Model evaluation best practices in Computer Vision - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to calculate the accuracy of a classification model.

Computer Vision

accuracy = accuracy_score(y_true, [1])

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

Ay_train

By_test

Cy_pred

Dy_val

Attempts:

3 left

💡 Hint

Common Mistakes

Using training labels instead of predicted labels.

Confusing true labels with predicted labels.

The accuracy_score function compares the true labels y_true with the predicted labels y_pred to calculate accuracy.

2fill in blank

medium

Complete the code to split the dataset into training and testing sets.

Computer Vision

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=[1], random_state=42)

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

A0.25

B0.5

C0.1

D0.75

Attempts:

3 left

💡 Hint

Common Mistakes

Setting test_size too high, leaving too little training data.

Using an integer instead of a float for test_size.

The test_size parameter defines the proportion of the dataset to include in the test split. 0.25 means 25% test data and 75% training data.

3fill in blank

hard

Fix the error in the code to compute the confusion matrix correctly.

Computer Vision

cm = confusion_matrix(y_true, [1])

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

Ay_train

By_test

Cy_val

Dy_pred

Attempts:

3 left

💡 Hint

Common Mistakes

Using test labels instead of predicted labels.

Mixing training labels with true labels.

The confusion matrix compares true labels y_true with predicted labels y_pred. Using other variables causes incorrect results or errors.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that maps each class to its precision score.

Computer Vision

precision_scores = {cls: precision_score(y_true, y_pred, labels=[cls], average=[1]) for cls in [2]

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

Amacro

Bmicro

Cweighted

Dclasses

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'micro' average which aggregates globally.

Using an undefined variable instead of class labels.

The average='macro' calculates precision for each class independently. The variable classes holds the list of class labels to iterate over.

5fill in blank

hard

Fill all three blanks to compute the F1 score with macro averaging and print the result.

Computer Vision

f1 = f1_score(y_true, [1], average=[2])
print('F1 Score:', [3])

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

Ay_pred

Bmicro

Cf1

Dmacro

Attempts:

3 left

💡 Hint

Common Mistakes

Using micro average which aggregates globally.

Printing the wrong variable.

The F1 score is computed using predicted labels y_pred and macro averaging to treat all classes equally. The variable f1 holds the score to print.