Given the dataset below, use information gain to calculate the first attribute the decision tree should split on.

import numpy as np from sklearn.datasets import load iris data iris, labels iris = load iris( return X y=True) print( 'There are samples with attributes.'.format ( data_iris.shape)) print('The labels consist of f classes'.format(len(np.unique(labels_iris)))) There are 150 samples with 4 attributes. The labels consist of 3 classes import numpy as np from sklearn.datasets import load iris data iris, labels iris = load iris( return X y=True) print( 'There are samples with attributes.'.format ( data_iris.shape)) print('The labels consist of f classes'.format(len(np.unique(labels_iris)))) There are 150 samples with 4 attributes. The labels consist of 3 classes