Preprocessing Script¶

Created on Thu Dec 3 01:51:16 2020 This is the script for preprocessing of MRS spectrums.

Users can switch between different methods using the method parameter of functions

Abdullah BAS abdullah.bas@boun.edu.tr BME Bogazici University Istanbul / Uskudar @author: abas

preprocess.normalizer(X, norm='l2')¶

Normalizer

Parameters

• X (float) – Input data

• norm (str, optional) – You can choose between L1 and L2. Defaults to ‘l2’.

Returns

X_norm is the normalized output of input X

Return type

[float]

preprocess.scaler(X, method=2)¶

In this function users can select a scaler among sklearn.preprocessing.scale,StandardScaler,MinMaxScaler,RobustScaler

Parameters

• X (float) – Input array.

• method (int, optional) – It is for choosing the scaler.0:scale,1:StandardScaler,2:MinMaxScaler,3:RobustScaler. Defaults to 2.

Returns

X_scaled scaled version of X [scaler]: Scaler with parameters estimated from input X.

Return type

[float]

preprocess.smoother(X, window=11, order=2)¶

Savitzky-Golay filtering

Parameters

• X (float) – Input data

• window (int, optional) – Smoothin kernel length. Defaults to 11.

• order (int, optional) – Filter order. Defaults to 2.

Returns

X_smooth is the smoothed output of input X.

Return type

[float]

preprocess.transformation(X, method=1, powerMet='yeo-johnson')¶

Power transformation

Parameters

• X (float) – Input data

• method (int, optional) – 0:QuantileTransformer,1:Power Transformer. Defaults to 1.

• powerMet (str, optional) – It is essential for method 1. Not included in method 0. Defaults to ‘yeo-johnson’.

Returns

X_tr transformed version of X. [transformer]: transformer with the estimated values from the input X

Return type

[float]