Hugging Face's logo

wuschelschulz
/
gipf-zero

Reinforcement Learning
PyTorch
board-game
Model card Files
xet
 Community
gipf-zero / MODEL_CARD.md
wuschelschulz's picture
wuschelschulz
Publish completed overnight evaluation report
55e998f verified
preview code
|
Raw History Blame Contribute Delete
722 Bytes

GIPF Zero checkpoint

Custom PyTorch state_dict checkpoint for the GIPF Zero source. Load with training.model.load_model.

Source SHA-256: aa533bb8922231a3e0dc53dd2ff947cab6899ef09a9245463c89ac99ea776ef0.

Evaluation evidence: [{"report": "reports/final/20260909T191906Z/final-vs-champion-duel.json", "opponent": "/workspace/gipf/checkpoints/final/20260909T191906Z/champion-before-final-a9b830a90847.pt", "wins": 51, "losses": 29, "cutoffs": 0, "unfinished": 0}, {"report": "reports/final/20260909T191906Z/final-vs-greedy.json", "opponent": "greedy", "wins": 40, "losses": 0, "cutoffs": 0, "unfinished": 0}]. These are engine-baseline results, not a human or expert rating.