ic-lab-duth/FusedGCN4HLS

Systolic Three Matrix Multiplier for Graph Convolutional Networks using High Level Synthesis

29
/ 100
Experimental

This project offers a specialized computational architecture designed to speed up how quickly Graph Convolutional Networks (GCNs) can analyze graph-structured data. It takes your graph's connections (adjacency matrix) and node information (feature matrix), along with learned network weights, to quickly produce updated node representations. This is for machine learning engineers or researchers working with GCNs to process large graph datasets more efficiently.

No commits in the last 6 months.

Use this if you need to accelerate the inference speed of Graph Convolutional Networks, especially for large graph-structured datasets like those found in social networks or recommendation systems.

Not ideal if you are developing new GCN models from scratch or if your primary need is for training GCNs rather than just performing inference.

graph-neural-networks machine-learning-inference high-performance-computing hardware-acceleration deep-learning-optimization
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 7 / 25

How are scores calculated?

Stars

23

Forks

2

Language

C++

License

MIT

Last pushed

Jul 29, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/ic-lab-duth/FusedGCN4HLS"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.