State preparation

View on GitHub

State preparation is a very important part of many quantum algorithms. How a state is prepared can have a huge effect on the circuit characteristics. Here we will prepare the same state twice with different approximations to see how that can affect the generated circuit.

from classiq import *

amplitudes = [
    0.2450344728207077,
    -0.1700813505441722,
    0.16769736623372838,
    -0.0239052232765222,
    0.34393605643711794,
    -0.2966929669473228,
    -0.2839167245810649,
    0.3305828537849893,
    -0.32376029368963405,
    0.25704647614347453,
    0.1531631416826692,
    -0.07516264976664182,
    -0.314744872342388,
    -0.24992215398592088,
    0.3526620654512262,
    -0.015039183255468514,
]

Create a circuit to prepare the amplitudes

Use the prepare_amplitudes() function to encode the amplitudes above into a quantum circuit. Use the defined bound in your prepare_amplitudes() function.

bound = 0.3


@qfunc
def main(
    # target: Output[QArray]
):
    # Your code here
    pass


qprog = synthesize(create_model(main))
show(qprog)

Opening: https://platform.classiq.io/circuit/0cf6b636-9189-4280-9e13-e25517c8e1eb?version=0.41.0.dev39%2B79c8fd0855

Change the bound for 0.001 and look at the effect.

Why is there a big difference in circuit characteristics

The full solution for your reference

bound = 0.3


@qfunc
def main(target: Output[QArray]):
    prepare_amplitudes(amplitudes=amplitudes, out=target, bound=bound)


qprog = synthesize(create_model(main))
show(qprog)

Opening: https://platform.classiq.io/circuit/b4e9649f-2578-4dbb-bc4c-de943266395a?version=0.41.0.dev39%2B79c8fd0855