Struct prophet::learn_config::LearnConfig

pub struct LearnConfig {
    pub act_fn: ActivationFn<f32>,
    // some fields omitted
}

Represents a configuration that directly controls the parameters of learning for a neural net.

A system that tries to optimize the way a neural net learns over the duration of a training session could make a good use of this structure as an interface to the underlying parameters.

Fields

act_fn: ActivationFn<f32>

the activation function used during training and predicting

Methods

impl LearnConfig

fn new(learn_rate: f32, learn_momentum: f32, act_fn: ActivationFn<f32>) -> Self

Creates a new instance of LearnConfig.

Panics

If learn_rate and learn_momentum ∉ (0, 1).

fn learn_rate(&self) -> f32

Returns the current learn rate.

fn learn_momentum(&self) -> f32

Returns the current learn rate.

fn update_learn_rate(&mut self, new_learn_rate: f32)

Sets the learn rate of this configuration to the given value.

Panics

If the given learn rate is ∉ (0, 1].

fn update_learn_momentum(&mut self, new_learn_momentum: f32)

Sets the learn momentum of this configuration to the given value.

Panics

If the given learn momentum ∉ [0, 1].

Trait Implementations

impl Copy for LearnConfig

impl Clone for LearnConfig

fn clone(&self) -> LearnConfig

Returns a copy of the value.

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

Performs copy-assignment from source.

impl Debug for LearnConfig

fn fmt(&self, __arg_0: &mut Formatter) -> Result

Formats the value using the given formatter.

impl Display for LearnConfig

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

Formats the value using the given formatter.