This is the source code for the paper "Graph-driven Deep Multi-view Clustering with Self-paced Learning". The paper has been accepted by the journal Knowledge-Based Systems.
The GDMVC algorithm described in the paper is implemented using the Python language and the PyTorch framework. The algorithm needs to use a GPU when training the model, and if you want to train on the CPU, you need to modify the code.
The dataset directory contains the datasets used. To prevent the compression file from becoming too large, only a portion of the datasets has been uploaded.
The main program file of the code is: train_duibi.py. Before running this program, make sure the dataset exists in the dataset directory.
The runtime environment is Python 3.8.18 and Torch 2.1.0+cu118.
If you have any questions, please feel free to contact me at yff235351@stu.xjtu.edu.cn.