DermaMNIST

Dataset Information

The DermaMNIST is based on the HAM10000, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. The dataset consists of 10,015 dermatoscopic images categorized as 7 different diseases, formulized as a multi-class classification task. We split the images into training, validation and test set with a ratio of 7:1:2. The source images of 3×600×450 are resized into 3×28×28.

Task: multi-class

Labels:

0: actinic keratoses and intraepithelial carcinoma, 1: basal cell carcinoma, 2: benign keratosis-like lesions, 3: dermatofibroma, 4: melanoma, 5: melanocytic nevi, 6: vascular lesions

Samples:

• Train: 7007
• Validation: 1003
• Test: 2005

Experiment Parameter

Learning Rate: 3e-4

Training Epoch: 100

Convergence Epoch: 25