willyfh/graph-transformer
An unofficial implementation of Graph Transformer (Masked Label Prediction: Unified Message Passing Model for Semi-Supervised Classification) - IJCAI 2021
This is a tool for machine learning practitioners and researchers who need to classify data points when only a small portion of them are labeled. It takes in graph-structured data, where nodes represent data points and edges represent relationships, and outputs predicted labels for the unlabeled nodes. This project is ideal for those working on semi-supervised classification tasks in various domains.
No commits in the last 6 months. Available on PyPI.
Use this if you need to classify interconnected data points efficiently, even when most of your data lacks labels, leveraging the relationships between them.
Not ideal if your data is not inherently structured as a graph or if you have a fully labeled dataset for traditional supervised learning.
Stars
35
Forks
6
Language
Python
License
MIT
Category
Last pushed
Apr 20, 2024
Commits (30d)
0
Dependencies
2
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/willyfh/graph-transformer"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.