FSPDF

Training

dataset (miniImagenet/cifar/CUB)

methodname (S2M2_R/rotation)

① first run rotation ② second run S2M2_R

1. train frequency-domain

python train_dct.py --dataset [DATASETNAME] --method [METHODNAME] --model WideResNet28_10 --train_aug --dct_status

2. train spatial-domain

python train_dct.py --dataset [DATASETNAME] --method [METHODNAME] --model WideResNet28_10 --train_aug

Save features

1. save frequency-domain features

python save_features.py --dataset [DATASETNAME] --method S2M2_R --model WideResNet28_10 --train_aug --dct_status

2. save spatial-domain features

python save_features.py --dataset [DATASETNAME] --method S2M2_R --model WideResNet28_10 --train_aug --dct_status

2. save dual-domain features

python save_features_both.py --dataset [DATASETNAME] --method S2M2_R --model WideResNet28_10 --train_aug

Testing

1. test frequency-domain

python test_dct.py --dataset [DATASETNAME] --method S2M2_R --model WideResNet28_10 --n_shot [1/5] --train_aug --dct_status	

1. test spatial-domain

python test_dct.py --dataset [DATASETNAME] --method S2M2_R --model WideResNet28_10 --n_shot [1/5] --train_aug	

1. test dual-domain

python test_dct_both.py --dataset [DATASETNAME] --method S2M2_R --model WideResNet28_10 --n_shot [1/5] --train_aug

Acknowledgment

Our project references the codes in the following repos.

• S2M2_R