ruqu-qgat-mol

Quantum Graph Attention for Molecules - Combines quantum mechanics with graph attention networks for molecular simulation.

Part of the ruQu quantum computing suite by ruv.io.

Features

• Quantum Orbital Features - Atomic orbital-inspired feature extraction
• Graph Attention - Multi-head attention over molecular graphs
• Property Prediction - Energy, HOMO-LUMO gap, dipole moments
• Message Passing - Quantum-enhanced node/edge updates
• Molecular Fingerprints - Learned representations for similarity

Installation

[dependencies]
ruqu-qgat-mol = "0.1"

Quick Start

use ruqu_qgat_mol::{MolecularGraph, QuantumGAT, GATConfig};

let config = GATConfig {
    n_layers: 3,
    hidden_dim: 64,
    n_heads: 4,
    ..Default::default()
};

let model = QuantumGAT::new(config)?;
let molecule = MolecularGraph::from_smiles("CCO")?; // Ethanol

let energy = model.predict_energy(&molecule)?;
println!("Predicted energy: {:.4} eV", energy);

License

MIT License - see LICENSE