Multi-scale Adaptive Task Attention Network for Few-Shot Learning(ICPR-2022)

This code implements the Multi-scale Adaptive Task Attention Network (MATANet).

Our code is based on CovaMNet.

Citation

If you find our work useful, please consider citing our work using the bibtex:

@inproceedings{chen2022icpr,
	author  = {Chen, Haoxing and Li, Huaxiong and Li, Yaohui and Chen, Chunlin},
	title   = {Multi-scale Adaptive Task Attention Network for Few-Shot Learning},
	booktitle = {International Conference on Pattern Recognition(ICPR)},
	year    = {2022},
}

Prerequisites

Linux
Python 3.6
Pytorch 1.0+
GPU + CUDA CuDNN
pillow, torchvision, scipy, numpy

Datasets

Dataset download link:

Note: You need to manually change the dataset directory.

Few-shot Classification

Train a 5-way 1-shot model based on Conv-128F (on miniImageNet dataset):

python MATA_Train.py --dataset_dir ./datasets/miniImageNet --data_name miniImageNet --way_num 5 --shot_num 1

Test model on the test set:

python MATA_Test.py --dataset_dir ./datasets/miniImageNet --data_name miniImageNet --way_num 5 --shot_num 1
./results/MATA_miniImageNet_MATA_5Way_1Shot_K5/model_best.pth.tar --basemodel MATA

Contacts

Please feel free to contact us if you have any problems.