PennyLane-Qiskit Plugin

Release:

0.40.0-dev

The PennyLane-Qiskit plugin integrates the Qiskit quantum computing framework with PennyLane’s quantum machine learning capabilities.

PennyLane is a cross-platform Python library for quantum machine learning, automatic differentiation, and optimization of hybrid quantum-classical computations.

Qiskit is an open-source framework for quantum computing.

Once the PennyLane-Qiskit plugin is installed, the Qiskit devices can be accessed straightaway in PennyLane, without the need to import new packages.

Devices

The following devices are available:

'qiskit.aer'

Qiskit's staple simulator with great features such as noise models.

'qiskit.basicsim'

A simple local Python simulator running the Qiskit ``BasicSimulator``.

'qiskit.remote'

Allows integration with any Qiskit backend.

For example, the 'qiskit.aer' device with two wires is called like this:

import pennylane as qml
dev = qml.device('qiskit.aer', wires=2)

Backends

Qiskit devices have different backends, which define the actual simulator or hardware used by the device. A backend instance should be initalized and passed to the device.

Different simulator backends are optimized for different purposes. To change what backend is used, a simulator backend can be defined as follows:

from qiskit_aer import UnitarySimulator

dev = qml.device('qiskit.aer', wires=<num_qubits>, backend=UnitarySimulator())

Note

For 'qiskit.aer', PennyLane chooses the aer_simulator as the default backend if no backend is specified. For more details on the aer_simulator, including available backend options, see Qiskit Aer Simulator documentation.

To access a real device, we can use the 'qiskit.remote' device. A real hardware backend can be defined as follows:

from qiskit_ibm_runtime import QiskitRuntimeService

QiskitRuntimeService.save_account(channel="ibm_quantum", token="<IQP_TOKEN>")

# To access saved credentials for the IBM quantum channel and select an instance
service = QiskitRuntimeService(channel="ibm_quantum", instance="my_hub/my_group/my_project")
backend = service.least_busy(operational=True, simulator=False, min_num_qubits=<num_qubits>)

# passing a string in backend would result in an error
dev = qml.device('qiskit.remote', wires=<num_qubits>, backend=backend)

Tutorials

Check out these demos to see the PennyLane-Qiskit plugin in action:

Ensemble classification with Forest and Qiskit devices

Quantum volume

Using PennyLane with IBM's quantum devices and Qiskit