facebookresearch / Pearl

A Production-ready Reinforcement Learning AI Agent Library brought by the Applied Reinforcement Learning team at Meta.

facebookresearch/Pearl Issues

cuda out of meomory
Closed 9 days ago13
Bug in tutorial single item recommender system: can't instantiate abstract class RecEnv
Closed 16 days ago3
Error when running the Single Item Recommender System Notebook
Closed 22 days ago5
Simplify actor-critic with shared layers
Updated a month ago2
Creating a project roadmap
Updated a month ago1
How is Pearl different from TorchRL?
Closed 2 months ago1
TorchRL Vs Pearl: What's the difference & when to use one?
Closed 2 months ago1
PPO on continuous actions
Updated 2 months ago1
Pip availability and clear versioning in branches/tags/releases
Updated 2 months ago1
custom network issue
Closed 2 months ago1
save the trained agent for reuse
Updated 2 months ago1
Functionality for Stein variational policy gradient and/or regularization through the use of prior policies
Updated 2 months ago2
Is there some limitation with the dimensions of actions and observations?
Updated 2 months ago5
Question: train_via_uniform_data
Closed 2 months ago10
More tutorials are needed
Updated 3 months ago5
Unable to export loss values from TD3
Closed 3 months ago12
Model Export
Updated 3 months ago5
Frozen lake
Closed 3 months ago5
Frozen-lake tutorial
Closed 3 months ago1
What is the policy of Pull Requests ?
Closed 3 months ago3
Issues when trying to use torch.jit.script()
Closed 3 months ago2
Use pearl with custom pytorch model
Closed 3 months ago6
Does Pearl support serving in TorchScript?
Closed 4 months ago1
NotImplementedError: The Gym space 'Tuple' is not yet supported in Pearl.
Updated 4 months ago1
LSTM History Summarization Poor Performance
Closed 4 months ago8
PPO + LSTM issues
Closed 4 months ago3
Examples
Closed 5 months ago12
STAC algorithm
Updated 4 months ago1
Will n-D observation states be supported?
Updated 4 months ago1
Custom space vectorizers
Updated 4 months ago1
Mixed Discrete + Continuous Action Spaces
Closed 4 months ago6
MultiDiscrete Action space not support
Closed 4 months ago1
online_learning function meet custom gym env without register
Closed 4 months ago1
FIFO On-Policy replay buffers lead to error with PPO
Closed 4 months ago2
REINFORCE actor update missing discount factor
Closed 4 months ago2
Hidden dim of LSTM history summarization module must be equal to observation dim
Closed 4 months ago2
Recommender System Example
Closed 4 months ago1
How to infer result of the tutorials?
Closed 4 months ago1
Demo DQN-LSTM dose not converge
Closed 4 months ago3
Multi-agent Reinforcement Learning?
Closed 4 months ago1
Typo in the docs
Closed 5 months ago1
Graph Neural Network Compatibility
Closed 5 months ago1
FAlcon
Closed 5 months ago1
Gymnasium?
Closed 5 months ago1
PolicyLearner batch_size versus episode_steps clarification
Closed 5 months ago4
Is this Q*?
Closed 5 months ago5
Q-transformer
Closed 5 months ago1
[Question] Offline evals
Closed 5 months ago1
Suggestion: Add linters, formatters, and type checkers to ensure code quality and coding style consistency
Closed 5 months ago2
Link to project site in README broken?
Closed 5 months ago2