Model weights and predictions for reproducible benchmarking experiments in MedMNIST v2

This data repository is associated with our GitHub code

1. weights_*.zip:
   • PyTorch, AutoKeras and Google AutoML Vision are provided for MedMNIST2D.
   • PyTorch and AutoKeras are provided for MedMNIST3D.
   • If you are using PyTorch model weights, please note that the ResNet18_224 / ResNet50_224 models are trained with images resized to 224 x 224 by PIL.Image.NEAREST.
   • Snapshots for auto-sklearn are not uploaded due to the embarrassingly large model sizes (lots of model ensemble).
2. predictions.zip: We also provide all standard prediction files by PyTorch, auto-sklearn, AutoKeras and Google AutoML Vision, which works with medmnist.Evaluator. Each file is named as {flag}_{split}_[AUC]{auc:.3f}_[ACC]{acc:.3f}@{run}.csv, e.g., bloodmnist_test_[AUC]0.997_[ACC]0.957@autokeras_3.csv.