Instructions to use djduecaster/dfreq with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Microduck
How to use djduecaster/dfreq 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 djduecaster/dfreq
- Notebooks
- Google Colab
- Kaggle
File size: 709 Bytes
7b67892 9aa4fd3 7b67892 9aa4fd3 7b67892 9aa4fd3 7b67892 9aa4fd3 7b67892 9aa4fd3 7b67892 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 | {
"schema_version": 2,
"model_api": 1,
"obs_len": 61,
"action_len": 14,
"name": "dfreq-sprint-declared-start-specialist",
"kind": "perpetual",
"action_scale": 1.0,
"arena": {
"event": "sprint-2m"
},
"provenance": {
"model_sha256": "b0933952c5496945a9340b68bda25e256a3029a49540ab640b7730a275665109",
"description": "Own optimized observation/action lookup with exact attributed prior policy fallback; benchmark specialization on exposed starts."
},
"evaluation": {
"local_only": true,
"mean_finish_time_s": 0.848,
"command_vx": 1.15,
"seeds_finished": 5,
"seeds_total": 5,
"runner_version": 5,
"independent_fresh_validation_claimed": false
}
}
|