Struct leaf::layers::activation::tanh::TanH

pub struct TanH;

TanH Activation Layer

Trait Implementations

impl Debug for TanH

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

Formats the value using the given formatter.

impl Clone for TanH

fn clone(&self) -> TanH

Returns a copy of the value.

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

Performs copy-assignment from source.

impl<B: IBackend + Tanh<f32>> ILayer<B> for TanH

fn exact_num_output_blobs(&self) -> Option<usize>

Returns the exact number of output blobs required by the layer, or None if no exact number is required.

fn exact_num_input_blobs(&self) -> Option<usize>

Returns the exact number of input blobs required by the layer, or None if no exact number is required.

fn reshape(
    &mut self,
    backend: Rc<B>,
    input_data: &mut Vec<ArcLock<SharedTensor<f32>>>,
    input_gradient: &mut Vec<ArcLock<SharedTensor<f32>>>,
    weights_data: &mut Vec<ArcLock<SharedTensor<f32>>>,
    weights_gradient: &mut Vec<ArcLock<SharedTensor<f32>>>,
    output_data: &mut Vec<ArcLock<SharedTensor<f32>>>,
    output_gradient: &mut Vec<ArcLock<SharedTensor<f32>>>
)

Adjust to shapes of the output blobs to fit the shapes of the input blobs.

fn init(&mut self, backend: Rc<B>)

Initialize the layer for computation.

fn resize_shared_workspace(
    &mut self,
    backend: Rc<B>,
    workspace: Option<ArcLock<SharedTensor<u8>>>
) -> Option<ArcLock<SharedTensor<u8>>>

Adjust size of shared workspace.

fn forward(
    &self,
    backend: &B,
    input_data: &[ArcLock<SharedTensor<f32>>],
    weights_data: &[ArcLock<SharedTensor<f32>>],
    output_data: &mut [ArcLock<SharedTensor<f32>>]
)

Compute the [feedforward][1] layer output using the provided Backend. [1]: https://en.wikipedia.org/wiki/Feedforward_neural_network

fn backward_input(
    &self,
    backend: &B,
    weights_data: &[ArcLock<SharedTensor<f32>>],
    output_data: &[ArcLock<SharedTensor<f32>>],
    output_gradients: &[ArcLock<SharedTensor<f32>>],
    input_data: &[ArcLock<SharedTensor<f32>>],
    input_gradients: &mut [ArcLock<SharedTensor<f32>>]
)

Compute the [backpropagation][1] input gradient using the provided backend. [1]: https://en.wikipedia.org/wiki/Backpropagation

fn backward_parameters(
    &self,
    backend: &B,
    output_data: &[ArcLock<SharedTensor<f32>>],
    output_gradients: &[ArcLock<SharedTensor<f32>>],
    input_data: &[ArcLock<SharedTensor<f32>>],
    weights_gradients: &mut [ArcLock<SharedTensor<f32>>]
)

Compute the [backpropagation][1] parameters gradient using the provided backend. [1]: https://en.wikipedia.org/wiki/Backpropagation

fn sync(
    &self,
    backend: &B,
    input_data: &mut [ArcLock<SharedTensor<f32>>],
    input_gradients: &mut [ArcLock<SharedTensor<f32>>],
    weights_data: &mut [ArcLock<SharedTensor<f32>>],
    weights_gradients: &mut [ArcLock<SharedTensor<f32>>],
    output_data: &mut Vec<ArcLock<SharedTensor<f32>>>,
    output_gradients: &mut Vec<ArcLock<SharedTensor<f32>>>
)

Synchronize the blobs before doing a forward or backward operation.

fn auto_output_blobs(&self) -> bool

Return whether "anonymous" output blobs are created automatically for the layer.

fn min_output_blobs(&self) -> usize

Returns the minimum number of output blobs required by the layer, or 0 if no minimum number is required.

fn auto_weight_blobs(&self) -> bool

Return whether weight blobs are created automatically for the layer.

fn allow_force_backward(&self, input_id: usize) -> bool

Return whether to allow force_backward for a given input blob index.

fn sync_native(&self) -> bool

Return wether a simple native backend should be used to [sync][1] instead of the default backend. [1]: #method.sync

fn compute_in_place(&self) -> bool

Return wether the computations of a layer should be done in-place (the output will be written where the input was read from).

fn is_container(&self) -> bool

Return wether the layer is a container.

fn loss_weight(&self, output_id: usize) -> Option<f32>

Return the associated loss weight for a given output blob index.

fn inputs_data(&self) -> Option<Vec<ArcLock<SharedTensor<f32>>>>

Return the input tensors of the layer.

fn inputs_gradients(&self) -> Option<Vec<ArcLock<SharedTensor<f32>>>>

Return the gradients of the input tensors of the layer.

fn outputs_data(&self) -> Option<Vec<ArcLock<SharedTensor<f32>>>>

Return the output tensors of the layer.

fn outputs_gradients(&self) -> Option<Vec<ArcLock<SharedTensor<f32>>>>

Return the gradients of the output tensors of the layer.

fn learnable_weights(&self) -> Option<Vec<ArcLock<SharedTensor<f32>>>>

Return the learnable weights inside the layer.

fn learnable_weights_gradients(&self) -> Option<Vec<ArcLock<SharedTensor<f32>>>>

Return the gradients for the learnable weights inside the layer.

fn learnable_weights_names(&self) -> Option<Vec<String>>

Return the names of the learnable weights inside the layer.

fn learnable_weights_lr(&self) -> Option<Vec<Option<f32>>>

Return the learning rates for the learnable weights inside the layer.

impl<B: IBackend + Tanh<f32>> ComputeOutput<f32, B> for TanH

fn compute_output(
    &self,
    backend: &B,
    _weights: &[&SharedTensor<f32>],
    input_data: &[&SharedTensor<f32>],
    output_data: &mut [&mut SharedTensor<f32>]
)

Compute output for given input and write them into output_data.

impl<B: IBackend + Tanh<f32>> ComputeInputGradient<f32, B> for TanH

fn compute_input_gradient(
    &self,
    backend: &B,
    weights_data: &[&SharedTensor<f32>],
    output_data: &[&SharedTensor<f32>],
    output_gradients: &[&SharedTensor<f32>],
    input_data: &[&SharedTensor<f32>],
    input_gradients: &mut [&mut SharedTensor<f32>]
)

Compute gradients with respect to the inputs and write them into input_gradients.

impl<B: IBackend + Tanh<f32>> ComputeParametersGradient<f32, B> for TanH

fn compute_parameters_gradient(
    &self,
    backend: &B,
    output_data: &[&SharedTensor<T>],
    output_gradients: &[&SharedTensor<T>],
    input_data: &[&SharedTensor<T>],
    parameters_gradients: &mut [&mut SharedTensor<T>]
)

Compute gradients with respect to the parameters and write them into parameters_gradients.