https://github.com/eurus-holmes/vgcn

Code for Virtual Graph Convolutional Networks.
https://github.com/eurus-holmes/vgcn

Last synced: 2 days ago
JSON representation

Code for Virtual Graph Convolutional Networks.

• Host: GitHub
• URL: https://github.com/eurus-holmes/vgcn
• Owner: Eurus-Holmes
• License: mit
• Created: 2020-10-14T04:00:03.000Z (about 4 years ago)
• Default Branch: main
• Last Pushed: 2021-11-14T07:58:33.000Z (about 3 years ago)
• Last Synced: 2023-03-01T22:32:16.167Z (over 1 year ago)
• Language: Python
• Homepage:
• Size: 595 KB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Virtual Graph Convolutional Networks (VGCN)

In the virtual graph, a virtual node represents a timepoint. The embedding of a virtual node represents the “infection situations” at the timepoint. The virtual edges connect two virtual nodes at two timepoints, and the edge weights measure the significance of the virtual edge. Since the timepoints connected by the virtual edges can be outside the time window, VGCN can break the limitation of the time window by learning from neighbor nodes and improve the predictive accuracy.