Hi, I recently try to employ the device noise model from IBMQ into the variational quantum circuit, but it failed. Here is the toy code.

```
import pennylane as qml
from pennylane.optimize import GradientDescentOptimizer
from pennylane import numpy as np
import qiskit
import qiskit_aer.noise as noise
def circuit(param, n_qubit):
for j in range(n_qubit):
qml.RZ(param[j, 0], wires=j)
qml.RY(param[j, 1], wires=j)
qml.RZ(param[j, 2], wires=j)
# CNOT
for j in range(0, n_qubit, 2):
if j+1 < n_qubit:
qml.CNOT(wires=[j, j+1])
for j in range(1, n_qubit, 2):
if j+1 < n_qubit:
qml.CNOT(wires=[j, j+1])
return qml.expval(qml.PauliZ(n_qubit-1))
if __name__ == '__main__':
n_qubit = 4
qiskit.IBMQ.load_account()
provider = qiskit.IBMQ.providers(group='open')[0]
backend = provider.get_backend('ibmq_quito')
noise_model = noise.NoiseModel.from_backend(backend)
dev = qml.device('qiskit.aer', wires=n_qubit, noise_model=noise_model)
qnode = qml.QNode(circuit, dev)
opt = GradientDescentOptimizer(0.1)
param = 2*np.pi*np.random.random((n_qubit, 3))
param = opt.step(lambda p: qnode(p, n_qubit), param)
```

The error information is as follows.

```
File "/usr/lib/python3.10/contextlib.py", line 79, in inner
return func(*args, **kwds)
File "/home/xxx/.local/lib/python3.10/site-packages/pennylane_qiskit/qiskit_device.py", line 467, in batch_execute
res = self.statistics(circuit)
File "/home/xxx/.local/lib/python3.10/site-packages/pennylane/_qubit_device.py", line 738, in statistics
if obs.return_type is Expectation:
AttributeError: 'RZ' object has no attribute 'return_type'
```

The version of installed python package is

```
python=3.10.6
pennylane=0.26.0
qiskit=0.39.2
```

By the way, I can run the same code several months ago successfully. So I wonder whether there is a new style to implement device noise model in updated pennylane.

Thanks in advance!