CorGIE is a visualization tool for graph neural networks. This repo hosts the frontend code of CorGIE, and the other repo hosts the backend code.
You can upload your own datasets here.
You need nodejs (version>=14.0) and the package management tool (either yarn or npm) to run our code locally.
To install dependencies, run yarn install or npm install in the project root directory.
To start the application, run yarn start or npm start. You will be able to open the page with the URL http://localhost:3000 on your browser.
Note: this would serve the data files locally from public/data/ statically, so you would not be able to use the upload page. To upload your own dataset, you need to install the backend or use our demo website.
Zipeng Liu, Yang Wang, Jürgen Bernard, Tamara Munzner. Visualizing Graph Neural Networks with CorGIE: Corresponding a Graph to Its Embedding. TVCG 2022.