Function kernel_target_alignment 

pub fn kernel_target_alignment(
    kernel_matrix: &[Vec<f64>],
    labels: &[f64],
) -> Result<f64>

Compute kernel-target alignment (KTA) between a kernel matrix and target labels.

KTA measures how well a kernel matrix aligns with the ideal kernel matrix derived from target labels. Higher values indicate better alignment.

Arguments

• kernel_matrix - The kernel matrix K
• labels - Binary labels (+1 or -1) for each sample

Returns

• Alignment score in range [-1, 1]

Examples

use tensorlogic_sklears_kernels::kernel_utils::kernel_target_alignment;

let K = vec![
    vec![1.0, 0.8, 0.2],
    vec![0.8, 1.0, 0.3],
    vec![0.2, 0.3, 1.0],
];
let labels = vec![1.0, 1.0, -1.0];

let alignment = kernel_target_alignment(&K, &labels).unwrap();
// High alignment means kernel separates classes well