Scikit-Learn maxabs_scale() for Data Preprocessing

Scaling features of a dataset is a common preprocessing step in machine learning.

maxabs_scale() scales each feature by its maximum absolute value, which is particularly useful for datasets with varying scales and sparse data.

from sklearn.preprocessing import maxabs_scale
import numpy as np

# create a synthetic dataset
data = np.array([[1, -1, 2],
                 [2, 0, 0],
                 [0, 1, -1]])

# apply maxabs_scale
scaled_data = maxabs_scale(data)

# display before and after scaling
print("Original data:")
print(data)

print("\nScaled data:")
print(scaled_data)

Running the example gives an output like:

Original data:
[[ 1 -1  2]
 [ 2  0  0]
 [ 0  1 -1]]

Scaled data:
[[ 0.5 -1.   1. ]
 [ 1.   0.   0. ]
 [ 0.   1.  -0.5]]

The steps are as follows:

1. A synthetic dataset is created using numpy.array() with mixed positive and negative values.
2. The maxabs_scale() function is applied to the dataset to scale each feature by its maximum absolute value.
3. The original dataset is printed to show its values before scaling.
4. The scaled dataset is printed to show the transformed values after applying maxabs_scale().

This example shows how to preprocess data using maxabs_scale() in scikit-learn, which is particularly useful for datasets with features of varying scales and sparse datasets.