-
Notifications
You must be signed in to change notification settings - Fork 310
Issues: pytorch/rl
[QUESTION] Using LSTMs with vectorized envs without ParallelEnv
#1493
opened Sep 5, 2023 by
ADebor
Open
7
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
[BUG] Something isn't working
Transform._call
is not called from reset
bug
#2595
opened Nov 22, 2024 by
codingWhale13
[BUG] Something isn't working
torchrl.objectives.SACLoss
internal objective does not mask out terminated states
bug
#2590
opened Nov 20, 2024 by
fmeirinhos
3 tasks done
[BUG] Something isn't working
torchrl.objectives.SACLoss
is broken when there is more than one qvalue_network
bug
#2589
opened Nov 20, 2024 by
fmeirinhos
3 tasks done
[BUG] SliceSampler should return unique IDs when sampling multiple times from the same trajectory
bug
Something isn't working
#2588
opened Nov 20, 2024 by
dmelcer9
3 tasks done
[Feature Request] Transform that stacks data for agents with identical specs
enhancement
New feature or request
#2566
opened Nov 14, 2024 by
kurtamohler
1 task done
[Feature Request] Use context managers to toggle the recurrent mode of RNN modules?
enhancement
New feature or request
#2562
opened Nov 14, 2024 by
thomasbbrunner
1 task done
[BUG] Vmap call in value estimate not compatible with torch_geometric radius graph
bug
Something isn't working
#2537
opened Nov 5, 2024 by
matteobettini
[Feature Request] Include group-level done fields in the VMAS environment
enhancement
New feature or request
#2536
opened Nov 5, 2024 by
thomasbbrunner
1 task done
[Feature Request] TorchRL for Python 3.13t
enhancement
New feature or request
#2535
opened Nov 4, 2024 by
patchmeifyoucan
[BUG] SyncDataCollector Crashes when init_random_frames=0
bug
Something isn't working
#2534
opened Nov 4, 2024 by
AlexandreBrown
3 tasks done
[BUG] RuntimeError: Setting 'advantage' via the constructor is deprecated, use .set_keys(<key>='some_key') instead .
bug
Something isn't working
#2528
opened Oct 30, 2024 by
jianzhnie
[BUG] Something isn't working
target_value_network_params
initialization bug in convert_to_functional()
bug
#2523
opened Oct 28, 2024 by
matinmoezzi
3 tasks done
[BUG] info['_weight'] device for Importance Sampling in PER
bug
Something isn't working
#2518
opened Oct 26, 2024 by
EladSharony
3 tasks done
[Feature Request] Loss New feature or request
make_value_estimator
takes a ValueEstimatorBase
class as argument
enhancement
#2503
opened Oct 18, 2024 by
louisfaury
1 task done
[Feature Request] Support compiling New feature or request
ReplayBuffer.extend/sample
without recompile
enhancement
#2501
opened Oct 18, 2024 by
kurtamohler
1 task done
[BUG]ClipPPOLoss encounters an bug when calculating the loss in a composite action space scenario, failing to produce the correct result.
bug
Something isn't working
#2487
opened Oct 10, 2024 by
Sui-Xing
[BUG] Problems with multidimensional Something isn't working
nvec
in MultiCategorical
bug
#2486
opened Oct 10, 2024 by
matteobettini
[BUG] Something isn't working
MaskedCategorical
missing mode
and deterministic_sample
properties.
bug
#2479
opened Oct 9, 2024 by
MorganeAyle
[BUG] GAE parameters (gamma, lmbda) seemed to get changed by ClipPPOLoss, advantage module does not calculate loss_critic
bug
Something isn't working
#2462
opened Oct 3, 2024 by
therealjoker4u
[BUG]UserWarning: deterministic_sample wasn't found when queried in <class 'torch.distributions.categorical.Categorical'>. CompositeDistribution is falling back on mode instead. For better code quality and efficiency, make sure to either provide a distribution with a deterministic_sample attribute or to change the InteractionMode to the desired value.
bug
Something isn't working
#2458
opened Sep 27, 2024 by
Sui-Xing
[BUG] Something isn't working
RenameTransform
of ParallelEnv
is not the same as ParallelEnv
of transformed environment
bug
#2445
opened Sep 19, 2024 by
thomasbbrunner
3 tasks done
GAE does not support LSTM-based value network.
enhancement
New feature or request
#2444
opened Sep 19, 2024 by
levelrin
1 task done
Document the TensorDict structure of the return of the _step() function for a multi agent environment
enhancement
New feature or request
#2425
opened Sep 8, 2024 by
zoetsekas
1 task done
Previous Next
ProTip!
Follow long discussions with comments:>50.