Machine_Learning_with_Chemistry 🧪💻

Welcome to the Machine_Learning_with_Chemistry repository! This repository contains tutorials and code examples on using machine learning methods to predict chemical properties. The goal of these tutorials is to demonstrate how machine learning can enhance our understanding of chemistry and improve predictions while keeping the chemical intuition as the foundation.

Directory Structure 📂

1. 🧪 Identify carbonyl groups like a machine! 🌳🔍 This code trains a decision tree model on IR spectra to predict whether molecules contain a carbonyl functional group. It demonstrates key steps in applying supervised machine learning to chemistry, including loading and preprocessing data, training a decision tree classifier, testing on holdout data, and analyzing model performance. With this model, you can see molecules through the eyes of a machine and predict the presence of specific chemical functional groups without the need for wet lab experiments!

2. DeepChem tutorials 🧪📚 This directory contains tutorials on how to use DeepChem, a deep learning library, to train and test models on molecular data. The code examples in this directory are based on the book "Deep Learning in Life Sciences" and provide practical implementations of deep learning techniques in chemistry.

3. GNNs 🌐🔬 Explore the world of Graph Neural Networks (GNNs) and their applications in predicting properties of molecules. This directory includes tutorials that provide a basic understanding of GNNs. The examples are based on the work of Dr. Pat Walter and Dr. Vijay Pandey, and they serve as a starting point for learning about GNNs in chemistry.

4. Machine Learning with Scikit Learn 🧪🔬 Dive into the world of machine learning with Scikit Learn, a popular machine learning library. This directory contains tutorials that cover various simple machine learning models, including

   • linear regression,
   • logistic regression,
   • k-nearest neighbors,
   • decision trees,
   • support vector machines,
   • random forests,
   • gradient boosted trees,
   • k-means clustering, and
   • principal component analysis (PCA).

5. Vertical Excitation Energiess with Random Forest 🌳🔍 This ML model demonstrates how to use a random forest model to predict vertical excitation energies of molecules in gas phase. The snapshots are obtained from molecular dynamics simulations and the VEEs are obtained quantum mechanical calculations.

Contributions and Updates 🚀

This repository is a work in progress, and contributions are welcome! If you have any suggestions, improvements, or additional code examples related to machine learning in chemistry, feel free to submit a pull request.

Let's learn and explore together!

Happy coding! 💻🔬🚀