cldnn::concatenation Struct Reference

#include <concatenation.hpp>

Inheritance diagram for cldnn::concatenation:

Collaboration diagram for cldnn::concatenation:

Public Types
enum	concatenation_axis { along_b = cldnn_concatenation_along_b, along_f = cldnn_concatenation_along_f, along_x = cldnn_concatenation_along_x, along_y = cldnn_concatenation_along_y }
typedef cldnn_concatenation_desc	dto

Public Member Functions
	concatenation (const primitive_id &id, const std::vector< primitive_id > &input, const concatenation_axis axis, const padding &output_padding=padding())
	concatenation (const dto *dto)
	Constructs a copy from C API (depth_concatenate)
Public Member Functions inherited from cldnn::primitive_base< concatenation, cldnn_concatenation_desc >
const cldnn_primitive_desc *	get_dto () const override
	Returns pointer to a C API primitive descriptor casted to cldnn_primitive_desc.
Public Member Functions inherited from cldnn::primitive
	primitive (const primitive_type_id &type, const primitive_id &id, const std::vector< primitive_id > &input, const padding &output_padding=padding())
	primitive (const cldnn_primitive_desc *dto)
	Constructs a copy from basic C API cldnn_primitive_desc.
std::vector< std::reference_wrapper< primitive_id > >	dependecies ()
	Returns references to all primitive ids on which this primitive depends - inputs, weights, biases, etc.
std::vector< primitive_id >	dependecies () const
	Returns copy of all primitive ids on which this primitive depends - inputs, weights, biases, etc.
	operator primitive_id () const
	Implicit conversion to primiitive id.

Static Public Member Functions
static primitive_type_id	type_id ()

Public Attributes
concatenation_axis	axis
	Dimension along which concatenation should take place.
Public Attributes inherited from cldnn::primitive
const primitive_type_id	type
	Primitive's type id.
const primitive_id	id
	Primitive's id.
fixed_size_vector_ref	input
	List of ids of input primitives.
padding	output_padding
	Requested output padding.

Additional Inherited Members
Protected Member Functions inherited from cldnn::primitive_base< concatenation, cldnn_concatenation_desc >
	primitive_base (const primitive_id &id, const std::vector< primitive_id > &input, const padding &output_padding=padding())
	primitive_base (const cldnn_concatenation_desc *dto)
Protected Member Functions inherited from cldnn::primitive
virtual std::vector< std::reference_wrapper< const primitive_id > >	get_dependencies () const
Protected Attributes inherited from cldnn::primitive
primitive_id_arr	_input

Detailed Description

Concatenation is used to concatenate multiple sources into one destination along specified dimension.

• all other dimensions (except the one along which concatenation take place) must have the same value in each source.
• order of arguments in primitive creation has impact on order of feature maps in output primitive.

Alogrithm:

int outputIdx = 0
for(i : input)
{
    for(f : i.features)
    {
        output[outputIdx] = f
        outputIdx += 1
    }
}

Where:

• input : data structure holding all source inputs for this primitive
• output : data structure holding output data for this primitive
• i.features : number of features in currently processed input
• outputIdx : index of destination feature

Definition at line 53 of file concatenation.hpp.

Constructor & Destructor Documentation

concatenation()

cldnn::concatenation::concatenation ( const primitive_id &  id, const std::vector< primitive_id > &  input, const concatenation_axis  axis, const padding &  output_padding = padding()  )

inline

• Constructs concatenation primitive.

  Parameters

  id	This primitive id.
  input	Vector of input primitives ids.
  axis	Selected dimension for concatenation.

Definition at line 69 of file concatenation.hpp.

---

The documentation for this struct was generated from the following file:

• C:/ML/OpenSource/clDNN/api/CPP/concatenation.hpp