Hierarchical few-shot learning with feature fusion
driven by data and knowledge.

✔️ Requirements

• Ubuntu 20.04
• Python 3.8.5
• CUDA 11.1
• PyTorch 1.8.0

⚙️ Conda environmnet installation

conda env create --name HFFDK --file environment.yml
conda activate HFFDK

📚 Datasets

cd datasets
bash download_miniimagenet.sh
bash download_cub.sh
bash download_cifar_fs.sh
bash download_tieredimagenet.sh

📌 Quick start: testing scripts

To test in the 5-way K-shot setting:

bash scripts/test/{dataset_name}_5wKs.sh

For example, to test HFFDK on the CIFAR-FS dataset in the 5-way 1-shot setting:

bash scripts/test/cifar_fs_5w1s.sh

python test.py -dataset cifar_fs -datadir /home/data/cifar_fs -gpu 0 -extra_dir your_run_set -temperature_attn 5.0 

🔥 Training scripts

To train in the 5-way K-shot setting:

bash scripts/train/{dataset_name}_5wKs.sh

For example, to train HFFDK on the CIFAR-FS dataset in the 5-way 1-shot setting:

bash scripts/train/cifar_fs_5w1s.sh

python train.py -batch 64 -dataset cifar_fs -datadir /home/data/cifar_fs -gpu 0 -extra_dir your_run_set -temperature_attn 5.0 -lamb 0.5

📜 Citing HFFDK

If you find our code or paper useful to your research work, please consider citing our work using the following bibtex:

@article{Wu2023Hierarchical,
title = {Hierarchical Few-Shot Learning with Feature Fusion Driven by Data and Knowledge},
author = {Zhiping Wu and Hong Zhao},
journal = {Information Sciences},
pages = {119012},
year = {2023},
issn = {0020-0255},
doi = {https://doi.org/10.1016/j.ins.2023.119012},
url = {https://www.sciencedirect.com/science/article/pii/S0020025523005972}
}