Struct vikos::teacher::GradientDescentAl

pub struct GradientDescentAl {
    pub l0: f64,
    pub t: f64,
}

Gradient descent with annealing learning rate

For the i-th event the learning rate is l = l0 * (1 + i/t)

Fields

l0: f64

Start learning rate

t: f64

Smaller t will decrease the learning rate faster

After t events the start learning rate will be a half l0, after two t events the learning rate will be one third l0 and so on.

Trait Implementations

impl<M> Teacher<M> for GradientDescentAl where M: Model

type Training = usize

Contains state which changes during the training, but is not part of the expertise

fn new_training(&self, _: &M) -> usize

Creates an instance holding all mutable state of the algorithm

fn teach_event<Y, C>(&self, num_events: &mut usize, model: &mut M, cost: &C, features: &M::Input, truth: Y) where C: Cost<Y>, Y: Copy

Changes models coefficents so they minimize the cost function (hopefully)