We provide a demo for GCN on MNIST dataset based on torch7 and also PyTorch.
To run this demo, you should install these dependencies:
luarocks install torchnet
luarocks install optnet
install GCN:
cd $DIR/GCN
bash install.sh
run this demo:
cd $DIR/MNIST_demo
bash ./scripts/Train_MNIST.sh
To run the pytorch branch, please refer to this branch.
The experimental results of the comparison algorithm are directly quoted from the corresponding papers.
This demo is partially based on the code of Orientation Response Networks. If you use this demo please cite our paper and ORN.
@article{GaborCNNs, title={Gabor Convolutional Networks}, author={Luan, Shangzhen and chen, chen and Zhang, Baochang* and han, jungong and Liu, Jianzhuang}, year={2018}, IEEE Trans. Image processing. }
@INPROCEEDINGS{Zhou2017ORN, title={Oriented Response Networks}, author={Zhou, Yanzhao and Ye, Qixiang and Qiu, Qiang and Jiao, Jianbin}, booktitle = {CVPR} year={2017}, }