Function trustworthiness 

pub fn trustworthiness<S1, S2>(
    x: &ArrayBase<S1, Ix2>,
    x_embedded: &ArrayBase<S2, Ix2>,
    n_neighbors: usize,
    metric: &str,
) -> Result<f64>
where
    S1: Data,
    S2: Data,
    S1::Elem: Float + NumCast,
    S2::Elem: Float + NumCast,

Calculate trustworthiness score for a dimensionality reduction

Trustworthiness measures to what extent the local structure is retained when projecting data from the original space to the embedding space.

A trustworthiness of 1.0 means all local neighborhoods are perfectly preserved.

Arguments

• x - Original data, shape (n_samples, n_features)
• x_embedded - Embedded data, shape (n_samples, n_components)
• n_neighbors - Number of neighbors to consider
• metric - Metric to use (currently only ‘euclidean’ is supported)

Returns

• Result<f64> - Trustworthiness score between 0.0 and 1.0