Crate linfa_reduction

Dimensionality reduction techniques

This crate provides algorithms for dimensionality reduction in data analysis. They can be used to transform data from a high-dimensional space into a lower dimensional space such that some property of the data is retained.

The following implementations are available:

• Principal Component Analysis - projects data linearily and retains the largest variance
• Diffusion Map - applies kernel methods and projects close regions together

Re-exports

pub use pca::Pca;
pub use utils::to_gaussian_similarity;

Modules

diffusion_map	Diffusion Map
error
pca	Principal Component Analysis
utils

Structs

DiffusionMap

Embedding of diffusion map technique