Skip to main content

PlateauLR

Struct PlateauLR 

Source
pub struct PlateauLR { /* private fields */ }
Available on crate feature alloc only.
Expand description

Reduce learning rate when the loss plateaus.

Monitors current_loss and reduces the rate by factor whenever the loss has not improved for patience consecutive steps. “Improved” means the new loss is strictly less than the best-seen loss.

The rate is floored at min_lr to prevent it from vanishing entirely.

§Example

use irithyll::ensemble::lr_schedule::{LRScheduler, PlateauLR};

let mut sched = PlateauLR::new(0.1, 0.5, 3, 0.001);

// Improving loss -- rate stays at 0.1.
assert!(crate::math::abs((sched.learning_rate(0, 1.0) - 0.1)) < 1e-12);
assert!(crate::math::abs((sched.learning_rate(1, 0.9) - 0.1)) < 1e-12);

// Stagnating loss for patience=3 steps.
assert!(crate::math::abs((sched.learning_rate(2, 0.95) - 0.1)) < 1e-12);
assert!(crate::math::abs((sched.learning_rate(3, 0.95) - 0.1)) < 1e-12);
assert!(crate::math::abs((sched.learning_rate(4, 0.95) - 0.1)) < 1e-12);

// Patience exhausted -- rate drops to 0.1 * 0.5 = 0.05.
assert!(crate::math::abs((sched.learning_rate(5, 0.95) - 0.05)) < 1e-12);

Implementations§

Source§

impl PlateauLR

Source

pub fn new(initial_lr: f64, factor: f64, patience: u64, min_lr: f64) -> Self

Create a plateau-aware scheduler.

§Arguments
  • initial_lr – Starting learning rate.
  • factor – Multiplicative reduction factor (e.g., 0.5 halves the rate).
  • patience – Number of non-improving steps to tolerate before reducing.
  • min_lr – Floor below which the rate will not be reduced.

Trait Implementations§

Source§

impl Clone for PlateauLR

Source§

fn clone(&self) -> PlateauLR

Returns a duplicate of the value. Read more
1.0.0 · Source§

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

Performs copy-assignment from source. Read more
Source§

impl Debug for PlateauLR

Source§

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

Formats the value using the given formatter. Read more
Source§

impl LRScheduler for PlateauLR

Source§

fn learning_rate(&mut self, _step: u64, current_loss: f64) -> f64

Return the learning rate for the given step and current_loss. Read more
Source§

fn reset(&mut self)

Reset the scheduler to its initial state. Read more

Auto Trait Implementations§

Blanket Implementations§

Source§

impl<T> Any for T
where T: 'static + ?Sized,

Source§

fn type_id(&self) -> TypeId

Gets the TypeId of self. Read more
Source§

impl<T> Borrow<T> for T
where T: ?Sized,

Source§

fn borrow(&self) -> &T

Immutably borrows from an owned value. Read more
Source§

impl<T> BorrowMut<T> for T
where T: ?Sized,

Source§

fn borrow_mut(&mut self) -> &mut T

Mutably borrows from an owned value. Read more
Source§

impl<T> CloneToUninit for T
where T: Clone,

Source§

unsafe fn clone_to_uninit(&self, dest: *mut u8)

🔬This is a nightly-only experimental API. (clone_to_uninit)
Performs copy-assignment from self to dest. Read more
Source§

impl<T> From<T> for T

Source§

fn from(t: T) -> T

Returns the argument unchanged.

Source§

impl<T, U> Into<U> for T
where U: From<T>,

Source§

fn into(self) -> U

Calls U::from(self).

That is, this conversion is whatever the implementation of From<T> for U chooses to do.

Source§

impl<T> ToOwned for T
where T: Clone,

Source§

type Owned = T

The resulting type after obtaining ownership.
Source§

fn to_owned(&self) -> T

Creates owned data from borrowed data, usually by cloning. Read more
Source§

fn clone_into(&self, target: &mut T)

Uses borrowed data to replace owned data, usually by cloning. Read more
Source§

impl<T, U> TryFrom<U> for T
where U: Into<T>,

Source§

type Error = Infallible

The type returned in the event of a conversion error.
Source§

fn try_from(value: U) -> Result<T, <T as TryFrom<U>>::Error>

Performs the conversion.
Source§

impl<T, U> TryInto<U> for T
where U: TryFrom<T>,

Source§

type Error = <U as TryFrom<T>>::Error

The type returned in the event of a conversion error.
Source§

fn try_into(self) -> Result<U, <U as TryFrom<T>>::Error>

Performs the conversion.