joisino/laf
Code for "Training-free Graph Neural Networks and the Power of Labels as Features" (TMLR 2024)
This project offers a novel approach to graph neural networks (GNNs) that can be used immediately without extensive training. It takes in structured data, like relationships between items or documents, and outputs predictions or classifications for those items. Data scientists and machine learning engineers can use this to quickly analyze complex interconnected data.
No commits in the last 6 months.
Use this if you need to rapidly classify or predict properties within a graph-structured dataset without the time and computational resources typically required for model training.
Not ideal if your primary goal is to train highly specialized, fine-tuned GNNs for problems where extensive labeled data and training time are readily available.
Stars
58
Forks
7
Language
Python
License
MIT
Category
Last pushed
Aug 15, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/joisino/laf"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
pyg-team/pytorch_geometric
Graph Neural Network Library for PyTorch
a-r-j/graphein
Protein Graph Library
raamana/graynet
Subject-wise networks from structural MRI, both vertex- and voxel-wise features (thickness, GM...
pykale/pykale
Knowledge-Aware machine LEarning (KALE): accessible machine learning from multiple sources for...
dmlc/dgl
Python package built to ease deep learning on graph, on top of existing DL frameworks.