Skip to content

Latest commit

 

History

History
18 lines (11 loc) · 1.07 KB

README.md

File metadata and controls

18 lines (11 loc) · 1.07 KB

This package is implementing the method in AAAI 2020 paper: Large-scale Multi-view Subspace Clustering in Linear Time. Please contact [email protected] if you have any questions.

A Brief Introduction to This Code

This package contains 5 matlab files and a dataset.

runlmv.m : If you just simply want to run this algorithm and see its performance, then visit this file. You can run this file after changing the dataset name in the code. Its performance will be recorded automatically. It is advisable to read the remarks in this file before you run it in matlab.

lmv.m : The core part of the code is implemented in this function. If you are seeking to learn more about the details of the mechanism of LMVSC and its programming implementation, see this file.

ClusteringMeasure : This function measures the performance of LMVSC with 3 indicators--Accuracy, NMI and Purity.

litekmeans : Perform K-Means clustering.

mySVD : Perform singular value decomposition.

dataset(dir) : Directory for datasets.