Struct NeuralOptimizer

pub struct NeuralOptimizer<F: Float + ScalarOperand> { /* private fields */ }

Neural network optimizer that uses scirs2-optimize stochastic methods

Implementations

impl<F> NeuralOptimizer<F>

where F: 'static + Send + Sync + Float + ScalarOperand,

pub fn new(method: StochasticMethod, options: StochasticOptions) -> Self

Create a new neural optimizer

pub fn sgd(learning_rate: F, max_iter: usize) -> Self

Create SGD optimizer for neural networks

pub fn adam(learning_rate: F, max_iter: usize) -> Self

Create Adam optimizer for neural networks

pub fn adamw(learning_rate: F, max_iter: usize) -> Self

Create AdamW optimizer for neural networks

pub fn step( &mut self, params: &mut NeuralParameters<F>, ) -> Result<(), OptimizeError>

Update neural network parameters using the selected optimizer

pub fn get_learning_rate(&self) -> f64

Get current learning rate

pub fn set_learning_rate(&mut self, lr: f64)

Set learning rate

pub fn reset(&mut self)

Reset internal state

pub fn method_name(&self) -> &'static str

Get optimizer method name