Mutual information for feature selection in ML Python - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the function for mutual information classification.

ML Python

from sklearn.feature_selection import [1]

Drag options to blanks, or click blank then click option'

Amutual_info_regression

Bmutual_info_classif

CSelectKBest

Dchi2

Attempts:

3 left

💡 Hint

Common Mistakes

Choosing mutual_info_regression which is for regression tasks.

Choosing SelectKBest which is a selector, not the mutual information function.

The function mutual_info_classif computes mutual information for classification tasks, which helps select important features.

2fill in blank

medium

Complete the code to calculate mutual information scores for features X and target y.

ML Python

mi_scores = [1](X, y)

Drag options to blanks, or click blank then click option'

Amutual_info_classif

Bf_classif

Cchi2

Dmutual_info_regression

Attempts:

3 left

💡 Hint

Common Mistakes

Using mutual_info_regression which is for regression problems.

Using chi2 or f_classif which are different feature scoring methods.

mutual_info_classif calculates mutual information scores between features and a classification target.

3fill in blank

hard

Fix the error in the code to select top 3 features based on mutual information scores.

ML Python

from sklearn.feature_selection import SelectKBest
selector = SelectKBest(score_func=[1], k=3)
X_new = selector.fit_transform(X, y)

Drag options to blanks, or click blank then click option'

Amutual_info_regression

Bchi2

Cmutual_info_classif

Df_regression

Attempts:

3 left

💡 Hint

Common Mistakes

Using chi2 which requires non-negative features.

Using mutual_info_regression which is for regression.

Use mutual_info_classif as the scoring function for classification feature selection.

4fill in blank

hard

Fill both blanks to create a dictionary of feature names and their mutual information scores, filtering scores greater than 0.1.

ML Python

mi_dict = {feature: [1] for feature, score in zip(feature_names, mi_scores) if score [2] 0.1}

Drag options to blanks, or click blank then click option'

Ascore

B>

C<

Dfeature

Attempts:

3 left

💡 Hint

Common Mistakes

Using feature instead of score as dictionary value.

Using '<' instead of '>' in the condition.

The dictionary maps each feature to its score, including only scores greater than 0.1.

5fill in blank

hard

Fill both blanks to select top 5 features using mutual information and transform the dataset.

ML Python

selector = SelectKBest(score_func=[1], k=[2])
selected_features = selector.fit_transform(X, y)
selected_feature_names = [name for i, name in enumerate(feature_names) if selector.get_support()[i]]

Drag options to blanks, or click blank then click option'

Amutual_info_classif

B3

Cindex

D5

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong scoring function.

Selecting wrong number of features.

Use mutual_info_classif as the scoring function, select top 5 features.