Skip to content

GPU implementation of GCN (Graph Convolutional Networks)

Notifications You must be signed in to change notification settings

hengdashi/cuda_gcn

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

54 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Parallelization of Graph Convolutional Network

This is the class project repo for CS 259 (High Performance Computing) Fall 2019 taught by Glenn Reinman. The sequential version of the code comes from here. Reddit dataset can be downloaded here. This project is a GPU acceleration of the Graph Convolutional Network.

Team members are: Zongze Li, Yuanhao Jia, Hengda Shi, Jintao Jiang

Setup

Convert reddit data format (networkx, numpy, scipy are required)

python3 reddit_preprocess.py

Run cora/citeseer/pubmed/reddit datasets on CPU

make seq
./gcn-seq cora

Run cora/citeseer/pubmed/reddit datasets on GPU

make cuda
./gcn-cuda cora

Clean up executable

make clean