from lerobot.policies.factory import make_policy
from lerobot.configs.policies import PreTrainedConfig

cfg = PreTrainedConfig.from_pretrained("DreamMachines/cube_picknplace_cube_picknplace_pretrain_act_480x640_v0_small", revision="v002000")
policy = make_policy(cfg)

Available: v002000, v004000, v006000, v008000, v010000, v012000, v014000, v016000, v018000, v020000

Quick eval on a real DK1

Dataset: <a href="https://huggingface.co/datasets/DreamMachines/cube_picknplace_480x640_v0_small" target="_blank" rel="noopener noreferrer" class="text-blue-400 hover:text-blue-300 underline underline-offset-2 transition-colors"> DreamMachines/cube_picknplace_480x640_v0_small
Step (this revision): 20000
Batch size: 32
Total steps target: 20000
Optimizer: adamw (lr 3e-05 )

make eval MODEL=DreamMachines/cube_picknplace_cube_picknplace_pretrain_act_480x640_v0_small

See docs/source/policy_pi05_README.md for π₀.₅-specific notes.