opennn::ForwardPropagation Struct Reference

Workspace holding the activations of every layer during a forward pass. More...

#include <forward_propagation.h>

Public Member Functions
	ForwardPropagation (const Index=0, NeuralNetwork *=nullptr)
	Constructs a workspace for the given batch size and network.
void	set (const Index=0, NeuralNetwork *=nullptr)
	Reconfigures the workspace for a new batch size or network; reuses allocations when possible.
TensorView	get_last_trainable_layer_outputs () const
	Returns the output tensor of the last trainable layer.
TensorView	get_outputs () const
	Returns the final output tensor of the network.
void	print () const
	Prints a human-readable summary of the workspace contents.

Public Attributes
Index	batch_size = 0
NeuralNetwork *	neural_network = nullptr
Buffer	data
vector< vector< vector< TensorView > > >	views

Detailed Description

Workspace holding the activations of every layer during a forward pass.

Constructor & Destructor Documentation

ForwardPropagation()

opennn::ForwardPropagation::ForwardPropagation	(	const Index	= 0,
		NeuralNetwork *	= nullptr )

Constructs a workspace for the given batch size and network.

Parameters

batch_size	Maximum number of samples per forward pass.
neural_network	Network whose layer specs drive buffer sizing (non-owning).

Member Function Documentation

get_last_trainable_layer_outputs()

TensorView opennn::ForwardPropagation::get_last_trainable_layer_outputs

(

)

const

Returns the output tensor of the last trainable layer.

get_outputs()

TensorView opennn::ForwardPropagation::get_outputs

(

)

const

Returns the final output tensor of the network.

print()

void opennn::ForwardPropagation::print

(

)

const

Prints a human-readable summary of the workspace contents.

set()

void opennn::ForwardPropagation::set	(	const Index	= 0,
		NeuralNetwork *	= nullptr )

Reconfigures the workspace for a new batch size or network; reuses allocations when possible.

Parameters

batch_size	Maximum number of samples per forward pass.
neural_network	Network whose layer specs drive buffer sizing (non-owning).

Member Data Documentation

batch_size

Index opennn::ForwardPropagation::batch_size = 0

data

Buffer opennn::ForwardPropagation::data

neural_network

NeuralNetwork* opennn::ForwardPropagation::neural_network = nullptr

views

vector<vector<vector<TensorView> > > opennn::ForwardPropagation::views