forked from PufferAI/PufferLib
-
Notifications
You must be signed in to change notification settings - Fork 22
Pull requests: Emerge-Lab/PufferDrive
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
WIP: fix stopped-agent metric dilution (active_step_count divisor)
#402
opened Apr 13, 2026 by
eugenevinitsky
Loading…
4 tasks
Align reward randomization bounds with GIGAFLOW paper
#401
opened Apr 12, 2026 by
eugenevinitsky
Loading…
4 tasks
[WIP] Trajviz: Vulkan offline trajectory renderer
#398
opened Apr 11, 2026 by
eugenevinitsky
Loading…
4 of 8 tasks
[WIP]Added collision types, at-fault collision percentage
#394
opened Apr 9, 2026 by
mpragnay
Loading…
Refactor policy network code and remove LSTM as default
#393
opened Apr 9, 2026 by
riccardosavorgnan
Collaborator
•
Draft
WIP: merge turbostream DO NOT MERGE
#384
opened Mar 31, 2026 by
eugenevinitsky
Loading…
3 of 5 tasks
Randomize agent positions on respawn in variable agent mode
#376
opened Mar 29, 2026 by
eugenevinitsky
Loading…
WIP DO NOT MERGE Fix: save_map_binary wrote scenario_id prefix that C never reads
#374
opened Mar 28, 2026 by
eugenevinitsky
Loading…
3 tasks
Fix classic trajectory replay and expert inference
#369
opened Mar 27, 2026 by
julianh65
Collaborator
Loading…
[WIP DO NOT MERGE OR REVIEW] Add trajectory rollout and rendering infrastructure
#364
opened Mar 24, 2026 by
eugenevinitsky
Loading…
3 tasks done
STOP behavior reward and learning masking.
#353
opened Mar 20, 2026 by
riccardosavorgnan
Collaborator
Loading…
Enabling injection of different parameters at environment reset
#321
opened Mar 2, 2026 by
riccardosavorgnan
Collaborator
•
Draft
Previous Next
ProTip!
Updated in the last three days: updated:>2026-04-10.