HampelDetector

HampelDetector(window_size=5, threshold=3)

Hampel filter implementation that works on numpy arrays, implemented with numba.

Parameters

Name	Type	Description	Default
window_size		The window range is from [(i - window_size):(i + window_size)], so window_size is half of the window, counted in number of array elements (as opposed to specify a time span, which is not supported by this implementation)	5
threshold		The threshold for marking an outlier. A low threshold “narrows” the band within which values are deemed as outliers. n_sigmas, default=3.0	3

Methods

Name	Description
detect	Detect anomalies
fit	Set detector parameters based on data.
save	Save a detector for later use.
validate	Check that input data is in correct format and possibly adjust.

detect

HampelDetector.detect(data)

Detect anomalies

Parameters

Name	Type	Description	Default
data	Union[pd.Series, pd.DataFrame]	Time series data with possible anomalies	required

Returns

Name	Type	Description
	pd.Series or pd.DataFrame	Time series with bools, True == anomaly.

fit

HampelDetector.fit(data)

Set detector parameters based on data.

Parameters

Name	Type	Description	Default
data	Union[pd.Series, pd.DataFrame]	Normal (non-anomalous) time series data for training. If DataFrame, must contain exactly one column	required

Returns

Name	Type	Description
	Self

save

HampelDetector.save(path)

Save a detector for later use.

Parameters

Name	Type	Description	Default
path	str or Path	File path to save the detector to.	required

validate

HampelDetector.validate(data)

Check that input data is in correct format and possibly adjust.

Parameters

Name	Type	Description	Default
data	pd.Series or pd.DataFrame	Input data to validate.	required

Returns

Name	Type	Description
	pd.DataFrame	Validated data.

Raises

Name	Type	Description
	WrongInputDataTypeError	If data is not a pd.Series or pd.DataFrame.