def __init__(self,
training_dataset,
test_dataset=None,
datapoints=None,
gamma=None,
multiclass_extension=None):
self.validate(locals())
super().__init__()
if training_dataset is None:
raise AquaError('Training dataset must be provided.')
is_multiclass = get_num_classes(training_dataset) > 2
if is_multiclass:
if multiclass_extension is None:
raise AquaError(
'Dataset has more than two classes. A multiclass extension must be provided.'
)
else:
if multiclass_extension is not None:
logger.warning(
"Dataset has just two classes. Supplied multiclass extension will be ignored"
)
if multiclass_extension is None:
svm_instance = _SVM_Classical_Binary(training_dataset,
test_dataset, datapoints,
gamma)
else:
svm_instance = _SVM_Classical_Multiclass(training_dataset,
test_dataset, datapoints,
gamma,
multiclass_extension)
self.instance = svm_instance
def __init__(
self,
training_dataset: Dict[str, np.ndarray],
test_dataset: Optional[Dict[str, np.ndarray]] = None,
datapoints: Optional[np.ndarray] = None,
gamma: Optional[int] = None,
multiclass_extension: Optional[MulticlassExtension] = None
) -> None:
# pylint: disable=line-too-long
"""
Args:
training_dataset: training dataset.
test_dataset: testing dataset.
datapoints: prediction dataset.
gamma: Used as input for sklearn rbf_kernel internally. See
`sklearn.metrics.pairwise.rbf_kernel
<https://scikit-learn.org/stable/modules/generated/sklearn.metrics.pairwise.rbf_kernel.html>`_
for more information about gamma.
multiclass_extension: if number of classes > 2 then
a multiclass scheme is needed.
Raises:
AquaError: If using binary classifier where num classes >= 3
"""
super().__init__()
if training_dataset is None:
raise AquaError('Training dataset must be provided.')
is_multiclass = get_num_classes(training_dataset) > 2
if is_multiclass:
if multiclass_extension is None:
raise AquaError('Dataset has more than two classes. '
'A multiclass extension must be provided.')
else:
if multiclass_extension is not None:
logger.warning(
"Dataset has just two classes. Supplied multiclass "
"extension will be ignored")
if multiclass_extension is None:
svm_instance = _SVM_Classical_Binary(training_dataset,
test_dataset, datapoints,
gamma)
else:
svm_instance = _SVM_Classical_Multiclass(training_dataset,
test_dataset, datapoints,
gamma,
multiclass_extension)
self.instance = svm_instance
