Generating and cutting advanced circuits

cutqc2 has a library of advanced quantum circuits in the cutqc2.library module. The functions in this module generate qiskit.QuantumCircuit objects that can be used as input to the CutCircuit class.

Circuits can be created using the generate_circ function, which takes the following parameters:

• num_qubits: The number of qubits in the circuit.

• depth: The depth of the circuit. This parameter may not apply to all circuit types, and can be set None if not needed for that circuit type.

• circuit_type: The type of circuit to generate. Options include "supremacy", "sycamore", "bv", "qft", "aqft", and others.

• reg_name: The name of the quantum register. Default is "q".

• connected_only: If True (the default), ensures that the circuit only contains gates between connected qubits. This option must be set to True if you expect to cut the circuit using cutqc2.

• seed: A seed for the random number generator to ensure reproducibility. Default None.

A typical way to generate and cut one of these circuit types would be:

from cutqc2.library.utils import generate_circ
from cutqc2.core.cut_circuit import CutCircuit


circuit = generate_circ(
    num_qubits=38,
    depth=20,
    circuit_type="aqft",
    reg_name="q", 
    connected_only=True,
    seed=42
)
cut_circuit = CutCircuit(circuit)
cut_circuit.cut(
    max_subcircuit_width=30,
    max_cuts=10,
    num_subcircuits=[2]
)

Supported Circuit Types

At the time of this writing, the following circuit types can be generated using cutqc2.library.utils.generate_circ:

supremacy

Quantum Supremacy Circuit based on the implementations in https://www.nature.com/articles/s41567-018-0124-x and https://github.com/sboixo/GRCS.)

sycamore

Quantum Supremacy Circuit as found in https://www.nature.com/articles/s41586-019-1666-5

hwea

Hardware efficient ansatz for the QAOA algorithm. Based on the community detection circuit implemented by Francois-Marie Le Régent. This ansatz uses the entangler+rotation block structure like that described in the paper by Nikolaj Moll et al. (http://iopscience.iop.org/article/10.1088/2058-9565/aab822)

bv

Bernstein-Vazirani algorithm circuit.

qft

Circuit to perform the Quantum Fourier Transform (or its inverse) as described in Mike & Ike Chapter 5.

aqft

An approximate QFT that ignores controlled-phase rotations where the angle is beneath a threshold. This is discussed in more detail in https://arxiv.org/abs/quant-ph/9601018 or https://arxiv.org/abs/quant-ph/0403071.

adder

An n-bit ripple-carry adder. Based on the specification given in Cuccaro, Draper, Kutin, Moulton. (https://arxiv.org/abs/quant-ph/0410184v1)

regular

A QAOA-style variational circuit built on a 3-regular random graph with num_qubits nodes.

erdos

A QAOA-style variational circuit (depth P = depth) on an Erdős–Rényi graph over num_qubits nodes.

random

A generic random circuit of given width and depth, with connection degree 0.5 and a fixed number of Hadamards inserted.