Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

DMNIST

Benchmarks

Task NameDataset NameSOTA ResultTrend
Image ClassificationDMNIST (test)
Test Accuracy84.28
16
Misclassification DetectionDMNIST-LT (rho = 0.01)
AUPR88.76
10
ClassificationDMNIST-LT rho = 0.01 (test)
Test Accuracy69.42
10
Misclassification DetectionDMNIST
AUPR (Conf)96.17
8
Showing 4 of 4 rows