2
$\begingroup$

I want to run a simulation using qasm with a realistic noise model from a backend, specifying a coupling_map as well as an initial_layout. I do that via a QuantumInstance which in turn is passed to an algorithm. Now, two questions regarding that qubit layout:

  1. I know that the transpiler may change the mapping of logical to physical qubits. Furthermore, I pass a circuit to the algorithm as an argument and the algorithm then adds an ancilla qubit to the circuit. Is there a way for me to verify the final logical to physical qubit mapping after transpiling?

  2. Could that automatic adding of an ancilla cause problems when using QuantumInstance with a specific initial_layout?

$\endgroup$

1 Answer 1

1
$\begingroup$

If you want to get the final qubit layout (regarding to your first question), you can look at the circuit data. Here is an example:

circuit = QuantumCircuit(4, 4)

for i in range(4):
    circuit.x(i)
circuit.measure([0,1,2,3], [0,1,2,3])

    
Circuit_Transpile = transpile(circuit, provider.get_backend('ibmq_16_melbourne') , 
                                initial_layout =None, optimization_level=3)

for quantum_register in Circuit_Transpile.data[:circuit.num_qubits]:
    print(quantum_register[1])

What you will see as output is something like:

[[Qubit(QuantumRegister(15, 'q'), 0)],
 [Qubit(QuantumRegister(15, 'q'), 9)],
 [Qubit(QuantumRegister(15, 'q'), 13)],
 [Qubit(QuantumRegister(15, 'q'), 14)]]

The number on the right correspond to the qubit number on the device of your interest. There are other ways to do this as well.

$\endgroup$
3
  • $\begingroup$ Thanks for your answer. Could you elaborate on the alternatives for doing this? The above has two caveats; It does not show the ordering of logical qubits (e.g. [Qubit(QuantumRegister(15, 'q1'), 9)]) and, furthermore. gives one such line for every gate in the circuit, making it highly repetitive for longer circuits. I am looking for something like a dictionary logical --> physical qubits, e.g., {'q0': 0, 'q1': 9, 'q2': 13, 'q3': 14}. $\endgroup$
    – Durd3nT
    Commented May 8, 2021 at 10:09
  • $\begingroup$ The other answer I was thinking of is extracting the qasm code... print(Circuit_Transpile.qasm()) but that does not give you what you want... $\endgroup$
    – KAJ226
    Commented May 8, 2021 at 15:51
  • $\begingroup$ @KAJ226 How do I know which logical qubit is the physical qubit on the right being mapped to? Also, what are the other ways to do this? $\endgroup$
    – bisarch
    Commented Jan 11, 2023 at 18:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.