Models

qrobot.models module class diagram:

The Model abstract class

This class is the parent class which defines the general features a perception model has to implement. It is used to define custom models as a child classes:

from qrobot.models import Model

class myModel(Model):
    (...)

class qrobot.models.Model(n, tau)

Model is an abstract class which embeds the general features needed in a model for QL perception.

Parameters:

• n (int) – Model’s dimension (must be greater than 0, 1 is a scalar)

• tau (int) – Number of samples of the temporal window (must be greater than 0)

n

Model’s dimension.

Type:

int

tau

Number of samples of the temporal window.

Type:

int

circ

Quantum circuit which implements the model.

Type:

qiskit.QuantumCircuit

clear() → None

Re-initialize the model with an empty circuit.

abstract decode() → str

Exploits the information encoded in the qubit.

abstract encode(scalar_input, dim) → None

Encodes the scalar input in the correspondent qubit.

Example

To encode a sequence of input vectors, given tau and n:

for t in range(model.tau): # loop through time
    for dim in range(model.n): # loop through dimensions
        model.encode(sequence[t][dim], dim)

get_density_matrix() → ndarray

Returns the simulated density matrix of the model.

Returns:

Model’s density matrix.

Return type:

numpy.ndarray

get_statevector() → ndarray

Returns the simulated state vector of the model.

Returns:

Model’s state vector.

Return type:

numpy.ndarray

measure(shots=1, backend: Backend | None = None) → Dict[str, int]

Measures the qubits using a IBMQ backend

Parameters:

• shots (int) – Number of times to repeat the measurement shot

• backend (qiskit Backend) – Quantum backend for the execution (if None, AER simulator is chosen as default)

Returns:

State occurrences counts in the form {“state”: count}

Return type:

dict

plot_state_mat() → None

Plots the state and density matrix of the quantum system (just the real parts).

Example

To plot a perfectly balanced superposition of states:

model = Model(n, tau) # change Model with the desired child class

for t in range(0,model.tau): # loop through time
    for dim in range(model.n): # loop through dimensions
        model.encode(.5, dim)

model.plot_state_mat()

Raises:

OverflowError – If the dimension of the model is 6 or greater, plotting fails due to the high number of basis states.

print_circuit() → None

Prints the quantum circuit on which the model is implemented.

abstract query(target_vector) → None

Changes the basis of the quantum system choosing target_vector as the basis state |00…0>.

The AngularModel class

This is a custom Model provided by the package. This model encodes the perceptual information in the angle of the qubits’ Bloch sphere representations.

class qrobot.models.AngularModel(n, tau)

AngularModel is a kind of Model which encodes the perceptual information in the angle of the qubits’ Bloch sphere representations

decode()

The decoding for the AngularModel is a single measurement.

Returns:

The string label corresponding to the decoded state

Return type:

str

encode(scalar_input, dim)

Encodes the scalar input in the correspondent qubit

Parameters:

• input (float) – The scalar input for a certain dimension, must be a number between 0 and 1 inclusive.

• dim (int) – The model’s dimension which the input belongs (values between 0 and n-1)

Returns:

The rotation angle applied to the qubit.

Return type:

float

query(target_vector)

Changes the basis of the quantum system choosing target_vector as the basis state |00…0>

Parameters:

target_vector (list) – The target_vector state, it must be a list containing n floats (between 0 and 1 inclusive).

The LinearModel class

This is a custom model derived from AngularModel, which corrects it with a nonlinear encoding. By means of its nonlinear encoding, this model provides a linear decoding (a notebook is provided in order to illustrate the difference between this and the angular one).

class qrobot.models.LinearModel(n, tau)

LinearModel corrects the AngularModel the encoding, allowing a linear decoding for single-event sequencies (tau=1).

Warning

The current implementation of LinearModel provides a linear dependency between encoded input and decoding probabilities only for tau = 1 (or for tau > 1 if and only if the input is always the same one). For a time-varying sequencee (tau > 1) the results obtained are again nonlinear, and similar to the one of AngularModel (see this notebook for more information)

encode(scalar_input, dim)

Encodes the scalar input in the correspondent qubit

Parameters:

• scalar_input (float) – The scalar input for a certain dimension, must be a number between 0 and 1 inclusive.

• dim (int) – The model’s dimension which the scalar input belongs.

Returns:

The rotation angle applied to the qubit.

Return type:

float