Migration to Qiskit Primitives

[robotastray]

Summary

This PR aims to resolve issue #83.

Details and comments

File: test_braket_backend.py

Function test_vqe():

• used from qiskit.algorithms.minimum_eigensolvers import VQE, VQEResult (Note: "The old VQE class has been superseded by the qiskit.algorithms.minimum_eigensolvers.VQE class.")

• substituted QuantumInstance with BackendEstimator

• used estimator as argument for VQE

• substituted qiskit.opflow with qiskit.quantum_info

• used SparsePauliOp to represent Pauli operators

File: test_adapter.py

• substituted qiskit.opflow with qiskit.quantum_info
• used SparsePauliOp to represent Pauli operators

File: test_braket_job.py

No changes

File: test_braket_provider.py

No changes

File: docs/tutorials/0_tutorial_qiskit-braket-provider_overview.ipynb

• used VQE from qiskit.algorithms.minimum_eigensolvers
• substituted QuantumInstance with BackendEstimator
• used SparsePauiliOp as alternative to qiskit.opflow operators

File: docs/tutorials/1_tutorial_vqe.ipynb

• used VQE from qiskit.algorithms.minimum_eigensolvers
• substituted QuantumInstance with BackendEstimator
• used SparsePauiliOp as alternative to qiskit.opflow operators

File: docs/tutorials/2_tutorial_hybrid_jobs_byoc.ipynb

• re-run ! cat ./data/2_hybrid_jobs/job_script.py which displayed the content of the new job_script.py

File: docs/tutorials/3_tutorial_minimum_eigen_optimizer.ipynb

• substituted QuantumInstance with BackendSampler
• used COBYLA to be used as optimiser for QAOA
• used QAOA, NumPyMinimumEigensolver from new qiskit.algorithms.minimum_eigensolvers
• added options={ "seed_simulator": algorithm_globals.random_seed, "seed_transpiler": algorithm_globals.random_seed, }, as options in BackendSampler()

File: /data/2_hybrid_jobs/job_script.py

• used VQE from qiskit.algorithms.minimum_eigensolvers
• substituted QuantumInstance with BackendEstimator
• used SparsePauiliOp as alternative to qiskit.opflow operators

[robotastray]

@IceKhan13 I have noticed something that might be unrelated to this issue. There is a problem in test_adaper.py to be specific function test_standard_gate_decomp(). I have made no changes to this function so far, and this error must have been present before this PR started. I have been working with Python 3.11. The tests fails for 3.10 version

======================================================================
ERROR: test_standard_gate_decomp (test_adapter.TestAdapter.test_standard_gate_decomp)
Tests adapter decomposition of all standard gates to forms that can be translated
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/Users/robotastray/Desktop/github/qiskit-braket-provider/tests/providers/test_adapter.py", line 202, in test_standard_gate_decomp
    qiskit_circuit = qiskit_circuit.assign_parameters(parameter_bindings)
                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/robotastray/Desktop/github/qiskit-braket-provider/venv/lib/python3.11/site-packages/qiskit/circuit/quantumcircuit.py", line 2762, in assign_parameters
    raise CircuitError(
qiskit.circuit.exceptions.CircuitError: 'Cannot bind parameters (0) not present in the circuit.'

----------------------------------------------------------------------

For more info, you can see here

[robotastray]

The error with test_standard_gate_decomp() in test_adapter.py is due to the following standard_gates:

XXMinusYYGate(Parameter("ϴ")),
 XXPlusYYGate(Parameter("ϴ")),

Note this is unrelated to this PR, it something I stumbled upon.
If one comments these gates out of the code there are no issues. I have commented them from the array for now. Let me know how you wish to proceed regarding this.

[IceKhan13]

Hey @robotastray ! Thank for this PR!

hmm, something is wrong with this gate specifically. You are right test failing in main branch too https://github.com/qiskit-community/qiskit-braket-provider/actions/runs/5768152751/job/15638813727 and only for python 3.10, but not for 3.7 🤔

Give me couple of days, I will try to figure out what is going on, so we can remove commented gates from this PR.

As a heads-up PR looks great! :)

[IceKhan13]

@robotastray fixed failing tests in #118. Make sure to merge changes from main, uncomment tests and we can move with PR review!

[robotastray]

Hello @IceKhan13,
Thank you for your message and help. I have completed the steps now.

[IceKhan13]

Sweet!

Thank you @robotastray !