Struct LarsOptimizer 

pub struct LarsOptimizer { /* private fields */ }

LARS optimizer (Layer-wise Adaptive Rate Scaling).

LARS scales the learning rate for each layer based on the ratio of the parameter norm to the gradient norm. This is particularly effective for large batch training.

Reference: You et al. “Large Batch Training of Convolutional Networks” (2017)

Implementations

impl LarsOptimizer

pub fn new(config: OptimizerConfig, trust_coef: f64, exclude_bias: bool) -> Self

Create a new LARS optimizer.

Arguments

• config - Optimizer configuration
• trust_coef - Trust coefficient for adaptive LR (default: 0.001)
• exclude_bias - Whether to exclude bias from LARS adaptation (default: true)

Trait Implementations

impl Debug for LarsOptimizer

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

Formats the value using the given formatter.

impl Optimizer for LarsOptimizer

fn step( &mut self, parameters: &mut HashMap<String, Array<f64, Ix2>>, gradients: &HashMap<String, Array<f64, Ix2>>, ) -> TrainResult<()>

Update parameters with computed gradients.

fn zero_grad(&mut self)

Zero all gradients.

fn get_lr(&self) -> f64

Get current learning rate.

fn set_lr(&mut self, lr: f64)

Set learning rate.

fn state_dict(&self) -> HashMap<String, Vec<f64>>

Get optimizer state for checkpointing.

fn load_state_dict(&mut self, state: HashMap<String, Vec<f64>>)

Load optimizer state from checkpoint.