Instructions to use ignite-dohyun/act_so101_cup_eraser_target_v1_10ep with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LeRobot
How to use ignite-dohyun/act_so101_cup_eraser_target_v1_10ep with LeRobot:
- Notebooks
- Google Colab
- Kaggle
metadata
library_name: lerobot
tags:
- robotics
- lerobot
- act
- so101
- bimanual
- imitation-learning
datasets:
- ignite-dohyun/so101_cup_eraser_target_v1
ACT SO-101 Bimanual Cup/Eraser Target v1 - 10 episodes
ACT policy trained with LeRobot on ignite-dohyun/so101_cup_eraser_target_v1.
- Robot: bimanual SO-101 (
bi_so_follower) - Dataset episodes: 10
- Cameras:
left_wrist,left_top,right_wrist - Training steps: 100000
- Local output:
/workspace/outputs/train/act_so101_cup_eraser_target_v1_10ep
Use with LeRobot by passing this repo as the policy path, or use the local checkpoint directory with --policy.path.