def __init__(
self,
n_neighbors=1,
weights="uniform",
distance="dtw",
distance_params=None,
**kwargs
):
self._distance_params = distance_params
if distance_params is None:
self._distance_params = {}
self.distance = distance
self.distance_params = distance_params
if isinstance(self.distance, str):
distance = distance_factory(metric=self.distance)
super(KNeighborsTimeSeriesClassifier, self).__init__(
n_neighbors=n_neighbors,
algorithm="brute",
metric=distance,
metric_params=None, # Extra distance params handled in _fit
**kwargs
)
BaseClassifier.__init__(self)
self.weights = _check_weights(weights)
# We need to add is-fitted state when inheriting from scikit-learn
self._is_fitted = False
def fit(self, X, y, **kwargs):
"""Wrap fit to call BaseClassifier.fit.
This is a fix to get around the problem with multiple inheritance. The
problem is that if we just override _fit, this class inherits the fit from
the sklearn class BaseTimeSeriesForest. This is the simplest solution,
albeit a little hacky.
"""
return BaseClassifier.fit(self, X=X, y=y, **kwargs)
def predict_proba(self, X, **kwargs):
"""Predict proba wrapper."""
return BaseClassifier.predict_proba(self, X, **kwargs)
def predict(self, X, **kwargs):
"""Predict wrapper."""
return BaseClassifier.predict(self, X, **kwargs)
def fit(self, X, y, **kwargs):
"""Override fit is required to sort out the multiple inheritance."""
return BaseClassifier.fit(self, X, y, **kwargs)
def predict_proba(self, X, **kwargs) -> np.ndarray:
"""Wrap predict_proba to call BaseClassifier.predict_proba."""
return BaseClassifier.predict_proba(self, X=X, **kwargs)
def predict(self, X, **kwargs) -> np.ndarray:
"""Predict wrapper."""
return BaseClassifier.predict(self, X, **kwargs)