Pradnya1208/Dimensionality-Reduction-Techniques ? reverse-engineered prompt

Build me a clean Jupyter notebook project that demonstrates the main dimensionality reduction techniques on the included data, especially PCA, factor analysis, t SNE, UMAP, Isomap, and a simple random forest based feature importance approach.

I want the notebook to load the provided train.csv and test.csv, do the basic preprocessing needed, and then show how each technique reduces the feature space and what the result looks like. Please include clear plots and short explanations so someone learning this can follow along, and make the notebook easy to run from top to bottom without missing steps. If a method needs a bit of setup or the latest docs, look it up online and use the current best practice.

Keep the code readable, organize the sections well, and save any generated visualizations in the output folder. If there are parts of the dataset that need cleaning or encoding first, handle that in the notebook too.