Sha-Lab / FEAT

The code repository for "Few-Shot Learning via Embedding Adaptation with Set-to-Set Functions"
MIT License
418 stars 84 forks source link

1 shot & 5 shot training #25

Closed Franklin-Yao closed 4 years ago

Franklin-Yao commented 4 years ago

Do you train 1 shot and 5 shot separately? Some papers use large numbers of shots, such as 15 to train a model once and test for 1 shot and 5 shot. MetaOptNet is the case.

Han-Jia commented 4 years ago

Hi, Franklin-Yao,

The 1-shot and 5-shot models are trained separately.