Struct neuro::optimizers::SGD

pub struct SGD { /* fields omitted */ }

Stochastic Gradient Descent

Methods

impl SGD

pub fn new(learning_rate: PrimitiveType) -> SGD

Creates a Stochastic Gradient Descent optimizer.

pub fn with_param(learning_rate: PrimitiveType, momentum: PrimitiveType) -> SGD

Creates a Stochastic Gradient Descent optimizer with momentum estimation.

Trait Implementations

impl Optimizer for SGD

fn name(&self) -> &str

fn update_parameters(&mut self, layer: &mut dyn Layer, layer_idx: usize)

fn initialize_parameters(&mut self, layers_dims: Vec<(Dim, Dim)>)

fn save(&self, file: &File) -> Result<(), Error>

fn update_time_step(&mut self)