A short and easy PyTorch implementation of E(n) Equivariant Graph Neural Networks

A Python package for data-mining the QM9 dataset

Predicting properties of small molecules using MPNN on QM9 dataset

Advanced Guidance Strategies for Conditional Molecule Generation with Flow Matching

Graph Neural Network creation module, implemented in Tensorflow 2 with examples using the module and the iGNNition library for fast GNN prototyping.

A from-scratch Message Passing Neural Network (MPNN) for molecular property prediction, built and benchmarked on QM9-style data. Classical AI/DL baseline for a quantum-enhanced drug discovery research project.

A Variational Autoencoder in Google Colab to generate and visualize novel molecular structures for potential drug discovery applications, using the QM9 dataset and SMILES representation.

A update QM9 benchmark, update form the recent articles

Raman Spectrum Predictor: A machine learning model trained on QM9-like molecules to estimate Raman peak positions near 1600 cm⁻¹. Includes training pipeline, Streamlit UI, and RDKit-based featurization.

Master's Thesis: Constrained Molecular Graph Generation with Diffusion Models

Topology-aware EGNNs with persistent homology features for HOMO–LUMO gap prediction on QM9

A graph neural network project for predicting molecular HOMO-LUMO gap values using QM9-style molecular graph data.

Does persistent homology help an E(3)-equivariant model predict molecular dipole moments and polarizability tensors? A controlled study with matched negative controls, whose answer in this configuration is no.