Hands-On Quantum Machine Learning with Python/

...

/

Multiple Results

Multiple Results

Learn how we can implement multiple functions like prepare_group and find_relatives.

We'll cover the following...

Passengers may not only have a single sibling, spouse, parent, or child. They can have many. For instance, the Carter family, where the father is Mr. William Ernest Carter, consists of the parental couple and two children.

Before we see what happens, let’s further structure our code. We define the prepare_group function. It takes a PassengerId and returns the corresponding passenger and a list of potential relatives.

Prepare the search

Press + to interact
def prepare_group(passengerId):
    current_passenger = train[train["PassengerId"].eq(passengerId)]
    last_name = current_passenger.Name.to_string(index=False).split(',')[0]
    train[train["Name"].str.contains(last_name)]
    ticket = current_passenger["Ticket"].to_string(index=False)
    passengerId = current_passenger["PassengerId"]
    group = train[
        train["PassengerId"].ne(passengerId) & (
        train["Name"].str.contains(last_name) |
        train["Ticket"].eq(ticket)
    )]
    return (current_passenger, group)

We also define the find_relatives function. It takes the passenger and the group and runs the search algorithm.

Run the search for a passenger’s relatives

Press + to interact
def find_relatives(current_passenger, group):
    q_pass = QuantumRegister(QUBITS, name='passengers')
    q_rules = QuantumRegister(RULES, name='rules')
    qc = QuantumCircuit(q_pass, q_rules)
    # put passenger qubits into superposition
    qc.h(q_pass)
    rounds(pi*sqrt(2**QUBITS)/4, qc, current_passenger, group, q_pass, q_rules)
    results = execute(qc,Aer.get_backend('statevector_simulator')).result()
    return plot_histogram(results.get_counts())

Let’s see what happens if we run our algorithm to search for the relatives of Mr. William Ernest Carter.

Search for the relatives of Mr. William Ernest Carter

Press + to interact
QUBITS=3
(current_passenger, group) = prepare_group(391)
find_relatives(current_passenger, group)

When we look at the table of potential relatives, we see that the algorithm found all but the actual relatives.

PassengerId

Survived 

Pclass 

Name 

Sex 

Age 

. . . 

Embarked 

249

250

0

2

Carter, Rev. Ernest Courtenay

male

54.0

. . . 

S

235 

336

1

1

Carter, Miss. Lucile Polk

female

14.0

. . . 

S

763 

764

1

1

Carter, Mrs. William Ernest (Lucile Polk) 

female

36.0

. . . 

S

.

.

.

.

.

.


.

.

.


.

.

.


.

.

.


.

.

.


.

.

.


.

.

.


.

.

.


The problem is the number of iterations. When we have more than one correct relative, calculating the required iterations is a little different. If we search for ...