Struct AggMo 

pub struct AggMo { /* private fields */ }

Aggregated Momentum optimizer.

AggMo maintains multiple momentum buffers with different decay rates and averages their contributions to improve convergence.

Implementations

impl AggMo

pub fn new(config: AggMoConfig) -> Self

Create a new AggMo optimizer.

pub fn with_defaults( learning_rate: f32, momentum_coefficients: Vec<f32>, ) -> Self

Create AggMo with default configuration.

pub fn get_config(&self) -> &AggMoConfig

Get the configuration.

pub fn num_momentum_buffers(&self) -> usize

Get the number of momentum buffers per parameter.

Trait Implementations

impl Debug for AggMo

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

Formats the value using the given formatter.

impl OptimizerState for AggMo

fn zero_grad(&mut self) -> Result<()>

Zero out gradients

fn step(&mut self, parameters: &mut [Tensor]) -> Result<()>

Perform optimization step

fn get_lr(&self) -> f32

Get current learning rate

fn set_lr(&mut self, lr: f32)

Set learning rate

fn state_dict(&self) -> Result<HashMap<String, Tensor>>

Save optimizer state to dictionary

fn load_state_dict(&mut self, state: HashMap<String, Tensor>) -> Result<()>

Load optimizer state from dictionary