Jupyter Notebook Tutorials for Machine Learning in Chemistry (MLChem)

This website hosts the Jupyter Notebook gallery for the undergraduate course Machine Learning in Chemistry (MLChem), first offered at the University of Wisconsin–Madison in Spring 2025 as CHEM 361, taught by Prof. Xuhui Huang. In this course, results from applying ML methods to chemical datasets were generated directly from executable Jupyter notebooks used in the lecture slides. These notebooks have been compiled into an online gallery with Sphinx. Each notebook is a self-contained tutorial that can be viewed as a rendered webpage or run interactively in Google Colab. The corresponding lecture slides are also embedded within each notebook.

Contents:

• Home
• Jupyter Notebook Tutorial Gallery
• Hands-on Homeworks
• How to Cite

Course Syllabus

Chapter	Contents	Tutorial Notebooks	Assignments
Chapter 1	Introduction and Setup of the Coding Environment	🎓 EnvSetup	💼 HW0
Chapter 2	From Molecules to Features – Preparing Training Data	💻 Descriptors, 💻 Fingerprints	💼 HW1
Chapter 3	Uncovering Patterns in Chemistry – Unsupervised Learning and Dimensionality Reduction	💻 KMeans, 💻 KCenter, 💻 DBSCAN, 💻 AgglomerativeClustering, 💻 PCA
Chapter 4	Predicting Chemical Outcomes – Supervised Learning for Classification and Regression	💻 LinReg, 💻 OverfitCV, 💻 Regularization, 💻 LogReg, 💻 kNN, 💻 DecisionTree	💼 HW2
Chapter 5	Neural Networks – Fundamentals and Applications in Chemical Modeling	💻 XOR, 💻 Softmax, 💻 MLP, 💻 OverfitNN
Chapter 6	Deep Neural Networks – Advanced Architectures for Chemical Applications	💻 CNN, 💻 GNN, 💻 RNN	💼 HW3
Chapter 7	Generating Chemical Data – AI Generative Models	💻 VAE
Chapter 8	Transforming Chemistry with Large Language Models – From Chemical to Protein Language Models	💻 ESM2	💼 HW4
Chapter 9	Machine Learning Force Fields – New Energy Functions in Chemical Simulations
Chapter 10	Reinforcement Learning for Chemical Discovery – Enabling Automated Synthesis
Chapter 11	Understanding Dynamics – ML Models for Molecular Simulations
Chapter 12	Predicting Protein Structure – AlphaFold and Its Applications

How to contribute

To maximize reproducibility and accessibility, we maintain the source code and notebooks at this repository xuhuihuang/mlchem. If you would like to contribute, please feel free to open an issue or a pull request to this repository.

How to cite

If you use materials in this repository, please cite:

“Machine Learning in Chemistry: A Data Centred, Hands-on Introductory Machine Learning Course for Undergraduate Students”, Mingyi Xue, Bojun Liu, and Xuhui Huang, ChemRxiv, DOI: 10.26434/chemrxiv-2025-9zldf.