direpack.preprocessing.robcent.VersatileScaler

class VersatileScaler(center='mean', scale='std', trimming=0)[source]

VersatileScaler Center and Scale data about classical or robust location and scale estimates

Parameters

  • center (str or callable, location estimator. String has to be name of the) – function to be used, or ‘None’.

  • scale (str or callable, scale estimator) –

  • trimming (trimming percentage to be used in location and scale estimation.) –

Arguments for methods

  • X: array-like, n x p, the data.

  • trimming: float, fraction to be trimmed (must be in (0,1)).

Remarks

Options for classical estimators ‘mean’ and ‘std’ also give access to robust trimmed versions.

__init__(center='mean', scale='std', trimming=0)[source]

Initialize values. Check if correct options provided.

Methods

__init__([center, scale, trimming])

Initialize values.

fit(X)

Estimate location and scale, store these in the class object.

fit_transform(X)

Estimate center and scale for training data and scale these data

get_params([deep])

Get parameters for this estimator.

inverse_transform([Xs])

Transform scaled data back to their original scale

predict(Xn)

Standardize new data on previously estimated location and scale.

set_params(**params)

Set the parameters of this estimator.

transform(X)

Center and/or scale training data to pre-estimated location and scale

Attributes