Model weights and predictions for reproducible benchmarking experiments in MedMNIST v2

<p>This data repository is associated with our&nbsp;<a href="https://github.com/MedMNIST/experiments">GitHub code</a></p> <ol> <li><code>weights_*.zip</code>: <ul> <li>PyTorch, AutoKeras and Google AutoML Vision are provided for MedMNIST2D.</li> <li>PyTorch and AutoKeras are provided for MedMNIST3D.</li> <li>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&nbsp;<code>PIL.Image.NEAREST</code>.</li> <li>Snapshots for&nbsp;<code>auto-sklearn</code>&nbsp;are not uploaded due to the embarrassingly large model sizes (lots of model ensemble).</li> </ul> </li> <li><code>predictions.zip</code>: We also provide all standard prediction files by PyTorch, auto-sklearn, AutoKeras and Google AutoML Vision, which works with&nbsp;<code>medmnist.Evaluator</code>. Each file is named as&nbsp;<code>{flag}_{split}_[AUC]{auc:.3f}_[ACC]{acc:.3f}@{run}.csv</code>, e.g.,&nbsp;<code>bloodmnist_test_[AUC]0.997_[ACC]0.957@autokeras_3.csv</code>.</li> </ol>