Struct OptimizerNetwork 

pub struct OptimizerNetwork {
    pub hidden_states: HashMap<String, Tensor>,
    pub cell_states: HashMap<String, Tensor>,
    pub weights: NetworkWeights,
    pub config: NeuralOptimizerConfig,
}

Simple neural network for learning optimization updates This is a simplified implementation for demonstration purposes

Fields

hidden_states: HashMap<String, Tensor>

LSTM-like state for each parameter

cell_states: HashMap<String, Tensor>

Cell states for LSTM

weights: NetworkWeights

Network weights

config: NeuralOptimizerConfig

Configuration

Implementations

impl OptimizerNetwork

pub fn new(config: NeuralOptimizerConfig) -> OptimizerResult<Self>

Create a new neural optimizer network

pub fn forward( &mut self, param_id: &str, gradient: &Tensor, parameter: &Tensor, ) -> OptimizerResult<Tensor>

Forward pass through the neural network to compute parameter update

pub fn reset_state(&mut self)

Reset network state

pub fn parameters(&self) -> Vec<&Tensor>

Get network parameters for meta-optimization

Trait Implementations

impl Clone for OptimizerNetwork

fn clone(&self) -> OptimizerNetwork

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for OptimizerNetwork

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

Formats the value using the given formatter.