Struct CrossEntropyLoss

pub struct CrossEntropyLoss { /* private fields */ }

Cross-entropy loss function.

The cross-entropy loss is defined as: L = -sum(y_true * log(y_pred)) where y_true are target probabilities and y_pred are predicted probabilities.

It is commonly used for classification problems.

Examples

use scirs2_neural::losses::CrossEntropyLoss;
use scirs2_neural::losses::Loss;
use ndarray::{Array, arr1, arr2};

let ce = CrossEntropyLoss::new(1e-10);

// One-hot encoded targets and softmax'd predictions for a 3-class problem
let predictions = arr2(&[
    [0.7, 0.2, 0.1],  // First sample, class probabilities
    [0.3, 0.6, 0.1]   // Second sample, class probabilities
]).into_dyn();

let targets = arr2(&[
    [1.0, 0.0, 0.0],  // First sample, true class is 0
    [0.0, 1.0, 0.0]   // Second sample, true class is 1
]).into_dyn();

// Forward pass to calculate loss
let loss = ce.forward(&predictions, &targets).unwrap();

// Backward pass to calculate gradients
let gradients = ce.backward(&predictions, &targets).unwrap();

Implementations

impl CrossEntropyLoss

pub fn new(epsilon: f64) -> Self

Create a new cross-entropy loss function

Arguments

• epsilon - Small value to add to predictions to avoid log(0)

Trait Implementations

impl Clone for CrossEntropyLoss

fn clone(&self) -> CrossEntropyLoss

Returns a copy of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for CrossEntropyLoss

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for CrossEntropyLoss

fn default() -> Self

Returns the “default value” for a type.

impl<F: Float + Debug> Loss<F> for CrossEntropyLoss

fn forward( &self, predictions: &Array<F, IxDyn>, targets: &Array<F, IxDyn>, ) -> Result<F>

Calculate the loss between predictions and targets

fn backward( &self, predictions: &Array<F, IxDyn>, targets: &Array<F, IxDyn>, ) -> Result<Array<F, IxDyn>>

Calculate the gradient of the loss with respect to the predictions

impl Copy for CrossEntropyLoss