I'm working on QSVM to know the difference between SVM and QSVM. How much better quantum machine learning can do when compared to classical machine learning algorithms?
Part of my code:
backend = BasicAer.get_backend('qasm_simulator', hub=None) quantum_instance = QuantumInstance(backend, shots=1024, seed=seed, seed_transpiler=seed) result = qsvm.run(quantum_instance) print("test accuracy: ", result['testing_accuracy'])
With the above code, I can get the accuracy of the test.
I'm trying to know whether I can draw a decision boundary and classify results from QSVM just like how classical SVM does.
print(result.keys()) dict_keys(['svm', 'kernel_matrix_testing', 'test_success_ratio', 'kernel_matrix_training', 'testing_accuracy']) print (result['svm'].keys()) dict_keys(['bias', 'alphas', 'support_vectors', 'yin'])
Given only these are the keys that result has, how can I look at QSVM results into two classes graphically? I could get the prediction results though! my new_test_data= 2d array to classify between these two classes : good=1/ bad=0
predictions = qsvm.predict(new_test_data,quantum_instance) print(predictions) [1 0 0 0 0 0 0 0 1 1 0 1 1 0 1 0]
Given I don't see how get decision function and be able to plot decision boundary and margins? Like this as we do in classical SVM:
Any help in clarifying how to read the results to plot a graph like this from running QSVM would be much appreciated.