search

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

Accuracy of pre-trained model for Tiered Imagenet Dataset #77

Closed mayug closed 2 years ago

mayug commented 2 years ago

Hi,

I am trying to recreate the pretrained resnet12 model for tieredimagenet. Could you please tell me what is the best_val_acc_sim and best_val_acc_dist you got when you pretrained on tiered imagenet?

Thanks and Regards Mayug

Han-Jia commented 2 years ago

Hi,

For the 1-shot test on all meta-val classes, the best val acc dist is around 16.62% and the best_val_acc_sim is around 18.86%.