Using Non-negative Matrix Factorization (NMF) and Variational Autoencoder (VAE) machine learning models to analyze spatial and spectral features of hyperspectral cathodoluminescence (CL) spectroscopy images taken from hybrid inorganic-organic perovskite materials on the nanometer scale.
The .h5py hyperspectral datasets are not publicly available at this time as they have not yet been published. However, when they are published, they will be included in this repository.
The Jupyter Notebooks in this project require Python version 3.12.2 or later.
Create a new virtual environment and clone the repository with git clone https://github.com/jonperk318/machine-learning-analysis-of-hyperspectral-data.
Open the cloned directory and run the command pip install -r requirements.txt. This contains necessary Jupyter core
packages as well as all other packages, including numpy, scikit-learn, pytorch, and pyroved. Then, run jupyter notebook
and navigate to the notebook you wish to view.