GGCM

Repository for the paper : GGCM: Gradient-Guided for Cross-Domain Few-Shot Learning

If you have any questions/advices/potential ideas, welcome to contact me by [email protected].

1 Dependencies

A anaconda envs is recommended:

conda create --name py36 python=3.6
conda activate py36
conda install pytorch torchvision -c pytorch
pip3 install scipy>=1.3.2
pip3 install tensorboardX>=1.4
pip3 install h5py>=2.9.0

2 datasets

We evaluate our methods on five datasets: mini-Imagenet works as source dataset, cub, cars, places, and plantae serve as the target datasets, respectively.

1. The datasets can be conviently downloaded and processed as in FWT.
2. Remember to modify your own dataset dir in the 'options.py'.
3. We follow the same the same auxiliary target images as in previous work meta-FDMixup, and the used jsons have been provided in the output dir of this repo.

3 pretraining

As in most of the previous CD-FSL methods, a pretrained feature extractor baseline is used.

• you can directly download it from this link, rename it as 399.tar, and put it to the ./output/checkpoints/baseline

4 Usages

Our method is target set specific, and we take the cub target set under the 5-way 1-shot setting as an example.

1. training for Baseline

python train_baseline.py --method Baseline --stop_epoch 400 --modelType Student --target_set cub --name Baseline/CUB/1shot --train_aug --warmup baseline --n_shot 1

• DATASET: cub/cars/places/plantae

2. testing for Baseline

python test.py --method Baseline --name  Baselien/CUB/1shot --dataset cub --save_epoch 399 --n_shot 1

• DATASET: cub/cars/places/plantae

3. training for GGCM

python train.py  --method GGCM --aug_rate 0.5 --meta_layers 3 4 --lconsist 5.0  --stop_epoch 400 --modelType Student --target_set cub --name GGCM/CUB/1shot \
--train_aug --warmup baseline --n_shot $shot

• DATASET: cub/cars/places/plantae

4. testing for GGCM

python test.py --method GGCM --name  GGCM/CUB/1shot --dataset cub --save_epoch 399 --n_shot 1

• DATASET: cub/cars/places/plantae