Instructions to use GinDuck/microduck-gin-sock-c17 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Microduck
How to use GinDuck/microduck-gin-sock-c17 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 GinDuck/microduck-gin-sock-c17
- Notebooks
- Google Colab
- Kaggle
gin-sock-c17
Composite counter, 4.781 s in the local replica [4.64, 5.095, 4.675, 4.68, 4.815]. Seeds 2-5 replay rumezhak's microduck-sock-heist-composite-v10 (run 01M7N0C6V63JMP4M9FM2G1). Seed 1 is jrubiosainz's microduck-sock-counter-s5 sequence (run 01M7JVJ71N7BABY0M9RMCZ) after an own tail climb (same 4.64 s). Fallback net: own. Credits to rumezhak and jrubiosainz.
A perpetual policy for the microduck (61-D observation, 14 actions, 50 Hz). Runs until told otherwise โ a gait, loaded into a policy slot.
Run it on a robot
sudo robotctl policy load <slot> GinDuck/microduck-gin-sock-c17
The observation normalizer is baked into policy.onnx; feed raw observations.
manifest.json follows schema 2 of the microduck policy manifest (docs/policy-manifest.md in the daemon repo).
Training
- repo:
pollen-robotics/microduck_rl - branch:
talent-3 - commit:
debcfcae0
- Downloads last month
- 20