normalize_operator

Operator to normalize the integer/float values in a dataframe

class tasrif.processing_pipeline.custom.normalize_operator.NormalizeOperator(feature_names='all', method='zscore', normalization_parameters=None)

Normalizes the values for the supplied feature_names using the specified algorithm. This operator works on a 2D data frames where the feature_names represent the features. The returned data frame contains normalized values in the specified feature_names

Examples

>>> import pandas as pd
>>> from tasrif.processing_pipeline.custom import NormalizeOperator
>>> df = pd.DataFrame([
>>>     [1, "2020-05-01 00:00:00", 1],
>>>     [1, "2020-05-01 01:00:00", 1],
>>>     [1, "2020-05-01 03:00:00", 2],
>>>     [2, "2020-05-02 00:00:00", 1],
>>>     [2, "2020-05-02 01:00:00", 1]],
>>>     columns=['logId', 'timestamp', 'sleep_level'])
>>>
>>> df['timestamp'] = pd.to_datetime(df['timestamp'])
>>> df = df.set_index('timestamp')
>>> op = NormalizeOperator('all', 'minmax', {'min': 0, 'max': 2})
>>> op.process(df)
[            sleep_level
timestamp
2020-05-01     1.333333
2020-05-02     1.000000]

__init__(feature_names='all', method='zscore', normalization_parameters=None)

Creates a new instance of NormalizeOperator

Parameters

• feature_names (list, str) – feature_names in the given dataframe to normalize

• method (str) – The normalization method (‘zscore’, ‘minmax’, ‘maxabs’, ‘robust’) to be used

• normalization_parameters (dict) –

  Dictionary containing parameters for a specific normalization.::

  Default normalization parameters are as follows:

  • If method is zscore

    normalization_params = {'with_mean': True, 'with_std': True}

  • If method is minmax

    normalization_params = {'feature_range': (0,1)}

  • If method is maxabs

    normalization_params = {}

  • If method is robust

    normalization_params = {'with_scaling': True, 'with_centering': True, 'quantile_range':(25.0, 75.0), 'unit_variance': False}

Raises

ValueError – parameter method unknown.

linear_fit_operator normalize_transform_operator