible method to record training data in TensorBoard
enhancement
Feature that is not a new algorithm or an algorithm enhancement
good first issue
Good for newcomers
doubt in compute_nstep_return in policy
bug
Something isn't working
not reproduced yet
Not yet tested or reproduced by a reviewer
question
Further information is requested
compute_episodic_return bug when v_s=None
performance issues
RNN+PPO bug in test/continuous/test_ppo.py
bug
Something isn't working
not reproduced yet
Not yet tested or reproduced by a reviewer
RNN
Temporary label to group all things RNN
#884
opened Jun 5, 2023 by
Caopeng17
[Questions] against PPO process_fn implementation: why not re-using forward's log_prob but re-compute instead?
question
Further information is requested
#883
opened Jun 4, 2023 by
spacegoing
2 of 8 tasks
Question with stack_num
bug
Something isn't working
not reproduced yet
Not yet tested or reproduced by a reviewer
#882
opened Jun 3, 2023 by
sNiper-Qian
Allow finding the corresponding episode from a sample in reply buffer
enhancement
Feature that is not a new algorithm or an algorithm enhancement
good first issue
Good for newcomers
RNN
Temporary label to group all things RNN
puzzle about policy learning of offline RL algorithms
question
Further information is requested
#877
opened May 27, 2023 by
GongYanfu
3 of 8 tasks
SAC implementation consider the reduction operator as a parameter as "min" in not always the best choice
algorithm enhancement
Not quite a new algorithm, but an enhancement to algo. functionality
blocked
Can't be worked on for now
not reproduced yet
Not yet tested or reproduced by a reviewer
#863
opened Apr 29, 2023 by
jamartinh
About the HER implementation in PickAndPlace-v2 environment
not reproduced yet
Not yet tested or reproduced by a reviewer
performance issues
Slow execution or poor-quality results
#862
opened Apr 29, 2023 by
shangdongyang
8 tasks done
how to return multiple rnn internal state value?
question
Further information is requested
RNN
Temporary label to group all things RNN
#855
opened Apr 20, 2023 by
db005
[Question] Best practice to save and resume training with PPO + reward normalization
question
Further information is requested
#846
opened Apr 5, 2023 by
sky0470
4 of 8 tasks
Plotting more metrics for PPO
blocked
Can't be worked on for now
enhancement
Feature that is not a new algorithm or an algorithm enhancement
Getting started example causes TypeError: object of type 'TimeLimit' has no len()
bug
Something isn't working
not reproduced yet
Not yet tested or reproduced by a reviewer
Errors with ParallelEnv and AECEnv
question
Further information is requested
#816
opened Feb 28, 2023 by
Franjrz
5 of 8 tasks
[question] LSTM for A2C with discrete action space
question
Further information is requested
RNN
Temporary label to group all things RNN
#814
opened Feb 27, 2023 by
cbschen
3 of 6 tasks
Fix handling of torch "device" association
bug
Something isn't working
good first issue
Good for newcomers
Possible Leak of Observations In Multi-Agent Policies
MARL
Temporary label to group all things MARL
question
Further information is requested
#806
opened Feb 15, 2023 by
uinversion
4 of 8 tasks
RNN support for TD3 and SAC
question
Further information is requested
RNN
Temporary label to group all things RNN
#795
opened Jan 12, 2023 by
qtomcatq
lstm+ppo/sac
question
Further information is requested
RNN
Temporary label to group all things RNN
#754
opened Oct 7, 2022 by
1900360
Implement Decision Transformer for offline RL
blocked
Can't be worked on for now
new algorithm
Adding a new RL algorithm
RNN
Temporary label to group all things RNN
#626
opened May 2, 2022 by
nuance1979
4 of 8 tasks
Improve discrete control offline RL benchmark
enhancement
Feature that is not a new algorithm or an algorithm enhancement
#612
opened Apr 25, 2022 by
nuance1979
4 of 8 tasks
question about DRQN
bug
Something isn't working
not reproduced yet
Not yet tested or reproduced by a reviewer
RNN
Temporary label to group all things RNN
#584
opened Apr 3, 2022 by
leao1995
Implementation design issues in SubprocVectorEnv
discussion
Discussion of a typical issue
enhancement
Feature that is not a new algorithm or an algorithm enhancement
refactoring
No change to functionality
#573
opened Mar 19, 2022 by
duburcqa
What paper or reference is the RNN implementation trying to replicate?
bug
Something isn't working
RNN
Temporary label to group all things RNN
#567
opened Mar 11, 2022 by
BFAnas
5 of 8 tasks
ProTip!
Type g i on any issue or pull request to go back to the issue listing page.