benedekrozemberczki/APPNP
A PyTorch implementation of "Predict then Propagate: Graph Neural Networks meet Personalized PageRank" (ICLR 2019).
This helps researchers, data scientists, or machine learning engineers improve how they classify interconnected items like research papers, social network users, or biological proteins. You provide information about how these items are linked (an edge list), descriptive features for each item, and some initial labels. The output is a more accurate classification of all items, even those without initial labels.
374 stars. No commits in the last 6 months.
Use this if you need to classify items that are connected in a network, and you want to leverage those connections for better prediction accuracy with a flexible and efficient model.
Not ideal if your data lacks explicit connections between items or if you are not comfortable with machine learning model training and evaluation.
Stars
374
Forks
52
Language
Python
License
GPL-3.0
Category
Last pushed
Nov 06, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/benedekrozemberczki/APPNP"
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.