Bagging concept in ML Python - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create multiple bootstrap samples for bagging.

ML Python

from sklearn.utils import resample

samples = [resample(data, replace=[1], n_samples=100) for _ in range(10)]

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

AFalse

B0

CNone

DTrue

Attempts:

3 left

💡 Hint

Common Mistakes

Using replace=False will create samples without replacement, which is not bagging.

Setting replace to None or 0 causes errors.

Bagging uses bootstrap samples with replacement, so replace=True is needed.

2fill in blank

medium

Complete the code to train multiple decision trees on bootstrap samples.

ML Python

from sklearn.tree import DecisionTreeClassifier

models = []
for sample in samples:
    clf = DecisionTreeClassifier()
    clf.fit(sample[[1]], sample[[2]])
    models.append(clf)

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

Atarget

BX

Cy

Ddata

Attempts:

3 left

💡 Hint

Common Mistakes

Using the whole data instead of target for labels.

Confusing features and target variables.

Decision trees are trained using features X and target y.

3fill in blank

hard

Fix the error in the code to aggregate predictions from all models by majority vote.

ML Python

import numpy as np

predictions = np.array([model.predict(X_test) for model in models])
final_prediction = np.apply_along_axis(lambda x: np.bincount(x).[1](), axis=0, arr=predictions)

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

Asum

Bargmax

Cmax

Dmean

Attempts:

3 left

💡 Hint

Common Mistakes

Using max returns the maximum count value, not the index.

Using sum or mean does not give the majority class.

np.bincount(x).argmax() returns the most common prediction (majority vote) among models.

4fill in blank

hard

Fill both blanks to create a bagging ensemble using scikit-learn's BaggingClassifier.

ML Python

from sklearn.ensemble import BaggingClassifier

bagging = BaggingClassifier(base_estimator=[1], n_estimators=10, random_state=42)
bagging.fit([2], y_train)

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

ADecisionTreeClassifier()

BRandomForestClassifier()

CX_train

DX_test

Attempts:

3 left

💡 Hint

Common Mistakes

Using RandomForestClassifier as base estimator is incorrect.

Using test data X_test for training causes errors.

The base estimator for bagging is usually a decision tree, and training uses X_train features.

5fill in blank

hard

Fill all three blanks to evaluate bagging model accuracy on test data.

ML Python

from sklearn.metrics import [1]

preds = bagging.predict([2])
accuracy = [3](y_test, preds)
print(f"Accuracy: {accuracy:.2f}")

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

Aaccuracy_score

BX_test

Dconfusion_matrix

Attempts:

3 left

💡 Hint

Common Mistakes

Using confusion_matrix instead of accuracy_score for accuracy.

Predicting on training data instead of test data.

Use accuracy_score metric and X_test features to predict and evaluate accuracy.