DaT SPECT abnormality classification — DrivenData comp 311, 3rd place
prize_submission.tar.gz (2.9 GB) — the complete prize package: code
(github.com/paul-bd/DAT_challenge @ main), pinned requirements, and the
shipped weights (50 fusion10 TorchScript members + striatal ellipse
localiser + calibration).
sha256: 9752e5b3273841b343b73131cf55a038fcffa501899c1b915981802a05e1da65
Unpack, pip install -r requirements.txt, then
DATA_DIR=/path/with/niftis OUTPUT_PATH=out.csv python inference/main.py.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support