Samanvith1404/MicroGNN
A tiny Graph Neural Network framework built from scratch with a minimal autograd engine.
This project helps machine learning engineers and students understand the inner workings of Graph Neural Networks (GNNs) and automatic differentiation. It takes raw graph data and node features, then processes them through a transparent GNN implementation to show how predictions are made and how the model learns. If you're studying machine learning or preparing for technical interviews, this provides a clear, step-by-step view of GNN mechanics.
Use this if you want to learn the fundamental, line-by-line mechanics of how Graph Neural Networks and their learning process (autograd) function without relying on complex deep learning libraries.
Not ideal if you need to build high-performance, production-ready GNN models for large datasets, as it prioritizes clarity and education over optimization and advanced features.
Stars
8
Forks
—
Language
Python
License
MIT
Category
Last pushed
Nov 15, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Samanvith1404/MicroGNN"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
JonathanRaiman/theano_lstm
:microscope: Nano size Theano LSTM module
google/tangent
Source-to-Source Debuggable Derivatives in Pure Python
ahrefs/ocannl
OCANNL: OCaml Compiles Algorithms for Neural Networks Learning
yoshoku/numo-openblas
Numo::OpenBLAS builds and uses OpenBLAS as a background library for Numo::Linalg
statusfailed/catgrad
a categorical deep learning compiler