Function binary_log_loss 

pub fn binary_log_loss<S1, S2, D1, D2>(
    y_true: &ArrayBase<S1, D1>,
    y_prob: &ArrayBase<S2, D2>,
    eps: f64,
) -> Result<f64>
where
    S1: Data<Elem = u32>,
    S2: Data<Elem = f64>,
    D1: Dimension,
    D2: Dimension,

Calculate binary log loss, also known as binary cross-entropy

This is the loss function used in (multinomial) logistic regression and neural networks (with softmax output).

Arguments

• y_true - Ground truth binary labels
• y_prob - Predicted probabilities for the positive class
• eps - Small value to avoid log(0)

Returns

• The log loss (float)

Examples

use scirs2_core::ndarray::array;
use scirs2_metrics::classification::binary_log_loss;

let y_true = array![0, 1, 1, 0];
let y_prob = array![0.1, 0.9, 0.8, 0.3];

let loss = binary_log_loss(&y_true, &y_prob, 1e-15).unwrap();