SmolVLA – iCub, MuJoCo (lift the blue block)
SmolVLA policy trained with LeRobot on iCub demonstrations recorded in MuJoCo.
- Task:
lift the blue block - Inputs:
observation.state,observation.images.left,observation.images.right,observation.images.front - Output:
action(chunk size 60) - Training: 20,000 steps, batch size 22
Usage
from lerobot.policies.smolvla.modeling_smolvla import SmolVLAPolicy
policy = SmolVLAPolicy.from_pretrained("mklovin/smolvla_icub_mujoco_train50")
- Downloads last month
- 19