How to calculate the expected value of a quantum operator in Qiskit (after deprecations)

Question

Background

In 2022 Qiskit released a very nice tutorial detailing how the expectation value of an operator could be found using CircuitOp. As of 2024, these techniques (and others that I have found on the quantum computing stack exchange) are unusable as they are deprecated. The migration guide suggests using Estimators. As a test, I tried writing some code to calculate the exact (non-simulated) expected value of $\langle0|H|0\rangle$ which, I believe, should be equal to $\dfrac{1}{\sqrt{2}}$. Instead, I got $0.9999999999999999$. I include my code below (as it is quite short).

The Code

from qiskit import QuantumCircuit
from qiskit.primitives import Estimator
testCircuit = QuantumCircuit(1)
testCircuit.h(0)
estimator = Estimator()
result = estimator.run(testCircuit, 'I').result()
print(result.values[0])

The Question

What is the best practice (in 2024) for calculating the exact expected value of any observable specified as a matrix and/or circuit using Qiskit?

Answer 1

This is how you do it with the latest Qiskit Packages:

Step 1: Expectation Value of an Operator

$$E = \langle \psi |O|\psi\rangle$$

Make the circuit you want to measure the $\psi$ with

from qiskit import QuantumCircuit, QuantumRegister, ClassicalRegister
circuit = QuantumCircuit(2)
circuit.x(0)
circuit.x(1)
#op = CircuitOp(circuit)
circuit.draw(style='iqx',output='mpl')

Step 2: The Operator

Make the operator you want to measure the expectation value of:

from qiskit.quantum_info import Pauli, SparsePauliOp
X = Pauli('X')
Y = Pauli('Y')
Z = Pauli('Z')
I = Pauli('I')
operator = SparsePauliOp(["II", "IZ", "ZI","ZZ","XX" ], coeffs = [-1.052373245772859, 0.39793742484318045,-0.39793742484318045,
                                                                  -0.01128010425623538,0.18093119978423156 ])
#op = (-1.052373245772859 * I^I) + (0.39793742484318045 * I^Z) +
(-0.39793742484318045 * Z^I) + (-0.01128010425623538 * Z^Z) +
#(0.18093119978423156 * X^X)
print(operator)

Step 3: The State

The state you want to measure against:

from qiskit.quantum_info import SparsePauliOp, Statevector
from qiskit.primitives import Estimator
psi = QuantumCircuit(2)
psi.x(0)
psi.x(1)
estimator = Estimator()
#psi = Statevector(psi)
expectation_value = estimator.run(psi, operator).result().values.real

The result

print(expectation_value)

This follows the youtube tutorial you linked, but with the latest qiskit packages, you can change as per you need.