Struct PcaResult 

pub struct PcaResult {
    pub components: Vec<Vec<f64>>,
    pub explained_variance: Vec<f64>,
    pub explained_variance_ratio: Vec<f64>,
    pub mean: Vec<f64>,
    pub std_dev: Vec<f64>,
    pub n_samples: usize,
    pub n_features: usize,
    pub iterations_used: usize,
}

PCA result containing components and explained variance

Fields

components: Vec<Vec<f64>>

Principal component vectors (n_components x n_features), row-major

explained_variance: Vec<f64>

Variance explained by each component (eigenvalues)

explained_variance_ratio: Vec<f64>

Proportion of total variance explained by each component

mean: Vec<f64>

Feature means used for centering (needed to project new data)

std_dev: Vec<f64>

Feature standard deviations (if scaling was used)

n_samples: usize

Number of samples in the input data

n_features: usize

Number of features in the input data

iterations_used: usize

Number of iterations used (last component for PowerIteration, power iters for Randomized)

Implementations

impl PcaResult

pub fn transform(&self, data: &[Vec<f64>]) -> Vec<Vec<f64>>

Project multiple data points into the reduced PCA space.

Uses ndarray matrix multiplication for efficiency.

pub fn transform_one(&self, point: &[f64]) -> Vec<f64>

Project a single data point into the reduced PCA space.