MEGY is a machine learning project aimed at predicting key magnetic properties of nanoparticles, specifically:

• Coercivity
• Remanence magnetization
• Saturation magnetization

This project is a part of a larger initiative, AI Lab Assistant, designed to assist with the prediction and optimization of nanoparticle synthesis. The project leverages a variety of descriptors for nanoparticles and includes models for both coated and uncoated nanoparticles.

Contains Jupyter notebooks (.ipynb files) with the following:

• Models: Trained machine learning models for predicting the magnetic properties.
• Visualizations: Scatter plots and feature analysis plots using SHAP (Shapley Additive Explanations) for interpreting the model’s predictions.

Includes the datasets used for training the models. These datasets contain information about nanoparticle properties and their corresponding magnetic characteristics.

To get started with MEGY, clone the repository and install the required dependencies:

git clone https://github.com/yourusername/MEGY.git
cd MEGY
pip install -r requirements.txt