Struct RAdam 

pub struct RAdam {
    pub lr: f32,
    pub beta1: f32,
    pub beta2: f32,
    pub eps: f32,
    pub weight_decay: f32,
    /* private fields */
}

Rectified Adam. Per-tensor state: two f32 buffers.

Fields

lr: f32

Learning rate.

beta1: f32

First-moment EMA decay β₁. Default 0.9.

beta2: f32

Second-moment EMA decay β₂. Default 0.999.

eps: f32

Denominator stability constant. Default 1e-8.

weight_decay: f32

L2 weight-decay coefficient (folded into the gradient — like classical Adam, not decoupled). Default 0.0.

Implementations

impl RAdam

pub fn new(lr: f32) -> Self

Construct with (β₁, β₂, ε, λ) = (0.9, 0.999, 1e-8, 0.0).

pub fn with_betas(self, b1: f32, b2: f32) -> Self

Override (β₁, β₂).

pub fn with_weight_decay(self, wd: f32) -> Self

Override the L2 weight-decay coefficient.

Trait Implementations

impl Clone for RAdam

fn clone(&self) -> RAdam

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for RAdam

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

Formats the value using the given formatter.

impl Optimizer for RAdam

fn step( &mut self, name: &str, _shape: &[usize], param: &mut [f32], grad: &[f32], )

fn end_iteration(&mut self)

Advance the global step counter. Most algorithms increment per call to [step], so most implementations leave this a no-op.

fn lr_scale(&self, _name: &str) -> f32

Per-tensor multiplier on the effective learning rate. Default is 1.0 for every name. Override when wrapping this crate to support per-name LR schedules (e.g. embedding-vs-attention splits, or the Gaussian-splat attribute-typed LR setup). The CPU impls in this crate currently honor this only when the caller passes a pre-scaled lr for the relevant call — backends are encouraged to consult it inside their fused kernel.