Creating a tSNE and UMAP embedding

t-Distributed Stochastic Neighbor Embedding (t-SNE) and Uniform Manifold Approximation and Projection (UMAP) are both dimensionality reduction techniques commonly used in ML and data visualization.

t-SNE is a non-linear technique that aims to visualize high-dimensional data in a lower-dimensional space while preserving the local structure of the data. t-SNE is particularly useful for revealing clusters or groups of data points that may not be immediately apparent in the original high-dimensional space.

UMAP is another non-linear dimensionality reduction technique that has gained popularity due to its scalability and efficiency. UMAP also focuses on preserving local structures and employs a different mathematical method based on graph theory and optimization techniques. UMAP is often used for visualizing large datasets with millions of data points.

Both algorithms suffer from being computationally expensive, especially when dealing with large...