benedekrozemberczki/ClusterGCN
A PyTorch implementation of Cluster-GCN, an algorithm for efficiently training large-scale graph convolutional networks on partitioned subgraphs.
Not currently ranked — collecting fresh signals.
star history
This repository provides a PyTorch implementation of the Cluster-GCN paper (KDD 2019), which uses graph clustering techniques to enable efficient SGD-based training of deep and large graph convolutional networks. It addresses the computational and memory challenges of training GCNs by sampling dense subgraphs identified through graph partitioning algorithms. The implementation includes the Amazon2M benchmark dataset with 2 million nodes and 61 million edges for scalability testing.
Frequently asked
- What is benedekrozemberczki/ClusterGCN?
- A PyTorch implementation of Cluster-GCN, an algorithm for efficiently training large-scale graph convolutional networks on partitioned subgraphs.
- Is ClusterGCN open source?
- Yes — benedekrozemberczki/ClusterGCN is open source, released under the GPL-3.0 license.
- What language is ClusterGCN written in?
- benedekrozemberczki/ClusterGCN is primarily written in Python.
- How popular is ClusterGCN?
- benedekrozemberczki/ClusterGCN has 805 stars on GitHub.
- Where can I find ClusterGCN?
- benedekrozemberczki/ClusterGCN is on GitHub at https://github.com/benedekrozemberczki/ClusterGCN.