The Self-Optimal-Transport Feature Transform

Benchmark Model Rank Results
few-shot-image-classification-on-cifar-fs-5-way-1-shotPT+MAP+SF+SOT (transductive)#1Accuracy: 89.94
few-shot-image-classification-on-cifar-fs-5-way-5-shotPT+MAP+SF+SOT (transductive)#2Accuracy: 92.83
few-shot-image-classification-on-cub-200-5-way-1-shotPT+MAP+SF+SOT (transductive)#1Accuracy: 95.80
few-shot-image-classification-on-cub-200-5-way-5-shotPT+MAP+SF+SOT (transductive)#2Accuracy: 97.12
few-shot-image-classification-on-mini-imagenet-5-way-1-shotPT+MAP+SF+SOT (transductive)#5Accuracy: 85.59
few-shot-image-classification-on-mini-imagenet-5-way-5-shotPT+MAP+SF+SOT (transductive)#7Accuracy: 91.34