Struct tangram_linear::EarlyStoppingOptions

pub struct EarlyStoppingOptions {
    pub early_stopping_fraction: f32,
    pub n_rounds_without_improvement_to_stop: usize,
    pub min_decrease_in_loss_for_significant_change: f32,
}

The parameters in this struct control how to determine whether training should stop early after each round or epoch.

Fields

early_stopping_fraction: f32

This is the fraction of the dataset that is set aside to compute the early stopping metric.

n_rounds_without_improvement_to_stop: usize

If this many rounds or epochs pass by without a significant improvement in the early stopping metric over the previous round or epoch, training will be stopped early.

min_decrease_in_loss_for_significant_change: f32

This is the minimum descrease in the early stopping metric for a round or epoch to be considered a significant improvement over the previous round or epoch.

Trait Implementations

impl Clone for EarlyStoppingOptions

fn clone(&self) -> EarlyStoppingOptions

Returns a copy of the value.

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

Performs copy-assignment from source.

impl Debug for EarlyStoppingOptions

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

Formats the value using the given formatter.