Struct RmsProp 

pub struct RmsProp {
    pub lr: f64,
    pub alpha: f64,
    pub eps: f64,
    pub momentum: f64,
    /* private fields */
}

RMSprop optimizer (Hinton 2012).

Maintains an exponential moving average of squared gradients: E[g²] ← α·E[g²] + (1-α)·g² params ← params - lr · g / (√E[g²] + ε)

With momentum > 0: vel ← momentum·vel + lr · g / (√E[g²] + ε) params ← params - vel

Fields

lr: f64

Base learning rate.

alpha: f64

Decay factor for the moving average (default 0.99).

eps: f64

Numerical stability constant.

momentum: f64

Momentum coefficient (0 = no momentum).

Implementations

impl RmsProp

pub fn new(lr: f64) -> Self

Create a new RMSprop optimizer with default α=0.99.

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

Perform one RMSprop update step.

pub fn reset_state(&mut self)

Reset accumulated state.

Trait Implementations

impl Clone for RmsProp

fn clone(&self) -> RmsProp

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for RmsProp

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

Formats the value using the given formatter.