livxue / algcn

This repository contains the author's implementation in PyTorch for the paper "Adaptive Label-aware Graph Convolutional Networks for Cross-Modal Retrieval".

Adaptive Label-aware Graph Convolutional Networks

This repository contains the author's implementation in PyTorch for the paper "Adaptive Label-aware Graph Convolutional Networks for Cross-Modal Retrieval".

Dependencies

• Python (>=3.7)

• PyTorch (>=1.2.0)

• Scipy (>=1.3.2)

Datasets

You can download the features of the datasets from:

• MIRFlickr,
• NUS-WIDE(top-21 concepts)

Implementation

Here we provide the implementation of ALGCN, along with datasets. The repository is organised as follows:

• data/ contains the necessary dataset files for NUS-WIDE and MIRFlickr;
• models.py contains the implementation of the ALGCN;

Finally, main.py puts all of the above together and can be used to execute a full training run on MIRFlcikr or NUS-WIDE.

Process

• Place the datasets in data/
• Change the dataset in main.py to mirflickr or NUS-WIDE-TC21.
• Train a model:

python main.py

• Modify the parameter EVAL = True in main.py for evaluation:

python main.py

Citation

If you find our work or the code useful, please consider cite our paper using:

@article{qian2021adaptive,
  title={Adaptive Label-aware Graph Convolutional Networks for Cross-Modal Retrieval},
  author={Qian, Shengsheng and Xue, Dizhan and Fang, Quan and Xu, Changsheng},
  journal={IEEE Transactions on Multimedia},
  year={2021},
  publisher={IEEE},
  pages={1-1},
  doi={10.1109/TMM.2021.3101642}
}

algcn's Issues

adj.mat

你好！，我看文件里面缺少adj的生成这一块，能否补充一下呢？感谢。

datasets

你好，能否提供一下您使用的数据集呢？如果全部数据不方便提供的话能否提供一下生成adj.mat文件的代码呢？