Instructions to use GinDuck/microduck-gin-sock-c18 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Microduck
How to use GinDuck/microduck-gin-sock-c18 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-c18
- Notebooks
- Google Colab
- Kaggle
gin-sock-c18
Composite of other players' seeds, 4.771 s in the local x86 replica [4.640, 5.090, 4.655, 4.680, 4.790]. Seeds 2, 3, 4 and 5 replay HlibPlakhtii's quicksilver-sock-heist-nov18-v5 unchanged (run 01M7NV7ASVGPZH7AP0YKBT). Seed 1 replays jrubiosainz's Counter s5 4.790 (run 01M7JVJ71N7BABY0M9RMCZ). No own seed improvement in this entry. Fallback net: own. Simulation only. Credits: HlibPlakhtii, jrubiosainz, and everyone their trajectories build on (rumezhak, lphatnguyen).
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-c18
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
- 16