• PCA (Principal Component Analysis)
• t-Sne (t-Distributed Stochastic Neighbor Embedding)
• UMAP (Uniform Manifold Approximation and Projection)

Dataset being used for above experimentation is the famous digit recognition dataset MNIST dataset. We have used the sklearn library for downloading the same.

Following are the visualizations generated by different algorithms:

• PCA (2d)
• PCA (3d)
• t-SNE
• UMAP

Use the package manager pip to install the requirements.

pip install pandas
pip install sklearn
pip install seaborn
pip install umap-learn

python src/main.py

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

Please make sure to update tests as appropriate.

MIT