Struct CustomScheduler 

pub struct CustomScheduler<A, F>
where
    A: Float + Debug + ScalarOperand,
    F: FnMut(usize) -> A,
{ /* private fields */ }

A custom scheduler that uses a closure to compute the learning rate

Implementations

impl<A, F> CustomScheduler<A, F>

where A: Float + Debug + ScalarOperand, F: FnMut(usize) -> A,

pub fn new(_initial_lr: A, lrfunc: F) -> CustomScheduler<A, F>

Create a new custom scheduler with the given initial learning rate and computation function

Arguments

• initial_lr - The initial learning rate (used only for documentation)
• lr_func - A function that takes the current step count and returns the learning rate

Example

use optirs_core::schedulers::{CustomScheduler, LearningRateScheduler};

// Create a scheduler that reduces the learning rate by 10% every 10 steps
let mut scheduler = CustomScheduler::new(0.1, |step| {
    0.1 * 0.9_f64.powi((step / 10) as i32)
});

assert_eq!(scheduler.get_learning_rate(), 0.1);
scheduler.step();

pub fn get_step_count(&self) -> usize

Get the current step count

Trait Implementations

impl<A, F> LearningRateScheduler<A> for CustomScheduler<A, F>

where A: Float + Debug + ScalarOperand, F: FnMut(usize) -> A,

fn get_learning_rate(&self) -> A

Get the learning rate at the current step

fn step(&mut self) -> A

Update the scheduler state and return the new learning rate

fn reset(&mut self)

Reset the scheduler state

fn step_with_metric(&mut self, metric: A) -> A

Update the scheduler state using an observed metric and return the new learning rate

fn apply_to<D, O>(&self, optimizer: &mut O)

where D: Dimension, O: Optimizer<A, D>, Self: Sized,

Apply the scheduler to an optimizer