feat(vla): align transformer training stack and rollout validation
This commit is contained in:
3
roboimi/vla/eval_utils.py
Normal file
3
roboimi/vla/eval_utils.py
Normal file
@@ -0,0 +1,3 @@
|
||||
def execute_policy_action(env, action):
|
||||
"""Execute policy outputs using EE-action semantics."""
|
||||
env.step(action)
|
||||
Reference in New Issue
Block a user