MLIPs Roadmap

mlips-roadmap is a collection of tutorials, examples, and workflows for building Machine Learning Interatomic Potentials (MLIPs) using modern frameworks.

This repository serves as a practical and incremental roadmap for anyone interested in training MLIPs for atomistic simulations, materials science, and catalysis.

---

📘 Contents

Tutorials (Notebooks)

• 01_MACE_Cu.ipynb
  Train a MACE potential for Cu, including dataset generation, training, and evaluation.

• 02_NequIP_Al.ipynb
  Train a NequIP potential for Al bulk systems and surface models.

More tutorials coming soon.

---

🎯 Goals

• Provide simple, reproducible MLIP tutorials.
• Offer a structured roadmap for beginners and researchers.
• Gradually evolve into a reusable Python package with shared utilities and scripts.
• Promote modern MLIP tools.

---

🚀 Getting Started

Clone the repository:

git clone https://github.com/mminotaki/mlips-roadmap.git
cd mlips-roadmap