graph-neural-networks-basics

Explain Graph Neural Network architectures and message passing for graph data.

14|3|Updated Oct 28, 2025

Search Tags:#GNN #graph neural networks #GAT #message passing #GCN #graph data #GraphSAGE

One-click install

npx skills add https://github.com/tachyon-beep/skillpacks --skill graph-neural-networks-basics

Or copy as Structured Prompt for Agent▼

Please help me install this Agent Skill.
Skill: graph-neural-networks-basics
Source: https://github.com/tachyon-beep/skillpacks/tree/main/plugins/yzmir-neural-architectures/skills/graph-neural-networks-basics
Command: npx skills add https://github.com/tachyon-beep/skillpacks --skill graph-neural-networks-basics

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires torch, torch_geometric.

What problem does it solve?

This Skill demystifies Graph Neural Networks (GNNs), explaining why traditional CNNs and RNNs fail on irregular graph data. It guides you through the message passing framework and helps you choose the right GNN architecture (GCN, GraphSAGE, GAT) for tasks like molecular property prediction or social network analysis, ensuring you leverage graph structure effectively.

Core Features & Use Cases

GNN Architecture Selection: Understand the strengths of GCN (baseline), GraphSAGE (scalability, inductive), and GAT (attention, interpretability).
Message Passing Explained: Learn the core mechanism of aggregating neighbor information and updating node representations.
Use Case: You're analyzing a social network to predict user interests. This skill helps you decide if GNNs are truly beneficial over simpler models and guides you to GraphSAGE for its inductive capabilities on large, evolving graphs.

Quick Start

I have a large social network graph and need to predict properties of new users. Which GNN architecture should I use?