Qrack::QNeuron Class Reference

#include <qneuron.hpp>

Public Member Functions
	QNeuron (QInterfacePtr reg, const std::vector< bitLenInt > &inputIndcs, bitLenInt outputIndx, QNeuronActivationFn activationFn=Sigmoid, real1_f alpha=ONE_R1_F, real1_f tol=FP_NORM_EPSILON)
	"QNeuron" is a "Quantum neuron" or "quantum perceptron" class that can learn and predict in superposition. More...
	QNeuron (const QNeuron &toCopy)
	Create a new QNeuron which is an exact duplicate of another, including its learned state. More...
QNeuron &	operator= (const QNeuron &toCopy)
void	SetAlpha (real1_f a)
	Set the "alpha" sharpness parameter of this QNeuron. More...
real1_f	GetAlpha ()
	Get the "alpha" sharpness parameter of this QNeuron. More...
void	SetActivationFn (QNeuronActivationFn f)
	Sets activation function enum. More...
QNeuronActivationFn	GetActivationFn ()
	Get activation function enum. More...
void	SetAngles (real1 *nAngles)
	Set the angles of this QNeuron. More...
void	GetAngles (real1 *oAngles)
	Get the angles of this QNeuron. More...
bitLenInt	GetInputCount ()
bitCapInt	GetInputPower ()
real1_f	Predict (bool expected=true, bool resetInit=true)
	Predict a binary classification. More...
real1_f	Unpredict (bool expected=true)
	"Uncompute" the Predict() method More...
real1_f	LearnCycle (bool expected=true)
void	Learn (real1_f eta, bool expected=true, bool resetInit=true)
	Perform one learning iteration, training all parameters. More...
void	LearnPermutation (real1_f eta, bool expected=true, bool resetInit=true)
	Perform one learning iteration, measuring the entire QInterface and training the resulting permutation. More...

Protected Member Functions
real1_f	LearnInternal (bool expected, real1_f eta, bitCapInt perm, real1_f startProb)

Static Protected Member Functions
static real1_f	applyRelu (real1_f angle)
static real1_f	negApplyRelu (real1_f angle)
static real1_f	applyGelu (real1_f angle)
static real1_f	negApplyGelu (real1_f angle)
static real1_f	applyAlpha (real1_f angle, real1_f alpha)
static real1_f	applyLeakyRelu (real1_f angle, real1_f alpha)
static real1_f	clampAngle (real1_f angle)

Protected Attributes
bitCapIntOcl	inputPower
bitLenInt	outputIndex
QNeuronActivationFn	activationFn
real1_f	alpha
real1_f	tolerance
std::vector< bitLenInt >	inputIndices
std::unique_ptr< real1[]>	angles
QInterfacePtr	qReg

Constructor & Destructor Documentation

QNeuron() [1/2]

Qrack::QNeuron::QNeuron ( QInterfacePtr  reg, const std::vector< bitLenInt > &  inputIndcs, bitLenInt  outputIndx, QNeuronActivationFn  activationFn = Sigmoid, real1_f  alpha = ONE_R1_F, real1_f  tol = FP_NORM_EPSILON  )

inline

"QNeuron" is a "Quantum neuron" or "quantum perceptron" class that can learn and predict in superposition.

This is a simple "quantum neuron" or "quantum perceptron" class, for use of the Qrack library for machine learning. See https://arxiv.org/abs/quant-ph/0410066 (and https://arxiv.org/abs/1711.11240) for the basis of this class' theoretical concept.

An untrained QNeuron (with all 0 variational parameters) will forward all inputs to 1/sqrt(2) * (|0> + |1>). The variational parameters are Pauli Y-axis rotation angles divided by 2 * Pi (such that a learning parameter of 0.5 will train from a default output of 0.5/0.5 probability to either 1.0 or 0.0 on one training input).

QNeuron() [2/2]

Qrack::QNeuron::QNeuron ( const QNeuron &  toCopy )

inline

Create a new QNeuron which is an exact duplicate of another, including its learned state.

Member Function Documentation

applyAlpha()

static real1_f Qrack::QNeuron::applyAlpha ( real1_f  angle, real1_f  alpha  )

inlinestaticprotected

applyGelu()

static real1_f Qrack::QNeuron::applyGelu ( real1_f  angle )

inlinestaticprotected

applyLeakyRelu()

static real1_f Qrack::QNeuron::applyLeakyRelu ( real1_f  angle, real1_f  alpha  )

inlinestaticprotected

applyRelu()

static real1_f Qrack::QNeuron::applyRelu ( real1_f  angle )

inlinestaticprotected

clampAngle()

static real1_f Qrack::QNeuron::clampAngle ( real1_f  angle )

inlinestaticprotected

GetActivationFn()

QNeuronActivationFn Qrack::QNeuron::GetActivationFn ( )

inline

Get activation function enum.

GetAlpha()

real1_f Qrack::QNeuron::GetAlpha ( )

inline

Get the "alpha" sharpness parameter of this QNeuron.

GetAngles()

void Qrack::QNeuron::GetAngles ( real1 *  oAngles )

inline

Get the angles of this QNeuron.

GetInputCount()

bitLenInt Qrack::QNeuron::GetInputCount ( )

inline

GetInputPower()

bitCapInt Qrack::QNeuron::GetInputPower ( )

inline

Learn()

void Qrack::QNeuron::Learn ( real1_f  eta, bool  expected = true, bool  resetInit = true  )

inline

Perform one learning iteration, training all parameters.

Inputs must be already loaded into "qReg" before calling this method. "expected" is the true binary output category, for training. "eta" is a volatility or "learning rate" parameter with a maximum value of 1.

In the feedback process of learning, default initial conditions forward untrained predictions to 1/sqrt(2) * (|0>

• |1>) for the output bit. If you want to initialize other conditions before "Learn()," set "resetInit" to false.

LearnCycle()

real1_f Qrack::QNeuron::LearnCycle ( bool  expected = true )

inline

LearnInternal()

real1_f Qrack::QNeuron::LearnInternal ( bool  expected, real1_f  eta, bitCapInt  perm, real1_f  startProb  )

inlineprotected

LearnPermutation()

void Qrack::QNeuron::LearnPermutation ( real1_f  eta, bool  expected = true, bool  resetInit = true  )

inline

Perform one learning iteration, measuring the entire QInterface and training the resulting permutation.

Inputs must be already loaded into "qReg" before calling this method. "expected" is the true binary output category, for training. "eta" is a volatility or "learning rate" parameter with a maximum value of 1.

In the feedback process of learning, default initial conditions forward untrained predictions to 1/sqrt(2) * (|0>

• |1>) for the output bit. If you want to initialize other conditions before "LearnPermutation()," set "resetInit" to false.

negApplyGelu()

static real1_f Qrack::QNeuron::negApplyGelu ( real1_f  angle )

inlinestaticprotected

negApplyRelu()

static real1_f Qrack::QNeuron::negApplyRelu ( real1_f  angle )

inlinestaticprotected

operator=()

QNeuron& Qrack::QNeuron::operator= ( const QNeuron &  toCopy )

inline

Predict()

real1_f Qrack::QNeuron::Predict ( bool  expected = true, bool  resetInit = true  )

inline

Predict a binary classification.

Feed-forward from the inputs, loaded in "qReg", to a binary categorical classification. "expected" flips the binary categories, if false. "resetInit," if true, resets the result qubit to 0.5/0.5 |0>/|1> superposition before proceeding to predict.

SetActivationFn()

void Qrack::QNeuron::SetActivationFn ( QNeuronActivationFn  f )

inline

Sets activation function enum.

SetAlpha()

void Qrack::QNeuron::SetAlpha ( real1_f  a )

inline

Set the "alpha" sharpness parameter of this QNeuron.

SetAngles()

void Qrack::QNeuron::SetAngles ( real1 *  nAngles )

inline

Set the angles of this QNeuron.

Unpredict()

real1_f Qrack::QNeuron::Unpredict ( bool  expected = true )

inline

"Uncompute" the Predict() method

Member Data Documentation

activationFn

QNeuronActivationFn Qrack::QNeuron::activationFn

protected

alpha

real1_f Qrack::QNeuron::alpha

protected

angles

std::unique_ptr<real1[]> Qrack::QNeuron::angles

protected

inputIndices

std::vector<bitLenInt> Qrack::QNeuron::inputIndices

protected

inputPower

bitCapIntOcl Qrack::QNeuron::inputPower

protected

outputIndex

bitLenInt Qrack::QNeuron::outputIndex

protected

qReg

QInterfacePtr Qrack::QNeuron::qReg

protected

tolerance

real1_f Qrack::QNeuron::tolerance

protected

---

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

• include/qneuron.hpp