TheMesocarp/koho
Full spectrum sheaf neural network over arbitrary CW complexes.
This project enables developers and researchers to build neural networks that can model complex data relationships across multi-dimensional structures, not just simple networks. It takes in topological data representations (like cell complexes) and outputs advanced diffusion models that understand connections beyond simple nodes, useful for those working with intricate data geometries. It is for machine learning researchers and advanced data scientists exploring novel neural network architectures.
Use this if you are a machine learning researcher or advanced data scientist needing to apply diffusion models to data structured as cellular complexes, where traditional graph neural networks are insufficient.
Not ideal if you are a practitioner looking for a ready-to-use tool for standard machine learning tasks, or if your data can be adequately modeled with simpler graph-based neural networks.
Stars
16
Forks
4
Language
Rust
License
AGPL-3.0
Category
Last pushed
Mar 01, 2026
Monthly downloads
4
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/TheMesocarp/koho"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
EnzymeAD/Enzyme
High-performance automatic differentiation of LLVM and MLIR.
Oxen-AI/Oxen
Lightning fast data version control system for structured and unstructured machine learning...
LaurentMazare/tch-rs
Rust bindings for the C++ api of PyTorch.
SunDoge/dlpark
A Rust Library for High-Performance Tensor Exchange with Python
Photoroom/datago
A natively parallel dataloader for Python, written in Rust. Serving data at GB/s speeds, while...