zhangsz1998 / Self-Expressive-Network

Self-Expressive Network

This repository contians the Pytorch implementation of our paper Learning a Self-Expressive Networl for Subspace Clustering

Usage

Download datasets.zip from

# dataset: MNIST, EMNIST, FashionMNIST, CIFAR10
python main.py --dataset=MNIST

Feature Generation

The preprocessed features (Scattering transform for MNIST/FashonMNIST/EMNIST, MCR$^2$ for CIFAR10) are already provided in ./datasets folder. Code for feature generation for MNIST, Fashion and EMNIST are provided in feature generation.py.

# dataset: MNIST, EMNIST, FashionMNIST, CIFAR10
python feature_generation.py --dataset=MNIST

For MCR$^2$ feature generation, please refer to ryanchankh/mcr2.

Update

Files under ./datasets were not properly uploaded, you may download the datasets from here.

Contact

Please contact [e-mail] if you have any question on the codes.