Struct TrainingResult 

pub struct TrainingResult {
    pub initial_loss: f64,
    pub final_loss: f64,
    pub epochs_completed: usize,
    pub convergence_achieved: bool,
    pub training_time_ms: u64,
}

Result of a completed training run.

Captures the loss before and after training, how many epochs actually ran, whether the loop converged early, and the wall-clock training time. This is the honest record of what the SGD loop did (not a stub).

Fields

initial_loss: f64

Mean squared error measured on the full dataset before the first weight update.

final_loss: f64

Mean squared error measured on the full dataset after the final epoch.

epochs_completed: usize

Number of epochs actually executed (may be less than config.epochs if converged).

convergence_achieved: bool

True if the loss plateaued below the convergence threshold before all epochs ran.

training_time_ms: u64

Wall-clock training time in milliseconds.

Trait Implementations

impl Clone for TrainingResult

fn clone(&self) -> TrainingResult

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for TrainingResult

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

Formats the value using the given formatter.