Instructions to use Swenlon/microduck-spin with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Microduck
How to use Swenlon/microduck-spin with Microduck:
# Replace SLOT with the slot specified in the model card (walk, stand, sitstand, ground_pick, kick_left, kick_right, roulade). sudo robotctl policy load SLOT Swenlon/microduck-spin
- Notebooks
- Google Colab
- Kaggle
Download inference_config.json from Swenlon/microduck-spin: direct link, hf CLI and curl.
- Browser
- Download file 1.31 kB
-
https://huggingface.co/Swenlon/microduck-spin/resolve/main/inference_config.json
- Command line
-
hf download hf://Swenlon/microduck-spin/inference_config.json
-
curl -L -o inference_config.json https://huggingface.co/Swenlon/microduck-spin/resolve/main/inference_config.json
1.31 kB
| { | |
| "schema_version": 1, | |
| "scope": "reference CPU MuJoCo rehearsal, not automatic robot daemon configuration", | |
| "policy_file": "policy.onnx", | |
| "policy_sha256": "c9bf6d76b327bc25d39c863ef0858cce477209cafd28f3f91a3f8233dcdf0033", | |
| "completed_updates": 20000, | |
| "scene": "src/mjlab_microduck/robot/microduck/scene_allcollisions.xml", | |
| "simulator_script": "scripts/infer_policy.py", | |
| "physics_timestep_s": 0.005, | |
| "control_period_s": 0.02, | |
| "actuator_delay": { | |
| "control_steps": 1, | |
| "seconds": 0.02, | |
| "cli": "--delay 1", | |
| "training_physics_steps": [3, 6], | |
| "training_seconds": [0.015, 0.03], | |
| "warning": "Do NOT copy training 3-6 physics steps into the CLI as --delay 3 6; CLI units are control periods." | |
| }, | |
| "observation_len": 61, | |
| "action_len": 14, | |
| "new_command_observations": true, | |
| "twist": [1, 0, 0], | |
| "twist_semantics": "constant behavior request, NOT forward speed 1 m/s", | |
| "head_command": [0, 0, 0, 0], | |
| "body_command": [0, 0, 0, 0, 0, 0], | |
| "action_scale": 1.0, | |
| "use_default_joint_pose_offset": true, | |
| "actuator_model": "BAM M6 XL330", | |
| "vin": 7.4, | |
| "vin_drop_gain": 0.1, | |
| "reference_versions": {"python": "3.12", "mujoco": "3.10.0", "onnxruntime": "1.24.4"}, | |
| "safety": {"hardware_validated": false, "safe_stop_validated": false, "fall_recovery_trained": false} | |
| } | |