Instructions to use tonghuiwang123/so100-act-60ep with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LeRobot
How to use tonghuiwang123/so100-act-60ep with LeRobot:
- Notebooks
- Google Colab
- Kaggle
so100-act-60ep
LeRobot policy for SO-ARM100 / SO100, trained on tonghuiwang123/test10
(40 episodes, 12948 frames, 30fps, 2 cameras @1280x720).
- Checkpoint step: 200000
- Task:
Grab the white cube to the white cup
真机部署(相机命名必须按下表,顺序不能错)
| 采数时的物理位置 | 本模型要求的命名 |
|---|---|
| front | front |
| side | side |
lerobot-record \
--robot.type=so100_follower --robot.port=/dev/ttyACM0 \
--robot.id=my_awesome_follower_arm \
--robot.cameras="{
front: {type: opencv, index_or_path: 0, width: 1280, height: 720, fps: 30},
side: {type: opencv, index_or_path: 1, width: 1280, height: 720, fps: 30},
}" \
--dataset.repo_id=tonghuiwang123/eval_so100-act-60ep \
--dataset.single_task="Grab the white cube to the white cup" \
--policy.path=tonghuiwang123/so100-act-60ep
index_or_path: 0 必须是采集数据时放在 front 位的那台相机,1 是 side 位的。
分辨率必须 1280x720,与训练数据一致。
- Downloads last month
- -