Struct LogCoshLoss

pub struct LogCoshLoss {
    pub reduction: LossReduction,
}

Computes the log cosh loss between inputs and targets.

Logcosh acts like L2 loss for small errors, ensuring stable gradients, and like the L1 loss for large errors, reducing sensitivity to outliers. This dual behavior offers a balanced, robust approach for regression tasks.

Fields

reduction: LossReduction

Reduction type. Default to LogCoshLoss::DEFAULT_REDUCTION

Implementations

impl LogCoshLoss

pub fn new() -> Self

Creates a new instance of LogCoshLoss.

impl LogCoshLoss

pub const DEFAULT_REDUCTION: LossReduction = LossReduction::None

Default value for the reduction parameter.

pub fn apply( &self, inputs: impl AsRef<Array>, targets: impl AsRef<Array>, ) -> Result<Array, Exception>

Computes the log cosh loss between inputs and targets.

Params

• inputs: predicted values
• targets: target values

Trait Implementations

impl Buildable for LogCoshLoss

type Builder = LogCoshLossBuilder

The builder type for this buildable type

impl Builder<LogCoshLoss> for LogCoshLossBuilder

type Error = Infallible

Error with building

fn build(self) -> Result<LogCoshLoss, Self::Error>

Build the type

impl Clone for LogCoshLoss

fn clone(&self) -> LogCoshLoss

Returns a copy of the value.

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

Performs copy-assignment from source.

impl Debug for LogCoshLoss

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

Formats the value using the given formatter.