Struct Adam 

pub struct Adam<T: RealScalar = f64, B: LinearAlgebra<T> = NalgebraProvider> { /* private fields */ }

Scalar- and linear-algebra-generic Adam optimizer.

Trait Implementations

impl<T, B, P, U, E> Algorithm<P, GradientStatus<T, B>, U, E> for Adam<T, B>

where T: RealScalar, B: LinearAlgebra<T>, P: Gradient<T, B, U, E>,

type Summary = MinimizationSummary<T, B>

A type which holds a summary of the algorithm’s ending state.

type Config = AdamConfig<T, B>

The configuration struct for the algorithm.

type Init = Vector<T, B>

The initialization payload for a single run.

fn initialize( &mut self, problem: &P, status: &mut GradientStatus<T, B>, args: &U, init: &Self::Init, config: &Self::Config, ) -> Result<(), E>

Any setup work done before the main steps of the algorithm should be done here.

fn step( &mut self, current_step: usize, problem: &P, status: &mut GradientStatus<T, B>, args: &U, config: &Self::Config, ) -> Result<(), E>

The main “step” of an algorithm, which is repeated until termination conditions are met or the max number of steps have been taken.

fn summarize( &self, _current_step: usize, _problem: &P, status: &GradientStatus<T, B>, _args: &U, init: &Self::Init, config: &Self::Config, ) -> Result<Self::Summary, E>

Generates a new Algorithm::Summary from the current state of the Algorithm, which can be displayed or used elsewhere.

fn reset(&mut self)

Reset the algorithm to its initial state.

fn default_callbacks() -> Callbacks<Self, P, GradientStatus<T, B>, U, E, Self::Config>

Provides a set of reasonable default callbacks specific to this Algorithm.

fn postprocessing( &mut self, problem: &P, status: &mut S, args: &U, config: &Self::Config, ) -> Result<(), E>

Runs any steps needed by the Algorithm after termination or convergence. This will run regardless of whether the Algorithm converged.

fn process<I, C>( &mut self, problem: &P, args: &U, init: I, config: Self::Config, callbacks: C, ) -> Result<Self::Summary, E>

where I: Into<Self::Init>, C: Into<Callbacks<Self, P, S, U, E, Self::Config>>, Self: Sized,

Process the given problem using this Algorithm.

fn process_with_default_callbacks<I>( &mut self, problem: &P, user_data: &U, init: I, config: Self::Config, ) -> Result<Self::Summary, E>

where I: Into<Self::Init>, Self: Sized,

Process the given problem using this Algorithm and the algorithm’s default callbacks.

fn process_default<I>( &mut self, problem: &P, user_data: &U, init: I, ) -> Result<Self::Summary, E>

where I: Into<Self::Init>, Self: Sized, Self::Config: Default,

Process the given problem using this Algorithm with default config and default callbacks.

impl<T: Clone + RealScalar, B: Clone + LinearAlgebra<T>> Clone for Adam<T, B>

fn clone(&self) -> Adam<T, B>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<T: Debug + RealScalar, B: Debug + LinearAlgebra<T>> Debug for Adam<T, B>

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

Formats the value using the given formatter.

impl<T, B> Default for Adam<T, B>

where T: RealScalar, B: LinearAlgebra<T>,

fn default() -> Self

Returns the “default value” for a type.

impl<T, B, P, U, E> Terminator<Adam<T, B>, P, GradientStatus<T, B>, U, E, AdamConfig<T, B>> for AdamEMATerminator<T>

where T: RealScalar, B: LinearAlgebra<T>, P: Gradient<T, B, U, E>,

fn check_for_termination( &mut self, _current_step: usize, algorithm: &mut Adam<T, B>, _problem: &P, status: &mut GradientStatus<T, B>, _args: &U, _config: &AdamConfig<T, B>, ) -> ControlFlow<()>

A termination check which is called on each step of an Algorithm.