Basic Functions¶

class sekupy.preprocessing.functions.Detrender(degree=1, chunks_attr='chunks', **kwargs)[source]¶

Detrend data.

Parameters:

degree (int, optional) – The polynomial degree of the detrending function (the default is 1)
chunks_attr (str, optional) – The attribute used to get data for the detrend (the default is ‘chunks’, which can be only a sample attribute of the dataset)

Detrend the dataset.

Parameters:: ds (Dataset) – The dataset to be detrended.
Returns:: ds – The detrended dataset
Return type:: Dataset

class sekupy.preprocessing.functions.FeatureStacker(stack_attr=['chunks'], keep_attr=['targets'], selection_dictionary={}, **kwargs)[source]¶

Stack features.

This function is used to stack features with different sample attribute keys, to use these features, jointly.

Parameters:

stack_attr (list, optional) –
This is the attribute to be used for stacking, the resulting dataset will have a sample attribute given by the union of unique attributes

(the default is ‘chunks’)
keep_attr (list, optional) – The attributes to keep, unique values of these attributes will be used to mask the dataset. (the default is [‘targets’])
selection_dictionary (dict, optional) – This will be used to filter the dataset see `SampleSlicer`.

transform(ds)[source]¶

Transform the provided dataset.

This method applies the transformation to the dataset and records the transformation in the dataset’s preprocessing history.

Parameters:: ds (Dataset) – The dataset to transform
Returns:: The transformed dataset
Return type:: Dataset

update_attribute(ds, ds_orig)[source]¶

class sekupy.preprocessing.functions.Resampler(up=1, down=1)[source]¶

transform(ds)[source]¶

Transform the provided dataset.

This method applies the transformation to the dataset and records the transformation in the dataset’s preprocessing history.

Parameters:: ds (Dataset) – The dataset to transform
Returns:: The transformed dataset
Return type:: Dataset

class sekupy.preprocessing.functions.SampleAttributeTransformer(attr=None, fx=None, **kwargs)[source]¶

transform(ds)[source]¶

Transform the provided dataset.

This method applies the transformation to the dataset and records the transformation in the dataset’s preprocessing history.

Parameters:: ds (Dataset) – The dataset to transform
Returns:: The transformed dataset
Return type:: Dataset

class sekupy.preprocessing.functions.SampleAverager(attributes)[source]¶

Init the transformer.

Parameters:: attributes (list) – List of sample attributes whose unique values will be used to identify the samples groups.

transform(ds)[source]¶

Average samples.

Parameters:: ds (Dataset) – The dataset to be transformed.
Returns:: ds – The transformed dataset
Return type:: Dataset

class sekupy.preprocessing.functions.SampleTransformer(attr={})[source]¶

Transforms samples.

This function is used when we need to lock SampleSlicer with TargetTransformer in order to be used with AnalysisIterator.

Parameters:: attr (dictionary) – [description]
Returns:: [description]
Return type:: [type]

transform(ds)[source]¶

Transform the provided dataset.

This method applies the transformation to the dataset and records the transformation in the dataset’s preprocessing history.

Parameters:: ds (Dataset) – The dataset to transform
Returns:: The transformed dataset
Return type:: Dataset

class sekupy.preprocessing.functions.TargetTransformer(attr='targets', fx=None, **kwargs)[source]¶

transform(ds)[source]¶

Transform the provided dataset.

This method applies the transformation to the dataset and records the transformation in the dataset’s preprocessing history.

Parameters:: ds (Dataset) – The dataset to transform
Returns:: The transformed dataset
Return type:: Dataset

class sekupy.preprocessing.functions.TemporalTransformer(attr='frame')[source]¶

Parameters:: attr (dictionary) – [description]
Returns:: [description]
Return type:: [type]

transform(ds)[source]¶

Transform the provided dataset.

This method applies the transformation to the dataset and records the transformation in the dataset’s preprocessing history.

Parameters:: ds (Dataset) – The dataset to transform
Returns:: The transformed dataset
Return type:: Dataset