5-way Image Classification on FC100 (test)
0.625AccuracyMetaOptNet-SVM-trainval
Evaluation results
| Method | Links | |
|---|---|---|
| MetaOptNet-SVM-trainvalbackbone=ResNet-12, shot=5-shot, training_set=meta-train + meta-validation2019.04 | 0.625 | |
| TADAMbackbone=ResNet-12, shot=5-shot2019.04 | 0.561 | |
| MetaOptNet-SVMbackbone=ResNet-12, shot=5-shot2019.04 | 0.555 | |
| MetaOptNet-RRbackbone=ResNet-12, shot=5-shot2019.04 | 0.553 | |
| ProtoNets (our backbone)backbone=ResNet-12, shot=5-shot2019.04 | 0.525 | |
| Prototypical Networksbackbone=64-64-64-64, shot=5-shot2019.04 | 0.486 | |
| MetaOptNet-SVM-trainvalbackbone=ResNet-12, shot=1-shot, training_set=meta-train + meta-validation2019.04 | 0.472 | |
| MetaOptNet-SVMbackbone=ResNet-12, shot=1-shot2019.04 | 0.411 | |
| MetaOptNet-RRbackbone=ResNet-12, shot=1-shot2019.04 | 0.405 | |
| TADAMbackbone=ResNet-12, shot=1-shot2019.04 | 0.401 | |
| ProtoNets (our backbone)backbone=ResNet-12, shot=1-shot2019.04 | 0.375 | |
| Prototypical Networksbackbone=64-64-64-64, shot=1-shot2019.04 | 0.353 |