Preprocessing

class hana_ml.algorithms.pal.auto_ml.Preprocessing(name, **kwargs)

Preprocessing class. Similar to the function preprocessing.

Parameters:
namestr

The preprocessing algorithm name. The options are:

  • "OneHotEncoder"

  • "FeatureNormalizer"

  • "KBinsDiscretizer"

  • "Imputer"

  • "Discretize"

  • "MDS"

  • "SMOTE"

  • "SMOTETomek"

  • "TomekLinks"

  • "Sampling"

**kwargs: dict

A dict of the keyword args passed to the object. Please refer to the documentation of the specific preprocessing algorithm for parameter information.

Examples

>>> result = Preprocessing(name="FeatureNormalizer").fit_transform(data=data, key="ID", features=["BMI"])
Attributes:
fit_hdbprocedure

Returns the generated hdbprocedure for fit.

predict_hdbprocedure

Returns the generated hdbprocedure for predict.

Methods

fit_transform(data[, key, features])

Execute the preprocessing algorithm and return the transformed dataset.
property fit_hdbprocedure

Returns the generated hdbprocedure for fit.

property predict_hdbprocedure

Returns the generated hdbprocedure for predict.

fit_transform(data, key=None, features=None, **kwargs)

Execute the preprocessing algorithm and return the transformed dataset.

Parameters:
dataDataFrame

Input data.

keystr, optional

Name of the ID column.

Defaults to the index column of data (i.e. data.index) if it is set.

featureslist, optional

The columns to be preprocessed.

Defaults to None.

**kwargs: dict

A dict of the keyword args passed to the fit_transform function. Please refer to the documentation of the specific preprocessing for parameter information.

Inherited Methods from PALBase

Besides those methods mentioned above, the Preprocessing class also inherits methods from PALBase class, please refer to PAL Base for more details.