NN Class Reference

A class containing basic methods for the decomposition process. More...

#include <NN.h>

Public Member Functions
void	create_randomized_parameters (int num_of_parameters, int qbit_num, int levels, Matrix_real &parameters, matrix_base< int8_t > &nontrivial_adaptive_layers)
	Call to construct random parameter, with limited number of non-trivial adaptive layers. More...
void	get_nn_chanels (const Matrix &Umtx, const int &target_qbit, Matrix_real &chanels)
	call retrieve the channels for the neural network associated with a single unitary More...
void	get_nn_chanels (int qbit_num, const Matrix &Umtx, Matrix_real &chanels)
	call retrieve the channels for the neural network associated with a single unitary More...
void	get_nn_chanels (int qbit_num, int levels, Matrix_real &chanels, matrix_base< int8_t > &nontrivial_adaptive_layers)
	call retrieve the channels for the neural network associated with a single, randomly generated unitary More...
void	get_nn_chanels (int qbit_num, int levels, int samples_num, Matrix_real &chanels, matrix_base< int8_t > &nontrivial_adaptive_layers)
	call retrieve the channels for the neural network associated with a single, randomly generated unitary More...
void	get_nn_chanels_from_kernel (Matrix &kernel_up, Matrix &kernel_down, Matrix_real &chanels)
	call retrieve the channels for the neural network associated with a single 2x2 kernel More...
	NN ()
	Nullary constructor of the class. More...
	NN (std::vector< matrix_base< int >> topology_in)
	Constructor of the class. More...

Public Attributes
int	tt
	number of gate blocks used in one shot of the optimization process More...

Protected Attributes
std::mt19937	gen
	Standard mersenne_twister_engine seeded with rd() More...
int	num_threads
	Store the number of OpenMP threads. (During the calculations OpenMP multithreading is turned off.) More...
std::random_device	rd
	Will be used to obtain a seed for the random number engine. More...
std::vector< matrix_base< int > >	topology
	connectivity between the wubits More...

Detailed Description

A class containing basic methods for the decomposition process.

Definition at line 61 of file NN.h.

Constructor & Destructor Documentation

NN() [1/2]

NN::NN

(

)

Nullary constructor of the class.

Returns

An instance of the class

Definition at line 34 of file NN.cpp.

NN() [2/2]

NN::NN

(

std::vector< matrix_base< int >>

topology_in

)

Constructor of the class.

Parameters

The	list of conenctions between the qubits

Returns

An instance of the class

Definition at line 56 of file NN.cpp.

Member Function Documentation

create_randomized_parameters()

void NN::create_randomized_parameters	(	int	num_of_parameters,
		int	qbit_num,
		int	levels,
		Matrix_real &	parameters,
		matrix_base< int8_t > &	nontrivial_adaptive_layers
	)

Call to construct random parameter, with limited number of non-trivial adaptive layers.

Parameters

num_of_parameters

The number of parameters

Definition at line 82 of file NN.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

get_nn_chanels() [1/4]

void NN::get_nn_chanels	(	const Matrix &	Umtx,
		const int &	target_qbit,
		Matrix_real &	chanels
	)

call retrieve the channels for the neural network associated with a single unitary

Parameters

Umtx	A unitary of dimension dim x dim, where dim is a power of 2.
target_qbit	The target qubit for which the chanels are calculated
chanels	output array containing the chanels prepared for th eneural network. The array has dimensions [ dim/2, dim/2, 4 ] (dimension "4" stands for theta_up, phi, theta_down , lambda)

Definition at line 222 of file NN.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

get_nn_chanels() [2/4]

void NN::get_nn_chanels	(	int	qbit_num,
		const Matrix &	Umtx,
		Matrix_real &	chanels
	)

call retrieve the channels for the neural network associated with a single unitary

Parameters

Umtx	A unitary of dimension dim x dim, where dim is a power of 2.
chanels	output array containing the chanels prepared for th eneural network. The array has dimensions [ dim/2, dim/2, 4 ] (dimension "4" stands for theta_up, phi, theta_down , lambda)
qbit_num	Th number of qubites
Umtx	A unitary of dimension dim x dim, where dim is a power of 2.
chanels	output array containing the chanels prepared for th eneural network. The array has dimensions [ dim/2, dim/2, 4 ] (dimension "4" stands for theta_up, phi, theta_down , lambda)

Definition at line 306 of file NN.cpp.

Here is the call graph for this function:

get_nn_chanels() [3/4]

void NN::get_nn_chanels	(	int	qbit_num,
		int	levels,
		Matrix_real &	chanels,
		matrix_base< int8_t > &	nontrivial_adaptive_layers
	)

call retrieve the channels for the neural network associated with a single, randomly generated unitary

Parameters

qbit_num	The number of qubits
levels	The number of adaptive levels to be randomly constructed
chanles	output argument to return with an array containing the chanels prepared for the neural network. The array has dimensions [ dim/2, dim/2, 4 ] (dimension "4" stands for theta_up, phi, theta_down , lambda)
parameters	output argument of the randomly created parameters

Definition at line 402 of file NN.cpp.

Here is the call graph for this function:

get_nn_chanels() [4/4]

void NN::get_nn_chanels	(	int	qbit_num,
		int	levels,
		int	samples_num,
		Matrix_real &	chanels,
		matrix_base< int8_t > &	nontrivial_adaptive_layers
	)

call retrieve the channels for the neural network associated with a single, randomly generated unitary

Parameters

qbit_num	The number of qubits
levels	The number of adaptive levels to be randomly constructed
samples_num	The number of samples
chanles	output argument to return with an array containing the chanels prepared for the neural network. The array has dimensions [ samples_num, dim/2, dim/2, 4 ] (dimension "4" stands for theta_up, phi, theta_down , lambda)
parameters	output argument of the randomly created parameters

Definition at line 463 of file NN.cpp.

Here is the call graph for this function:

get_nn_chanels_from_kernel()

void NN::get_nn_chanels_from_kernel	(	Matrix &	kernel_up,
		Matrix &	kernel_down,
		Matrix_real &	chanels
	)

call retrieve the channels for the neural network associated with a single 2x2 kernel

Returns

return with an 1x4 array containing the chanels prepared for the neural network. (dimension 4 stands for theta_up, phi, theta_down , lambda)

Definition at line 160 of file NN.cpp.

Here is the caller graph for this function:

Member Data Documentation

gen

std::mt19937 NN::gen

protected

Standard mersenne_twister_engine seeded with rd()

Definition at line 75 of file NN.h.

num_threads

int NN::num_threads

protected

Store the number of OpenMP threads. (During the calculations OpenMP multithreading is turned off.)

Definition at line 79 of file NN.h.

rd

std::random_device NN::rd

protected

Will be used to obtain a seed for the random number engine.

Definition at line 73 of file NN.h.

topology

std::vector<matrix_base<int> > NN::topology

protected

connectivity between the wubits

Definition at line 77 of file NN.h.

tt

int NN::tt

number of gate blocks used in one shot of the optimization process

Definition at line 67 of file NN.h.

---

The documentation for this class was generated from the following files:

• NN.h
• NN.cpp