def predict(self,
X: np.ndarray,
return_instance_score: bool = True) \
-> Dict[Dict[str, str], Dict[np.ndarray, np.ndarray]]:
"""
Compute outlier scores and transform into outlier predictions.
Parameters
----------
X
Batch of instances.
return_instance_score
Whether to return instance level outlier scores.
Returns
-------
Dictionary containing 'meta' and 'data' dictionaries.
'meta' has the model's metadata.
'data' contains the outlier predictions and instance level outlier scores.
"""
# compute outlier scores
iscore = self.score(X)
# values above threshold are outliers
outlier_pred = (iscore > self.threshold).astype(int)
# populate output dict
od = outlier_prediction_dict()
od['meta'] = self.meta
od['data']['is_outlier'] = outlier_pred
if return_instance_score:
od['data']['instance_score'] = iscore
return od
def predict(self,
X: np.ndarray,
outlier_type: str = 'instance',
outlier_perc: float = 100.,
batch_size: int = int(1e10),
return_feature_score: bool = True,
return_instance_score: bool = True) \
-> Dict[Dict[str, str], Dict[np.ndarray, np.ndarray]]:
"""
Compute outlier scores and transform into outlier predictions.
Parameters
----------
X
Univariate or multivariate time series.
outlier_type
Predict outliers at the 'feature' or 'instance' level.
outlier_perc
Percentage of sorted feature level outlier scores used to predict instance level outlier.
batch_size
Batch size used when making predictions with the seq2seq model.
return_feature_score
Whether to return feature level outlier scores.
return_instance_score
Whether to return instance level outlier scores.
Returns
-------
Dictionary containing 'meta' and 'data' dictionaries.
'meta' has the model's metadata.
'data' contains the outlier predictions and both feature and instance level outlier scores.
"""
# compute outlier scores
fscore, iscore = self.score(X,
outlier_perc=outlier_perc,
batch_size=batch_size)
if outlier_type == 'feature':
outlier_score = fscore
elif outlier_type == 'instance':
outlier_score = iscore
else:
raise ValueError(
'`outlier_score` needs to be either `feature` or `instance`.')
# values above threshold are outliers
outlier_pred = (outlier_score > 0).astype(int)
# populate output dict
od = outlier_prediction_dict()
od['meta'] = self.meta
od['data']['is_outlier'] = outlier_pred
if return_feature_score:
od['data']['feature_score'] = fscore
if return_instance_score:
od['data']['instance_score'] = iscore
return od
def predict(self,
X: np.ndarray,
outlier_type: str = 'instance',
batch_size: int = int(1e10),
return_feature_score: bool = True,
return_instance_score: bool = True) \
-> Dict[Dict[str, str], Dict[np.ndarray, np.ndarray]]:
"""
Predict whether instances are outliers or not.
Parameters
----------
X
Batch of instances.
outlier_type
Predict outliers at the 'feature' or 'instance' level.
batch_size
Batch size used when making predictions with the generative model.
return_feature_score
Whether to return feature level outlier scores.
return_instance_score
Whether to return instance level outlier scores.
Returns
-------
Dictionary containing 'meta' and 'data' dictionaries.
'meta' has the model's metadata.
'data' contains the outlier predictions and both feature and instance level outlier scores.
"""
# compute outlier scores
fscore, iscore = self.score(X, batch_size=batch_size)
if outlier_type == 'feature':
outlier_score = fscore
elif outlier_type == 'instance':
outlier_score = iscore
else:
raise ValueError(
'`outlier_score` needs to be either `feature` or `instance`.')
# values above threshold are outliers
outlier_pred = (outlier_score > self.threshold).astype(int)
# populate output dict
od = outlier_prediction_dict()
od['meta'] = self.meta
od['data']['is_outlier'] = outlier_pred
if return_feature_score:
od['data']['feature_score'] = fscore
if return_instance_score:
od['data']['instance_score'] = iscore
return od
def predict(
self,
X: np.ndarray,
outlier_type: str = "instance",
outlier_perc: float = 100.0,
batch_size: int = int(1e10),
return_feature_score: bool = True,
return_instance_score: bool = True,
) -> Dict[Dict[str, str], Dict[np.ndarray, np.ndarray]]:
assert return_instance_score == self.expected_return_instance_score
assert return_feature_score == self.expected_return_feature_score
od = outlier_prediction_dict()
od["data"]["is_outlier"] = self.expect_return_is_outlier
return od
def predict(
self,
df: pd.DataFrame,
return_instance_score: bool = True,
return_forecast: bool = True
) -> Dict[Dict[str, str], Dict[pd.DataFrame, pd.DataFrame]]:
"""
Compute outlier scores and transform into outlier predictions.
Parameters
----------
df
DataFrame with columns `ds` with timestamps and `y` with values which
need to be flagged as outlier or not.
return_instance_score
Whether to return instance level outlier scores.
return_forecast
Whether to return the model forecast.
Returns
-------
Dictionary containing 'meta' and 'data' dictionaries.
'meta' has the model's metadata.
'data' contains the outlier predictions, instance level outlier scores and the model forecast.
"""
# compute outlier scores
forecast = self.score(df)
iscore = pd.DataFrame(data={
'ds': df['ds'].values,
'instance_score': forecast['score']
})
# values above threshold are outliers
outlier_pred = pd.DataFrame(
data={
'ds': df['ds'].values,
'is_outlier': (forecast['score'] > 0.).astype(int)
})
# populate output dict
od = outlier_prediction_dict()
od['meta'] = self.meta
od['data']['is_outlier'] = outlier_pred
if return_instance_score:
od['data']['instance_score'] = iscore
if return_forecast:
od['data']['forecast'] = forecast
return od
def predict(self,
X: np.ndarray,
t: np.ndarray = None,
return_instance_score: bool = True) \
-> Dict[Dict[str, str], Dict[np.ndarray, np.ndarray]]:
"""
Compute outlier scores and transform into outlier predictions.
Parameters
----------
X
Uniformly sampled time series instances.
t
Equidistant timestamps corresponding to each input instances (i.e, the array should contain
numerical values in increasing order). If not provided, the timestamps will be replaced by an array of
integers `[0, 1, ... , N - 1]`, where `N` is the size of the input time series.
return_instance_score
Whether to return instance level outlier scores.
Returns
-------
Dictionary containing `meta` and `data` dictionaries.
- `meta` - has the model's metadata.
- `data` - contains the outlier predictions and instance level outlier scores.
"""
if t is None:
t = np.arange(X.shape[0])
# compute outlier scores
iscore = self.score(X, t)
# values above threshold are outliers
outlier_pred = (iscore > self.threshold).astype(int)
# populate output dict
od = outlier_prediction_dict()
od['meta'] = self.meta
od['data']['is_outlier'] = outlier_pred
if return_instance_score:
od['data']['instance_score'] = iscore
return od
