You can deploy this model and start earning money today!

wbell7/
starkhacks_act_cell_step10k

Name: starkhacks_act_cell_step10k
Availability: InStock

wbell7/starkhacks_cell_pickplace

apache-2.0

Views

No views yet

StarkHacks 2026 — ACT policy, step10k

Task: Pick the orange cell and place it into the empty slot of the green module.

Rank: #1 (primary — best generalization). Still actively learning; epoch 11 is the classic ACT sweet spot for small-demo datasets. Evaluate this first.

Trained on wbell7/starkhacks_cell_pickplace (58 episodes, 29K frames, cameras: top + wrist, 30 fps, 640×480).

Training run: https://wandb.ai/wrbell7/starkhacks/runs/nbyv34do

Metrics at this checkpoint

metric	value
step	10000
epoch	10.97
train loss	0.143
batch size	32
optimizer	AdamW, lr=1e-5, wd=1e-4
mixed precision	bf16
hardware	AMD MI300X VF (ROCm 7.2, torch 2.7.1+rocm6.3)
policy	ACT, ResNet-18 backbone, 52M params

Usage

bash

lerobot-record \
  --robot.type=so101_follower \
  --robot.port=/dev/so101_follower \
  --robot.id=my_follower_arm \
  --robot.cameras="{top: {type: opencv, index_or_path: /dev/cam_ugreen, width: 640, height: 480, fps: 30}, wrist: {type: opencv, index_or_path: /dev/cam_arc, width: 640, height: 480, fps: 30}}" \
  --dataset.repo_id=local/eval_step10k \
  --dataset.single_task="Pick the orange cell and place it into the empty slot of the green module." \
  --dataset.num_episodes=3 \
  --policy.path=wbell7/starkhacks_act_cell_step10k \
  --dataset.push_to_hub=false \
  --display_data=false