def init_params(cls, params, algo_input):
"""Constructor from params."""
feature_dimension = get_feature_dimension(algo_input.training_dataset)
fea_map_params = params.get(Pluggable.SECTION_KEY_FEATURE_MAP)
fea_map_params['feature_dimension'] = feature_dimension
feature_map = get_pluggable_class(
PluggableType.FEATURE_MAP,
fea_map_params['name']).init_params(params)
multiclass_extension = None
multiclass_extension_params = params.get(
Pluggable.SECTION_KEY_MULTICLASS_EXT)
if multiclass_extension_params is not None:
multiclass_extension_params['params'] = [feature_map]
multiclass_extension_params['estimator_cls'] = _QSVM_Estimator
multiclass_extension = \
get_pluggable_class(PluggableType.MULTICLASS_EXTENSION,
multiclass_extension_params['name']).init_params(params)
logger.info("Multiclass classifier based on %s",
multiclass_extension_params['name'])
return cls(feature_map, algo_input.training_dataset,
algo_input.test_dataset, algo_input.datapoints,
multiclass_extension)
def init_params(cls, params, algo_input):
"""Constructor from params."""
num_qubits = get_feature_dimension(algo_input.training_dataset)
fea_map_params = params.get(QuantumAlgorithm.SECTION_KEY_FEATURE_MAP)
fea_map_params['num_qubits'] = num_qubits
feature_map = get_pluggable_class(
PluggableType.FEATURE_MAP,
fea_map_params['name']).init_params(fea_map_params)
multiclass_extension = None
multiclass_extension_params = params.get(
QuantumAlgorithm.SECTION_KEY_MULTICLASS_EXTENSION, None)
if multiclass_extension_params is not None:
multiclass_extension_params['params'] = [feature_map]
multiclass_extension_params[
'estimator_cls'] = _QSVM_Kernel_Estimator
multiclass_extension = get_pluggable_class(
PluggableType.MULTICLASS_EXTENSION,
multiclass_extension_params['name']).init_params(
multiclass_extension_params)
logger.info("Multiclass classifier based on {}".format(
multiclass_extension_params['name']))
return cls(feature_map, algo_input.training_dataset,
algo_input.test_dataset, algo_input.datapoints,
multiclass_extension)
)
n = 2 # dimension of each data point
sample_Total, training_input, test_input, class_labels = wine(training_size=50,
test_size=6,
n=n,
plot_data=False)
temp = [test_input[k] for k in test_input]
total_array = np.concatenate(temp)
try:
aqua_globals.random_seed = 10598
backend = BasicAer.get_backend('qasm_simulator')
feature_map = ZZFeatureMap(
feature_dimension=get_feature_dimension(training_input),
reps=2,
entanglement='linear')
svm = QSVM(feature_map,
training_input,
test_input,
total_array,
multiclass_extension=AllPairs())
quantum_instance = QuantumInstance(
backend,
shots=1024,
seed_simulator=aqua_globals.random_seed,
seed_transpiler=aqua_globals.random_seed)
result = svm.run(quantum_instance)
for k, v in result.items():
