feat(training): release V0.9.1 避障训练与基础策略迁移
This commit is contained in:
@@ -7,3 +7,7 @@ mujoco-warp==3.5.0
|
||||
warp-lang==1.15.0
|
||||
# RSL-RL 5.0.1 仍向 wandb.Settings 传递 start_method,0.29 已删除该字段。
|
||||
wandb==0.28.2
|
||||
# 基础策略迁移依赖已验证的 actor/normalizer/checkpoint 契约。
|
||||
rsl-rl-lib==5.0.1
|
||||
# 基础策略身份校验只使用 CPU ONNX Runtime。
|
||||
onnxruntime==1.29.0
|
||||
|
||||
Reference in New Issue
Block a user