dstallenberg / in-phase

QEP: Phase Estimation

In-Phase

A quantum phase estimation tool

This README introduces In-Phase: a tool to implement Quantum Phase Estimation (QPE) for Quantum Inspire, a quantum circuit simulator. In-Phase creates a quantum circuit to estimate the phase of any given unitary matrix. This tool takes two limitations of real quantum circuits into account: phase errors caused by quantum gates and the fixed topology of mapped qubits on a real quantum circuit.

Collaborators:

• Beer de Zoeten
• Daniel Vlaardingerbroek
• Dimitri Stallenberg
• Mio Poortvliet

Installing from source

Clone the project

git clone https://github.com/dstallenberg/QEP-PhaseEstimation.git

Open the project in your favorite IDE

Gather requirements

pip install .

Run main.py

Usage

The most important function to use is as follows:

qasm_code = generate_qasm_code(nancillas, qubits, unitary)

This function has several possible arguments:

• The first argument is the amount of nancillas these dictate the accuracy of the output phase.
• The second argument is the amount of qubits that your unitary uses.
• The third and most important argument is the unitary which is also the only required argument. This unitary can be one of the following:
  • A string containing a singular gate and one argument if the gate requires it.

  unitary = 'Rz 0.5'

  Note: Do not add anything to the format: "[gate] [argument]". If there is no argument the format is "[gate]".

  • A string containing a circuit in qasm code prefixed by the QASM key word.

  unitary = '''
  QASM
  H q[0]
  X q[1]
  CNOT q[0], q[1]
  '''

  Note: Make sure to not add any additional spaces or enters to the qasm code, otherwise the tool won't be able to optimise the resulting qasm code.

  • A unitary numpy matrix where the dimensions are given by d = 2**q where q is the amount of qubits.

  unitary = np.array([[1, 0], 
                      [0, -1]])

  Note: Make sure that the dimensions follow the given formula and make sure that the matrix is unitary.

Look at the examples in the examples directory for a more explicit explanation.

Features

In-Phase also contains several other functionalities, most of which can be found in the examples directory

Error Estimation

Example:

from src.quantum_phase_estimation.util_functions import error_estimate

desired_bit_accuracy = 5
minimum_chance_of_success = 0.5

nancillas, p_succes = error_estimate(desired_bit_accuracy, minimum_chance_of_success)

Expected outcome:

nancillas == 7
p_succes == 0.75

Finding amount of qubits

Example:

from src.quantum_phase_estimation.util_functions import find_qubits_from_unitary

unitary = 'Z'
nancillas = 3

qubits, extra_empty_bits = find_qubits_from_unitary(unitary, nancillas)

Expected outcome:

qubits == 1
extra_empty_bits == 0

Note: extra_empty_bits is always zero unless the optional argument 'topology' is given

Optimizing cQASM

Example:

from src.qasm_optimizer.optimizer import optimize

qasm_code = '''
X q[0]
X q[0]
H q[0]
'''
nancillas = 3
qubits = 1
extra_empty_bits = 0

qasm_code = optimize(qasm_code, nancillas, qubits, extra_empty_bits)

Expected outcome:

qasm_code == '''
H q[0]
'''

Topology mapping

Example:

from src.qasm_topology_mapper.mapping import map_to_topology

topology = [[0, 1],
            [0, 2]
            [1, 3]
            [2, 3]]
qasm_code = '''
CNOT q[0], q[3]
'''

qasm_code = map_to_topology(topology, qasm_code)

Expected outcome:

qasm_code == '''
SWAP q[0], q[1]
CNOT q[1], q[3]
SWAP q[0], q[1]
'''

Error introduction

Example:

from src.qasm_error_introducer.error_introducer import introduce_error

qasm_code = '''
Rz q[0], 0.1
'''
mu = 0
sigma = 0.2

qasm_code = introduce_error(qasm_code, mu, sigma)

Expected outcome:

qasm_code == '''
Rz q[0], 0.3
Rz q[0], 0.1
'''

cQASM to ProjectQ conversion

Example:

from src.qasm_to_projectq.converter import qasm_to_projectq

qasm_code = '''
X q[0]
'''

projecq_code = qasm_to_projectq(qasm_code)

Expected outcome:

projecq_code == '''
X | q0
'''

Sources

We were inspired by quantum inspire and used the provided sdk to apply our generated qasm code to a quantum backend:

• https://www.quantum-inspire.com/

We used the following repository to convert unitary matrices to qasm code:

• https://github.com/fedimser/quantum_decomp/