Your First Simulation

The first circuit sampled measurement outcomes. This tutorial asks for more: the exact state, then an exact expectation value, and then how MENOS planned the job.

1. The exact statevector

Drop the measurements and ask for the statevector with want_statevector=True. Because nothing is sampled, shots is 0.

import numpy as np
from qiskit import QuantumCircuit

from menos.qiskit import MENOSBackend

qc = QuantumCircuit(2)
qc.h(0)
qc.cx(0, 1)

result = MENOSBackend().run(qc, shots=0, want_statevector=True).result()
state = np.asarray(result.get_statevector())
print(state)

assert np.allclose(state, [2**-0.5, 0, 0, 2**-0.5])

The amplitudes are exact (no sampling noise). Only the dense engine can produce a statevector, so this request is always planned onto it.

2. An expectation value

An expectation value such as <ZZ> can be computed without building the state at all. Use MENOSEstimator with a SparsePauliOp:

from qiskit import QuantumCircuit
from qiskit.quantum_info import SparsePauliOp

from menos.qiskit import MENOSEstimator

qc = QuantumCircuit(2)
qc.h(0)
qc.cx(0, 1)

value = float(MENOSEstimator().run([(qc, SparsePauliOp("ZZ"))]).result()[0].data.evs)
print(value)
assert abs(value - 1.0) < 1e-12

For a Bell state <ZZ> is exactly 1. See Expectation Values.

3. How was it planned?

The Qiskit layer returns a standard Qiskit result and does not expose the plan. The native API does. Here is the same Bell circuit, measured, run natively:

from menos import Circuit, Opcode, RunOptions, run

circuit = Circuit(2, 2)
circuit.append(Opcode.H, [0])
circuit.append(Opcode.CX, [0, 1])
circuit.append_measure(0, 0)
circuit.append_measure(1, 1)

options = RunOptions()
options.shots = 100
options.seed = 1

report = run(circuit, options).explain()
print(report)
assert "strategy" in report

The output looks like this:

MENOS EXECUTION REPORT
circuit    qubits=2 clbits=2 operations=4 depth=3 global_phase=0
light_cone qubits 2->2 operations 4->4
candidate 0strategy=direct_dense feasible=true winner=false peak_bytes=64 work_units=32
candidate 1strategy=direct_stabilizer feasible=true winner=true peak_bytes=40 work_units=16
candidate 2strategy=direct_mps feasible=true winner=false peak_bytes=128 work_units=34
strategy   direct_stabilizer (exact)
region 0   qubits=[0,1] engine=stabilizer backend=cpu
estimated  state_bytes=40 work_units=16
actual     state_bytes=40 execute_s=0.000046 sample_s=0.000000

Three strategies were feasible. The stabilizer engine won because it had the fewest work units (16, against 32 for dense and 34 for MPS). The actual line is what was measured after running.

A formatting detail

Labels are padded to 11 characters, so a label that is 11 characters or longer runs straight into its value: candidate 0strategy=…. That is the format as shipped, not a typo in this page. See Plan Explanation.

Where next

• Plan Explanation: read every line of that report.
• Qiskit Integration: what the backend supports.
• Examples: runnable programs for every engine.