act_open_lamp_door

Standard LeRobot ACT trained from scratch on SO101 data.

Policy	ACT (policy.type=act)
Dataset	villekuosmanen/armnetbench_open_lamp_door
Task	open_lamp_door
Action dim	12 (bimanual)
Cameras	top, left_wrist, right_wrist
Training	GCloud A100-80GB (alpha-act-a100-2), batch 16, action horizon 30
Loss @ 10k	0.164
W&B project	act_open_lamp_door
W&B run	do4hnyiv

Checkpoints

Archived copies:

Step	Path
010000	checkpoints/010000/pretrained_model/
008000	checkpoints/008000/pretrained_model/

The latest step (010000) is also copied to the repository root for direct loading.

Usage

from lerobot.policies.act.modeling_act import ACTPolicy
policy = ACTPolicy.from_pretrained("pravsels/act_open_lamp_door")