Struct AdaGrad 

pub struct AdaGrad {
    pub lr: f64,
    pub eps: f64,
    pub weight_decay: f64,
    /* private fields */
}

AdaGrad optimizer (Duchi et al. 2011).

Accumulates the sum of squared gradients per parameter and scales the learning rate accordingly: G ← G + g² params ← params - lr · g / (√G + ε)

Fields

lr: f64

Base learning rate.

eps: f64

Numerical stability constant.

weight_decay: f64

L2 weight-decay (coupled).

Implementations

impl AdaGrad

pub fn new(lr: f64) -> Self

Create a new AdaGrad optimizer.

pub fn step( &mut self, params: &mut Vec<f64>, grad: &[f64], ) -> OptimizeResult<()>

Perform one AdaGrad update step.

pub fn reset_state(&mut self)

Reset accumulated state.

Trait Implementations

impl Clone for AdaGrad

fn clone(&self) -> AdaGrad

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for AdaGrad

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

Formats the value using the given formatter.