12 Star 214 Fork 109

GVPAscend/MindSpeed-RL

actor_log_p 和 rule reward计算加速 可合并
ci-pipeline-passed
ascend-cla/yes
stat/needs-squash
!248 liuxinyang 18
2025-04-24 17:23
dev add profiler 可合并
ci-pipeline-failed
ascend-cla/yes
!240 tangmengcheng 17
2025-04-22 15:20
确定性计算以及TD保序特性 可合并
ci-pipeline-failed
ascend-cla/yes
stat/needs-squash
!239 tanwenqin 44
2025-04-22 00:57
新增partial rollout特性 存在冲突
ci-pipeline-failed
ascend-cla/yes
stat/needs-squash
!237 LookAround 11
2025-04-21 18:14
reward 可合并
ascend-cla/yes
!234 朱意霖 1
2025-04-21 15:31
!233 Nurxat 11
2025-04-21 09:27
feat: 支持td profiling 可合并
ci-pipeline-failed
ascend-cla/yes
stat/needs-squash
!231 sunjianping 60
2025-04-20 12:04
fix: 临时方案,td所在节点增加多个ray客户端分担调用 存在冲突
ascend-cla/yes
stat/needs-squash
!217 sunjianping 1
2025-04-17 22:04
[resharding] support EP and EP_expend 可合并
ascend-cla/yes
!216 zhoubeirong 2
2025-04-17 15:41
1
https://gitee.com/ascend/MindSpeed-RL.git
git@gitee.com:ascend/MindSpeed-RL.git
ascend
MindSpeed-RL
MindSpeed-RL

搜索帮助