Feature Engineering Made Easy

The min-max scaling method

Min-max scaling is similar to z-score normalization in that it will replace every value in a column with a new value using a formula. In this case, that formula is:

m = (x -x_min) / (x_max -x_min)

Where:

• m is our new value
• x is the original cell value
• x_min is the minimum value of the column
• x_max is the maximum value of the column

Using this formula, we will see that the values of each column will now be between zero and one. Let's take a look at an example using a built-in scikit-learn module:

# import the sklearn modulefrom sklearn.preprocessing import MinMaxScaler#instantiate the classmin_max = MinMaxScaler()# apply the Min Max Scalingpima_min_maxed = pd.DataFrame(min_max.fit_transform(pima_imputed), columns=pima_column_names) ...