Struct neuro::optimizers::Adam

pub struct Adam { /* fields omitted */ }

Adaptive moments estimation

Methods

impl Adam

pub fn new(learning_rate: PrimitiveType) -> Adam

Creates an Adam optimizer.

The exponential decay rates for the first and second moment estimates are set to 0.9 and 0.999 respectively. The epsilon value used for numerical stability is 1e-8.

pub fn with_param(
    learning_rate: PrimitiveType,
    beta1: PrimitiveType,
    beta2: PrimitiveType,
    eps: PrimitiveType
) -> Adam

Creates an Adam optimizer with the given parameters.

Arguments

• learning_rate - learning rate used to update the parameters of the layers.
• beta1 - exponential decay rate for the first moment estimate.
• beta2 - exponential decay rate for the second moment estimate.
• eps - small constant used for numerical stability.

Trait Implementations

impl Optimizer for Adam

fn name(&self) -> &str

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

fn update_time_step(&mut self)

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

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