Skip to content

Pull requests: vwxyzjn/cleanrl

New pull request New
39 Open 300 Closed
39 Open 300 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

docs: correct TensorBoard working directory
#570 opened Aug 22, 2026 by quinnarnold Loading…
5 of 11 tasks
1
1
Fix KeyError on Gymnasium final_obs for off-policy scripts
#567 opened Jul 16, 2026 by Fstarnb Loading…
3 of 8 tasks
1
2
Fix enjoy.py crash: re-add missing --capture_video argument
#561 opened Jun 24, 2026 by YangNuoCheng Loading…
3 of 6 tasks
1
1
Add unit tests for ppo_continuous_action GAE computation
#560 opened Jun 24, 2026 by YangNuoCheng Loading…
4 of 7 tasks
1
Add unit tests for PPO GAE computation
#559 opened Jun 24, 2026 by YangNuoCheng Loading…
4 of 7 tasks
1
ci: add dedicated ruff lint workflow
#558 opened Jun 14, 2026 by dashitongzhi Loading…
1 of 3 tasks
6
Fix Rainbow segment tree sampling for non-power-of-two capacities
#557 opened Jun 13, 2026 by discobot Loading…
4 of 18 tasks
1
1
Fix: reuse training obs normalization stats in ppo_continuous_action eval
#556 opened Jun 13, 2026 by discobot Loading…
4 of 18 tasks
1
1
perf: avoid tensor memory copy in ppo_atari_envpool, pqn_atari_envpool, ppo_rnd_envpool, and ppo_procgen
#552 opened May 4, 2026 by srygaard Loading…
5 of 19 tasks
1
fix(enjoy): restore --capture-video argument removed in #424
#550 opened Apr 20, 2026 by mturan33 Contributor Loading…
4 tasks done
1
1
Security: Command injection via unsanitized Docker tag in shell command
#548 opened Apr 15, 2026 by tuanaiseo Loading…
3 tasks
1
fix: update JAX dependency stack to modern compatible versions
#544 opened Mar 19, 2026 by Okyumi Loading…
1
1
Fix #508: Use per-env logging_step for episodic return logging
#539 opened Mar 2, 2026 by gspeter-max Loading…
1
3
Fixes Issue #533 - Fix tuner dropping the first CLI argument.
#536 opened Jan 12, 2026 by hliu-ai Loading…
3 of 18 tasks
1
2
Refactor: Enforce explicit reshape and consistent 2D output in swap_and_flatten
#535 opened Jan 2, 2026 by ZHYsfl Loading…
3 of 8 tasks
1
Add ES-C51 (Expected Sarsa based C51)
#529 opened Oct 9, 2025 by Rijul-Tandon Loading…
5 of 18 tasks
19
[Enhancement] Log mean/max KL across PPO update phase
#523 opened Aug 27, 2025 by pjain03 Loading…
8 of 18 tasks
1
Add SAC Jax Implementation (reloaded)
#519 opened Jul 17, 2025 by araffin • Draft
1 of 18 tasks
1
Update to Gymnasium v1.0
#516 opened Jul 5, 2025 by pseudo-rnd-thoughts Collaborator Loading…
4 of 18 tasks
2
feat: port isaacgym example to isaaclab
#513 opened Jun 17, 2025 by lvjonok Loading…
3 of 18 tasks
1
[Non-performance-impacting update] Use Pytorch DDP in ppo_atari_multigpu
#495 opened Jan 14, 2025 by realAsma Loading…
4 of 18 tasks
4
Adding Munchausen Reinforcement Learning
#466 opened Jun 30, 2024 by Paul-antoineLeTolguenec Loading…
6 of 18 tasks
1
Change actor_update_interval to policy_frequency in SAC comment
#458 opened Apr 22, 2024 by JinayJain Loading…
1 of 18 tasks
1
Enhancing Termination and Truncation Handling in CleanRL's PPO Algorithm
#448 opened Feb 29, 2024 by Josh00-Lu Loading…
6 of 20 tasks
10
add accelerate example
#446 opened Feb 10, 2024 by edbeeching • Draft
1 of 18 tasks
17
ProTip! What’s not been updated in a month: updated:<2026-08-01.