DreamMachines/cube_pnp_d1_act_r1

cube_pnp/d1_act_r1

ACT (Action Chunking with Transformers) — imitation learning that predicts short action chunks instead of single steps. See the paper.

Trained with LeDream (a fork of LeRobot).

Training

• Dataset: DreamMachines/cube_pnp_view_d1
• Step (this revision): 80000
• Batch size: 32
• Total steps target: 80000
• Optimizer: adamw (lr 3e-05)

Available revisions

Every training checkpoint is uploaded as its own git revision; the latest is also on main. Load a specific step with:

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

cfg = PreTrainedConfig.from_pretrained("DreamMachines/cube_pnp_d1_act_r1", revision="v002000")
policy = make_policy(cfg)

Available: v080000

Quick eval on a real DK1

make eval MODEL=DreamMachines/cube_pnp_d1_act_r1

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