Skip to content
humanoidrobots.training

HopeJR · published recipe

hand manipulation via ACT (train-your-own recipe)

Source ↗HopeJR robot page

Record

Robot
HopeJR huggingface/hopejr
Framework
LeRobot (ACT) · LeRobot
Policy license
not stated by the package
File
no weights file named
Source
https://huggingface.co/docs/lerobot/hope_jr
Note
No pretrained HopeJR checkpoint is published; the docs give the lerobot-train command (policy.repo_id=<USER>/hand_test_policy placeholder), an example dataset nepyope/hand_record_test_with_video_data, and a W&B run for reference.

Not in the tab yet

To replay this policy in the browser we need an ONNX export of a trained checkpoint, the observation and action contract copied from the authors' own sim2sim or deploy script, and a MuJoCo model of the robot that the contract matches. Then it must walk 60 s in our runtime without a fall. Until then the source link is the way in.