config

n_obs_steps: int = 1
chunk_size: int = 30
n_action_steps: int = 1
n_decoder_layers: int = 6
optimizer_lr: float = 5e-5

use

lerobot v0.5.2

lerobot-rollout   --strategy.type=base   --policy.path=JiaMinEsc/act_so101_pick_grape_put_bowl   --robot.type=so101_follower   --robot.port=/dev/ttyACM1   --robot.cameras="{ hand: {type: opencv, index_or_path: 0, width: 640, height: 480, fps: 30, fourcc: 'MJPG'}, env: {type: opencv, index_or_path: 4, width: 640, height: 480, fps: 30, fourcc: 'MJPG'}}"   --task="Pick up the grapes and put them in the bowl"   --duration=60

lerobot under v0.5.1

Downloads last month
74
Safetensors
Model size
78.5M params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Dataset used to train JiaMinEsc/act_so101_pick_grape_put_bowl