To bring values into range `[-1, 1]` we have to use the next formula:

Here we have the following values:
- **x_scaled** - normalized feature element,
- **x** - unnormalized feature element,
- **max(x)** -- maximum feature element.

There is a function in the **sklearn** library that normalizes data according to the formula given above: `MaxAbsScaler()`. In order to work with this function, it must first be imported in such a way:

from sklearn.preprocessing import MaxAbsScaler

Let's look at an example of how we apply this normalization to a very simple array.

from sklearn.preprocessing import MaxAbsScaler
data = [[10, 5, -6],[11, -9, 4],[-10, 0, 1]]

# Normalizer initialization
scaler = MaxAbsScaler()
# Dataset transfer and transformation
scaler.fit(data)
scaled_data = scaler.transform(data)
print('Data before normalization', data)
print('Data after normalization', scaled_data)

If you run this code you will get two different arrays: before and after normalization. And this function really works, because you can make sure that data after using `MaxAbsScaler()` function really lie within an interval `[-1, 1]`. Look below.v

MaxAbsScaler

解答