issues Search Results · repo:pytorch/rl language:Python
Filter by
676 results
(63 ms)676 results
inpytorch/rl (press backspace or delete to remove)Describe the bug
I see very cool advancements in the direction of LLM RL training in the repo, awesome work! :)
After playing a bit with the LLMEnv I got the following error when passing dialogue data ...
bug
albertbou92
- 1
- Opened 4 days ago
- #2875
Motivation
We need a collector that fits well the LLM space.
We will need to simplify the rollout function greatly - I would rewrite it from scratch. The LLMEnv and vLLMWrapper can
be used to simulate ...
enhancement
vmoens
- Opened 4 days ago
- #2872
Describe the bug
In the section Storing trajectories, an error is reported because a variable is undefined.
To Reproduce
from torchrl.data import SliceSampler
rb = TensorDictReplayBuffer(
storage=LazyTensorStorage(size), ...
bug
BrikeX
- Opened 10 days ago
- #2860
Describe the bug
When running experiments with multiprocess-based sampling of trajectories on macOS, the initialization of the data
collectors fail
To Reproduce
from torchrl.envs.libs.gym import GymEnv ...
bug
LCarmi
- Opened 11 days ago
- #2858
Describe the bug
When calling step_mdp on a TensorDict that contains NonTensorData inside the next key, the NonTensorData field is not
copied over to the new TensorDict.
To Reproduce
import dataclasses ...
bug
belfaunder
- 2
- Opened 19 days ago
- #2842
Describe the bug
When using BraxEnv with requires_grad=True, there appears to be a memory leak on the CPU side. The memory usage keeps
increasing over time, which can be observed using tools like htop. ...
bug
mondeg0
- 2
- Opened 23 days ago
- #2837
Describe the bug
Creating an instance of VideoRecorder crashes when we pass fps.
To Reproduce
from torchrl.record import VideoRecorder
VideoRecorder(None, None, , None, fps=1)
Traceback (most recent ...
bug
AlexandreBrown
- Opened 25 days ago
- #2826
I tried to replace/update some parts of the data stored in ReplayBuffer with new tensors.
Although the functionality was add (https://github.com/pytorch/rl/issues/2209), I found an unexpected behavior ...
bug
tmparticle
- 3
- Opened on Feb 25
- #2810
Following up on a discussion post.
TLDR; Trying to create an example notebook for UnityMLAgentsEnv/Wrapper but am unable to find docs or reference to how
to interact with the keys that are produced for ...
kylelevy
- 9
- Opened on Feb 12
- #2781
In torchrl/objectives/value/utils.py#L287-L290, the code currently uses tensor.shape[-2] to choose between torch.int16
and torch.int32. This should use the time dimension (time_dim) instead, especially ...
KubaMichalczyk
- 2
- Opened on Feb 5
- #2758
Learn how you can use GitHub Issues to plan and track your work.
Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub IssuesProTip!
Restrict your search to the title by using the in:title qualifier.Learn how you can use GitHub Issues to plan and track your work.
Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub IssuesProTip!
Restrict your search to the title by using the in:title qualifier.