DLR-RM/stable-baselines3

Issues

• [Question] Error Installing stable-baselines3[extra] on Windows

  #2033 opened 18 days ago by bajramienes
  2

• How can i change a Distribution?

  #2032 opened 19 days ago by CAI23sbP
  1

• [Bug] importing stable baselines 3 on linux and windows directory issue

  #2034 opened 18 days ago by Feelfeel20088
  1

• [Feature Request] When are you planning to upgrade to Gymnasium v1.0.0

  #2023 opened 19 days ago by drulye
  2

• [Bug]: No metrics logged when using wandb integrations

  #1995 opened 2 months ago by XiaobenLi00
  4

• [Question] Can a model be used in environments with different observation_space sizes?

  #2031 opened 22 days ago by SummerDiver
  2

• [Question] Can I load a RL model with Mac trained on windows platform?

  #2029 opened 25 days ago by Ssstirm
  5

• [Question] Batch Size Selection for a Finite MDP

  #2024 opened a month ago by DavidLudl
  4

• [Question] TD3 algorithm， During training，why limit the next_actions

  #2022 opened a month ago by Danny551
  1

• [Question] How to customize the loss calculation for PPO

  #2028 opened a month ago by olmoulin
  3

• [Bug]: When using SubprocVecEnv for parallel training of agents, the ep_rew_mean is no longer being recorded.

  #2027 opened a month ago by didu11
  0

• [Bug]: How to avoid saving an external LLM model while saving a cutomized dqn policy

  #2025 opened a month ago by chrisgao99
  2

• [Question] WSL Freeze with Rocm

  #2021 opened a month ago by darkavatar23
  2

• [Feature Request] request title jax API and gymnax

  #2020 opened a month ago by quanmissq
  1

• [Question] How to change my loaded model's "exploration_initial_eps", "exploration_final_eps", "exploration_fraction" parameters?

  #2018 opened a month ago by Melih96
  2

• [Question] Can stable-baselines3 be installed through pip without cuda dependencies? Is the CPU only docker image the only alternative?

  #2019 opened a month ago by joemc94
  1

• [Question] Auto-regressive manner policy network?

  #2010 opened 2 months ago by wadmes
  1

• Is possible to filter experience from the episode whose length is longer than a specified value to add into replay_buffer?

  #1999 opened 2 months ago by CornfileChase
  1

• [Question] About the logger

  #1997 opened 2 months ago by XiaobenLi00
  1

• Logger information

  #1998 opened 2 months ago by XiaobenLi00
  1

• [Feature Request] Warn users when using GPU with `A2C`/`PPO` + update documentation

  #2012 opened 2 months ago by jws-1
  5

• can not load PPO model when I use custom net_arch

  #2015 opened 2 months ago by krishdotn1
  1

• [Question] Manually Controlling Actions During PPO Training

  #2014 opened 2 months ago by wayne-weiwei
  2

• [Bug]: obs_as_tensor RuntimeError with numpy==2.1.1 on MacOS13

  #2016 opened 2 months ago by nquetschlich
  2

• [bug] Adaptive SAC: using logarithm of entropy coefficient to compute temperature objective instead of entropy coefficient

  #2013 opened 2 months ago by Mattia-sony
  1

• [Question] The entropy value is a negative number, and the entropy loss is a positive number.

  #2004 opened 2 months ago by YSAA1
  3

• [Bug]: Episode start flag is never set for off policy algorithms

  #2011 opened 2 months ago by josndan
  1

• A succesfully ppo trained agent demands some steps of re-training to make good predictions

  #2009 opened 2 months ago by tanielsfranklin
  1

• [Feature Request] Safe Reinforcement Learning & Multi-Objective Reinforcement Learning

  #2008 opened 2 months ago by cherrywoods
  2

• [Question] Shared feature extractor and gradient

  #2006 opened 2 months ago by brn-dev
  2

• [Question] Issues withe monitor not having seed parameter

  #2002 opened 2 months ago by AbhayGoyal
  2

• [Bug]: bug title SubprocVecEnv TypeError: reset() got an unexpected argument 'seed'

  #2001 opened 2 months ago by ccleavinger
  1

• [Question] MARL using Stable Baselines3

  #2007 opened 2 months ago by Hamza-101
  1

• [Feature Request] add_scalars to wirte func in TensorBoardOutputFormat in logger

  #1994 opened 3 months ago by shimonShouei
  2

• [Feature Request] Add support for optional environment wrapping in base_class.py

  #1996 opened 3 months ago by hasan-yaman
  2

• Clarification on Dependency Between Elements in Action Generation

  #1989 opened 3 months ago by fardinabbasi
  4

• [Question] Passing arguments to an environment that can't be pickled.

  #1990 opened 3 months ago by abhineet-gupta
  3

• observation_space does not match reset() observation and The environment is being initialised with render_mode='human' that is not in the possible render_modes ([])

  #1992 opened 3 months ago by XiaobenLi00
  5

• That WORKED !!

  #1991 opened 3 months ago by XiaobenLi00
  0

• [Вопрос] how to train 2 models in parallel?

  #1982 opened 3 months ago by kozlolet
  1

• [Question] About the output layer of algorithms

  #1988 opened 3 months ago by abdulkadrtr
  1

• Custom actor and critic network

  #1985 opened 3 months ago by krishdotn1
  4

• [Question] New to this and can't get it installed plz help

  #1987 opened 3 months ago by Misticfury
  1

• [Bug]: Possible inconsistencies with the PPO implementation

  #1986 opened 4 months ago by hexonfox
  2

• [Feature Request] Temporal Convolutional network

  #1984 opened 4 months ago by tty666
  1

• [Feature Request] Additional Callback before stepping the optimizer

  #1980 opened 4 months ago by kcorder
  2

• [Bug]: 'CarRacing' object has no attribute 'num_envs'

  #1983 opened 4 months ago by kuds
  1

• I would like to know what is the network structure of the SAC algorithm

  #1976 opened 4 months ago by hjg857
  1

• [Question] Questions about CNN policy input channel

  #1977 opened 4 months ago by DavidLudl
  2

• [Bug]: Load a PPO model and re-start learning

  #1974 opened 4 months ago by nrigol
  1