stable-baselines3 · Issues· 89 open
Open on GitHubLocally synced open issues (discussions stay on GitHub)
- #2285
[Bug] gSDE+tanh reconstruction changes PPO replay gradients despite a unit initial ratio
LLM generatedUpdated Sep 17, 2026 - #1790
[Bug]: Reset options ignored when resetting due to termination / truncation from within wrapper's `step`
bugdocumentationhelp wantedUpdated Sep 15, 2026 - #2282
[Bug]: Monitor with override_existing=False writes no header when the file is new
Updated Sep 4, 2026 - #2185
[Bug]: ReplayBuffer cannot detect the real memory constraints in Kubernetes.
documentationcheck the checklistUpdated Aug 28, 2026 - #2279
[Question] Why does VecNormalize normalize rewards by returns instead of rewards?
duplicatequestionUpdated Aug 25, 2026 - #2268
[Bug]: VecEnv sub-environment seeds (seed + i) overlap across runs with adjacent base seeds
bugUpdated Jul 24, 2026 - #2090
[Bug]: `is_image_space` works poorly with Gymnasium's `FrameStackObservation`
bugdocumentationhelp wantedUpdated May 27, 2026 - #1692
[Bug]: Rendering with EvalCallback does not render the initial or final state
bugUpdated May 26, 2026 - #627
[Bug] HER is not updating the done flag of HER transitions
bugUpdated May 25, 2026 - #2255
[Bug]: Render Tests Failure (SDL_RumbleMotor Deps Duplicate)
bugUpdated May 23, 2026 - #2142
BetaDistribution policy for bounded continuous action spaces to avoid Gaussian clipping bias and improve training stability
enhancementUpdated May 19, 2026 - #2256
[Question] why self.env.reset doesn't return an obs variable in FireResetEnv wrapper reset function, while it does in EpisodicLifeEnv and NoopResetEnv wrapper reset function
questionUpdated May 15, 2026 - #2145
[question] current_rewards in evaluations.py
questionUpdated May 2, 2026 - #2245
[Bug]: Left navigation moves when right content scrolls.
documentationUpdated Apr 21, 2026 - #2238
[Question] question title
questionUpdated Apr 8, 2026