Struct Adam 

pub struct Adam {
    pub lr: f64,
    pub beta1: f64,
    pub beta2: f64,
    pub eps: f64,
    pub weight_decay: f64,
    /* private fields */
}

Adam optimizer (Kingma & Ba 2015).

Update rule: m ← β₁ m + (1-β₁) g v ← β₂ v + (1-β₂) g² m̂ = m / (1-β₁ᵗ), v̂ = v / (1-β₂ᵗ) params ← params - lr · m̂ / (√v̂ + ε)

Fields

lr: f64

Learning rate.

beta1: f64

First moment decay (default 0.9).

beta2: f64

Second moment decay (default 0.999).

eps: f64

Numerical stability constant (default 1e-8).

weight_decay: f64

L2 weight-decay (added to gradient, not decoupled).

Implementations

impl Adam

pub fn new(lr: f64) -> Self

Create a new Adam optimizer with default β₁=0.9, β₂=0.999, ε=1e-8.

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

Perform one Adam update step.

pub fn reset_state(&mut self)

Reset all state (moments and step counter).

Trait Implementations

impl Clone for Adam

fn clone(&self) -> Adam

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Adam

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

Formats the value using the given formatter.