Name-sorted parameters in `qiskit.opflow`

What is the expected enhancement?

Binding parameters to a parameterized Opflow objects requires passing a dictionary with {parameter_instance: value} pairs. More often than not, this is used in the following setting

my_circuit = # ...
expression = # some opflow expression using the circuit

param_dict = dict(zip(my_circuit.parameters, my_values))
bound_expression = expression.bind_parameters(param_dict)

which is a bit cumbersome as we either have to carry along the parameters we’re interested in or have to write an extra line defining the param_dict.

Instead if would be nice to allow binding the parameters using only an array. Since we introduced a global parameter order in #5759, which says that parameters are sorted by name, that would be easy to implement. Using the same ordering convention as for the circuits, we could simply write

my_circuit = # ...
expression = # some opflow expression using the circuit

bound_expression = expression.bind_parameters(values)

Since Opflow objects already implement the parameter attribute, the steps to implement this could for instance be

1. ensure the parameter attribute returns a ParameterView object with parameters in the right order
2. if an array is passed into bind_parameters, construct the parameter dictionary for the user using the parameter order from parameters

I’m labelling this as “good first issue”, but some previous experience with qiskit.opflow would certainly be helpful!

[Cryoris]

Hi @pradkrish, for the QuantumCircuit we’re already able to pass my_values. But here’s a slightly modified example where we are currently not able to pass an array but would like to

from qiskit.circuit import Parameter, QuantumCircuit
from qiskit.opflow import StateFn, Z

phase = Parameter("phase")
x = Parameter("x")
y = Parameter("y")
z = Parameter("z")

qc = QuantumCircuit(1, global_phase=phase)
qc.rx(x, 0)
qc.rz(z, 0)
qc.ry(y, 0)

a = Parameter("a")

my_values = [-1, 0.1, 10, 20, 30]  # a, phase, x, y ,z

expectation = StateFn(a * Z, is_measurement=True) @ StateFn(qc)
expectation.assign_parameters(my_values)  # not possible right now!

[pradkrish]

Hello, is this issue still open? I am new to this project, happy to work on this with some guidance.