Appendix
Test the UMAP dimensionality reduction algorithm with your own data. This can be useful to find items with similar characteristics comparing across multiple dimensions. All of this will run live on your browser, so I won't recommend it with datasets with more than 1000 records.
From the UMAP library description: Uniform Manifold Approximation and Projection (UMAP) is a dimension reduction technique that can be used for visualisation similarly to t-SNE, but also for general non-linear dimension reduction.
The default example uses the USDA's nutrients dataset by default. You can load your own data in CSV format. The first row should contain the column names. The application will automatically detect the data types and will allow you to select the columns you want to use for the UMAP algorithm. You can also filter the data using navio.dev
Reactive and Reusable Widgets
This is a modular application built with many interconnected widgets. You can study the 👩🏻💻 source code on this observable Notebook, and to learn more about how to build them visit visit reactive and reusable widgets guide.
The following is an overlay of some of the widgets used in this application and their connections
Copyright 2024 John Alexis Guerra Gómez under MIT License