Struct LSTMOptimizer 

pub struct LSTMOptimizer<T: Float + Debug + Send + Sync + 'static> { /* private fields */ }

LSTM-based neural optimizer with meta-learning capabilities

Implementations

impl<T: Float + Default + Clone + Send + Sync + Sum + for<'a> Sum<&'a T> + ScalarOperand + Debug> LSTMOptimizer<T>

pub fn new(config: LearnedOptimizerConfig) -> Result<Self>

Create a new LSTM optimizer

pub fn lstm_step<S, D>( &mut self, parameters: &ArrayBase<S, D>, gradients: &ArrayBase<S, D>, loss: Option<T>, ) -> Result<Array<T, D>>

where S: Data<Elem = T>, D: Dimension + Clone,

Perform LSTM-based optimization step

pub fn meta_learning_step(&mut self, tasks: &[MetaTask<T>]) -> Result<T>

Meta-learning step for optimizer adaptation

pub fn transfer_to_domain( &mut self, target_tasks: &[MetaTask<T>], ) -> Result<TransferResults<T>>

Transfer learning to new optimization domain

pub fn get_metrics(&self) -> &LSTMOptimizerMetrics

Get current performance metrics

pub fn get_state_analysis(&self) -> OptimizationStateAnalysis<T>

Get optimization state analysis

Trait Implementations

impl<T: Clone + Float + Debug + Send + Sync + 'static> Clone for LSTMOptimizer<T>

fn clone(&self) -> LSTMOptimizer<T>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<T: Debug + Float + Debug + Send + Sync + 'static> Debug for LSTMOptimizer<T>

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

Formats the value using the given formatter.