One-vs-rest and one-vs-one strategies in ML Python - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a one-vs-rest classifier using scikit-learn.

ML Python

from sklearn.multiclass import [1]
from sklearn.svm import SVC

model = [1](estimator=SVC())

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

AOneVsRestClassifier

BOneVsOneClassifier

CRandomForestClassifier

DKNeighborsClassifier

Attempts:

3 left

💡 Hint

Common Mistakes

Using OneVsOneClassifier instead of OneVsRestClassifier

Using a classifier that is not designed for multi-class wrapping

The OneVsRestClassifier wraps a binary classifier to handle multi-class problems by training one classifier per class.

2fill in blank

medium

Complete the code to create a one-vs-one classifier using scikit-learn.

ML Python

from sklearn.multiclass import [1]
from sklearn.svm import SVC

model = [1](estimator=SVC())

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

AOneVsRestClassifier

BLogisticRegression

CDecisionTreeClassifier

DOneVsOneClassifier

Attempts:

3 left

💡 Hint

Common Mistakes

Confusing OneVsRestClassifier with OneVsOneClassifier

Using a base estimator that is not compatible

The OneVsOneClassifier trains one classifier for every pair of classes, useful for multi-class classification.

3fill in blank

hard

Fix the error in the code to correctly train a one-vs-rest classifier.

ML Python

from sklearn.multiclass import OneVsRestClassifier
from sklearn.svm import SVC

model = OneVsRestClassifier(SVC())
model.[1](X_train, y_train)

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

Apredict

Bfit

Ctrain

Dscore

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'train' instead of 'fit' to train the model

Using 'predict' or 'score' methods for training

The method to train a scikit-learn model is fit, not 'train'.

4fill in blank

hard

Fill both blanks to create a one-vs-one classifier and train it.

ML Python

from sklearn.multiclass import [1]
from sklearn.svm import SVC

model = [2](estimator=SVC())
model.fit(X_train, y_train)

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

AOneVsOneClassifier

BOneVsRestClassifier

CRandomForestClassifier

DKNeighborsClassifier

Attempts:

3 left

💡 Hint

Common Mistakes

Mixing OneVsRestClassifier and OneVsOneClassifier

Importing one class but instantiating another

Both blanks require OneVsOneClassifier to correctly create and use the one-vs-one strategy.

5fill in blank

hard

Fill all three blanks to create a one-vs-rest classifier, train it, and predict on test data.

ML Python

from sklearn.multiclass import [1]
from sklearn.svm import SVC

model = [1](estimator=SVC())
model.[2](X_train, y_train)
predictions = model.[3](X_test)

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

AOneVsRestClassifier

Bfit

Cpredict

DOneVsOneClassifier

Attempts:

3 left

💡 Hint

Common Mistakes

Using OneVsOneClassifier instead of OneVsRestClassifier

Using 'train' instead of 'fit'

Using 'fit' instead of 'predict' for predictions

Use OneVsRestClassifier to create the model, fit to train it, and predict to get predictions.