Struct burn::nn::Dropout

pub struct Dropout { /* private fields */ }

Set at random some elements of the input tensor to zero during training.

This is an effective regularization technique as describe in the paper Improving neural networks by preventing co-adaptation of feature detectors.

The input is also scaled during training to 1 / (1-p).

Implementations

impl Dropout

pub fn new(config: &DropoutConfig) -> Self

Trait Implementations

impl Clone for Dropout

fn clone(&self) -> Dropout

Returns a copy of the value.

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

Performs copy-assignment from source.

impl Debug for Dropout

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

Formats the value using the given formatter.

impl<B: Backend, const D: usize> Forward<Tensor<B, D>, Tensor<B, D>> for Dropout

fn forward(&self, input: Tensor<B, D>) -> Tensor<B, D>

The forward method mapping the input to the output.