My retinopathy model scored QWK 0.91. A classifier that never saw a retina scored 0.65.
I trained an EfficientNet-B0 to grade diabetic retinopathy on APTOS 2019. The five-fold ensemble reached QWK 0.909 on the held-out test split. Public solutions go past 0.93, so the score itself was not the interesting part. What I wanted to know was whether the model earned it for reasons that woul…
Read the full story at DEV Community — Machine Learning ↗
Timeline · 1 report
- 2026-11-18 16:27 · DEV Community — Machine Learning
My retinopathy model scored QWK 0.91. A classifier that never saw a retina scored 0.65.