def __init__(self, config):
super().__init__(config)
self._class_encoder = SKLabelEncoder()
self._feat_vectorizer = DictVectorizer()
self._feat_selector = self._get_feature_selector()
self._feat_scaler = self._get_feature_scaler()
self._meta_type = None
self._meta_feat_vectorizer = DictVectorizer(sparse=False)
self._base_clfs = {}
self.cv_loss_ = None
self.train_acc_ = None
def setup_model(self, config):
if config.model_settings is None:
selector_type = None
scale_type = None
else:
selector_type = config.model_settings.get('feature_selector')
scale_type = config.model_settings.get('feature_scaler')
self.class_encoder = SKLabelEncoder()
self.feat_vectorizer = DictVectorizer()
self._feat_selector = self._get_feature_selector(selector_type)
self._feat_scaler = self._get_feature_scaler(scale_type)
def __init__(self, config):
if not _is_module_available('torch'):
raise ImportError(
"Install the extra 'torch' library by runnning "
"'pip install mindmeld[torch]' to use pytorch based neural models"
)
super().__init__(config)
self._label_encoder = get_label_encoder(self.config)
self._class_encoder = SKLabelEncoder()
self._query_text_type = None
self._clf = None
def deserialise_encoder(
encoder: acton_pb.Database.LabelEncoder
) -> sklearn.preprocessing.LabelEncoder:
"""Deserialises a LabelEncoder protobuf.
Parameters
----------
encoder
LabelEncoder protobuf.
Returns
-------
sklearn.preprocessing.LabelEncoder
LabelEncoder (or None if no encodings were specified).
"""
encodings = []
for encoding in encoder.encoding:
encodings.append((encoding.class_int, encoding.class_label))
encodings.sort()
encodings = numpy.array([c[1] for c in encodings])
encoder = SKLabelEncoder()
encoder.classes_ = encodings
return encoder