QSVM+Image classification+ DCP errors

[ZainabASA]

. I use QSVM to classify image dataset … I don't understand this error, How to fix it.
error display

---

DCPError Traceback (most recent call last)
in ()
5 svm.random_seed = 10598
6 quantum_instance = QuantumInstance(backend, shots=1024, seed_simulator=10598, seed_transpiler=10598)
----> 7 result = svm.run(quantum_instance)

9 frames
/usr/local/lib/python3.6/dist-packages/cvxpy/reductions/solvers/intermediate_chain.py in construct_intermediate_chain(problem, candidates, gp)
94 append += ("\nHowever, the problem does follow DQCP rules. "
95 "Consider calling solve() with qcp=True.")
---> 96 raise DCPError("Problem does not follow DCP rules. Specifically:\n" + append)
97
98 elif gp and not problem.is_dgp():

DCPError: Problem does not follow DCP rules. Specifically:

[ArthiBoss]

Have you tried using statevector_simulator, in most case qasm_simulator shows such error but satevector_simulator wont show DCP error and the prediction accuracy also good.

[praveentn]

@ArthiBoss thank you!

Changing the backend from

backend = BasicAer.get_backend('qasm_simulator')

to

backend = BasicAer.get_backend('statevector_simulator')

worked for me!