Trait SecondOrderOptimizer 

pub trait SecondOrderOptimizer<A: Float + ScalarOperand + Debug + Send + Sync, D: Dimension> {
    // Required methods
    fn step_second_order(
        &mut self,
        params: &Array<A, D>,
        gradients: &Array<A, D>,
        hessian_info: &HessianInfo<A, D>,
    ) -> Result<Array<A, D>>;
    fn reset(&mut self);
}

Trait for second-order optimization methods

Required Methods

fn step_second_order( &mut self, params: &Array<A, D>, gradients: &Array<A, D>, hessian_info: &HessianInfo<A, D>, ) -> Result<Array<A, D>>

Update parameters using second-order information

fn reset(&mut self)

Reset optimizer state

Implementors

impl<A: Float + ScalarOperand + Debug + Send + Sync> SecondOrderOptimizer<A, Dim<[usize; 1]>> for Newton<A>

impl<A: Float + ScalarOperand + Debug + Send + Sync, D: Dimension + Send + Sync> SecondOrderOptimizer<A, D> for LBFGS<A, D>