List your company
ResearchFormerly WizWand

Image Classification on Melanoma Private Kaggle 2020 (test)

94.38AUCOur Submission
92.591293.055693.5293.9844Dec 6, 2020
Updated 8mo ago

Evaluation results

MethodLinks
Our SubmissionTTA=30, Backbone=EfficientNetB52020.1294.38
AUC-M (Meta)Loss=AUC Margin with Metadata, TTA=30, Backbone=EfficientNetB52020.1294.23
AUC-S (Meta)Loss=AUC Surrogate with Metadata, TTA=30, Backbone=EfficientNetB52020.1294.09
AUC-MLoss=AUC Margin Loss, TTA=30, Backbone=EfficientNetB52020.1293.93
AUC-M (Meta)Loss=AUC Margin with Metadata, TTA=None, Backbone=EfficientNetB52020.1293.8
AUC-SLoss=AUC Surrogate Loss, TTA=30, Backbone=EfficientNetB52020.1293.64
AUC-S (Meta)Loss=AUC Surrogate with Metadata, TTA=None, Backbone=EfficientNetB52020.1293.58
AUC-MLoss=AUC Margin Loss, TTA=None, Backbone=EfficientNetB52020.1293.57
CELoss=Cross-Entropy, TTA=30, Backbone=EfficientNetB52020.1293.45
AUC-SLoss=AUC Surrogate Loss, TTA=None, Backbone=EfficientNetB52020.1293.32
FocalLoss=Focal Loss, TTA=30, Backbone=EfficientNetB52020.1293.03
CELoss=Cross-Entropy, TTA=None, Backbone=EfficientNetB52020.1292.85
FocalLoss=Focal Loss, TTA=None, Backbone=EfficientNetB52020.1292.66