Trait ReluPointwise 

pub trait ReluPointwise<F>: NN<F> {
    // Required methods
    fn relu_pointwise(&self, x: &mut SharedTensor<F>) -> Result<(), Error>;
    fn relu_pointwise_grad(
        &self,
        x: &SharedTensor<F>,
        x_diff: &mut SharedTensor<F>,
    ) -> Result<(), Error>;
}

Provides the functionality for pointwise ReLU operations (overwrites the input with the result of the operation).

Required Methods

fn relu_pointwise(&self, x: &mut SharedTensor<F>) -> Result<(), Error>

Computes the [Rectified linear units][relu] over the input Tensor x. [relu]: https://en.wikipedia.org/wiki/Rectifier_(neural_networks)

Saves the result back to x.

fn relu_pointwise_grad( &self, x: &SharedTensor<F>, x_diff: &mut SharedTensor<F>, ) -> Result<(), Error>

Computes the gradient of [ReLU][relu] over the input Tensor x. [relu]: https://en.wikipedia.org/wiki/Rectifier_(neural_networks)

Saves the result back to x_diff.

Dyn Compatibility

This trait is not dyn compatible.

In older versions of Rust, dyn compatibility was called "object safety", so this trait is not object safe.

Implementations on Foreign Types

impl ReluPointwise<f32> for Backend<Native>

fn relu_pointwise(&self, x: &mut SharedTensor<f32>) -> Result<(), Error>

fn relu_pointwise_grad( &self, x: &SharedTensor<f32>, x_diff: &mut SharedTensor<f32>, ) -> Result<(), Error>

impl ReluPointwise<f64> for Backend<Native>

fn relu_pointwise(&self, x: &mut SharedTensor<f64>) -> Result<(), Error>

fn relu_pointwise_grad( &self, x: &SharedTensor<f64>, x_diff: &mut SharedTensor<f64>, ) -> Result<(), Error>

impl<T> ReluPointwise<T> for Backend<Cuda>

where T: Float + Default + DataTypeInfo,

fn relu_pointwise(&self, x: &mut SharedTensor<T>) -> Result<(), Error>

fn relu_pointwise_grad( &self, x: &SharedTensor<T>, x_diff: &mut SharedTensor<T>, ) -> Result<(), Error>

Implementors