bach.DataFrame.minmax_scale

minmax_scale​

(feature_range=(0, 1))

​[source]

Scales all numeric series based on a given range.

Parameters​

feature_range (Tuple[int, int]) – tuple(min, max) desired range to use for scaling.

Returns​

DataFrame

Return type​

bach.dataframe.DataFrame

Each transformation per feature is performed as follows:

In:
range_min, range_max = (0, 1)
feature_std = (feature - min_feature) / (max_feature - min_feature)
scaled_feature = feature_std * (range_max - range_min) + range_min
scaled_feature.to_pandas()
Out:
index
a    0.000000
b    0.333333
c    0.666667
d    1.000000
Name: feature, dtype: float64

Where:

  • feature is the series to be scaled
  • feature_min is the minimum value of feature
  • feature_max is the maximum value of feature
  • range_min, range_max = feature_range