-
Notifications
You must be signed in to change notification settings - Fork 4.3k
Pull requests: Unity-Technologies/ml-agents
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
A new training mode called "both" has been added. This allows simultaneous training with both PPO and SAC, each on a separate instance. For example, if you have 4 environments, the even-numbered ones will use PPO, and the odd-numbered ones will use SAC. Therefore, it's necessary to specify even values for num_envs.
#6233
opened Aug 3, 2025 by
VictorBarbosa
Loading…
1 of 10 tasks
[DON'T MERGE] Normalize GAIL encoder inputs
#4670
opened Nov 19, 2020 by
ervteng
Loading…
1 of 10 tasks
Reorder angles in RayPerceptionSensor
#5746
opened May 15, 2022 by
RedTachyon
Loading…
3 of 10 tasks
Release 22/inconsistent device tensor action in trainers [Don't Merge]
#6225
opened Jul 20, 2025 by
Jkho80
Loading…
1 of 10 tasks
Refactor ELO rating update logic in GhostTrainer
#6138
opened Aug 28, 2024 by
OMS1996
Loading…
1 of 10 tasks
Pettingzoo wrapper changed to gymnasium interface
#6211
opened Jun 13, 2025 by
alexander-zap
Loading…
5 of 10 tasks
added more options to registry environments
#5506
opened Aug 21, 2021 by
emmaajordan
Loading…
1 of 10 tasks
tighten test_process_pixels_gray bound
#5690
opened Jan 24, 2022 by
austereantelope
Loading…
1 of 10 tasks
[bug-fix] Fix sequence length buffer truncate
#5456
opened Jul 15, 2021 by
ervteng
Loading…
3 of 10 tasks
Added the "timeout_wait" parameter to the CLI ("--timeout-wait=").
#5812
opened Nov 13, 2022 by
max-krichenbauer
Loading…
2 of 10 tasks
Previous Next
ProTip!
Exclude everything labeled
bug
with -label:bug.