Struct Lamb 

pub struct Lamb { /* private fields */ }

Layer-wise Adaptive Moments optimizer for Batch training (LAMB).

Combines Adam-style adaptive moment estimation with layer-wise trust ratio scaling for stable large-batch training.

Implementations

impl Lamb

pub fn new(lr: f32) -> Result<Self, OptimError>

Creates LAMB with required learning rate.

pub fn with_betas(self, beta1: f32, beta2: f32) -> Result<Self, OptimError>

Sets beta coefficients (beta1, beta2) used for computing running averages.

pub fn with_weight_decay(self, weight_decay: f32) -> Result<Self, OptimError>

Sets L2 weight decay factor in [0, +inf).

pub fn with_epsilon(self, epsilon: f32) -> Result<Self, OptimError>

Sets epsilon value, must be finite and > 0.

pub fn clear_state(&mut self)

Drops optimizer state (for example when restarting training).

pub fn learning_rate(&self) -> f32

Returns current learning rate.

pub fn set_learning_rate(&mut self, lr: f32) -> Result<(), OptimError>

Overrides current learning rate.

pub fn step( &mut self, parameter_id: u64, weights: &mut Tensor, grad: &Tensor, ) -> Result<(), OptimError>

Applies one update to raw tensor weights.

pub fn step_graph_node( &mut self, graph: &mut Graph, node: NodeId, ) -> Result<(), OptimError>

Applies one update to a trainable graph node by its NodeId.

Trait Implementations

impl Clone for Lamb

fn clone(&self) -> Lamb

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Lamb

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

Formats the value using the given formatter.

impl LearningRate for Lamb

fn learning_rate(&self) -> f32

Returns current optimizer learning rate.

fn set_learning_rate(&mut self, lr: f32) -> Result<(), OptimError>

Sets optimizer learning rate after validation.

impl StepOptimizer for Lamb

fn step( &mut self, parameter_id: u64, weights: &mut Tensor, grad: &Tensor, ) -> Result<(), OptimError>