Struct NesterovAcceleratedGradient 

pub struct NesterovAcceleratedGradient {
    pub max_iter: usize,
    pub tol: Float,
    pub learning_rate: Float,
    pub adaptive_lr: bool,
    pub momentum: Float,
    pub lr_decay: Float,
}

Accelerated Gradient Descent with Nesterov momentum

Fields

max_iter: usize

Maximum number of iterations

tol: Float

Convergence tolerance

learning_rate: Float

Learning rate

adaptive_lr: bool

Whether to use adaptive learning rate

momentum: Float

Momentum parameter (typically 0.9)

lr_decay: Float

Learning rate decay factor

Implementations

impl NesterovAcceleratedGradient

pub fn adaptive() -> Self

Create optimizer with adaptive learning rate

pub fn minimize<F, GradF>( &self, f: F, grad_f: GradF, x0: Array1<Float>, ) -> Result<Array1<Float>, String>

where F: Fn(&Array1<Float>) -> Float, GradF: Fn(&Array1<Float>) -> Array1<Float>,

Minimize smooth objective using Nesterov accelerated gradient descent

Trait Implementations

impl Default for NesterovAcceleratedGradient

fn default() -> Self

Returns the “default value” for a type.