Evaluation and confusion matrix in Computer Vision - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the function that computes the confusion matrix.

Computer Vision

from sklearn.metrics import [1]

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

Atrain_test_split

Bconfusion_matrix

Cclassification_report

Daccuracy_score

Attempts:

3 left

💡 Hint

Common Mistakes

Importing accuracy_score instead of confusion_matrix

Importing train_test_split which is for splitting data

The confusion_matrix function from sklearn.metrics is used to compute the confusion matrix for classification results.

2fill in blank

medium

Complete the code to calculate accuracy from true and predicted labels.

Computer Vision

accuracy = sum(y_true == y_pred) / [1]

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

Alen(y_pred)

Bsum(y_pred)

Clen(y_true)

Dsum(y_true)

Attempts:

3 left

💡 Hint

Common Mistakes

Dividing by sum of predictions instead of total samples

Using length of predictions which might differ if data is inconsistent

Accuracy is the number of correct predictions divided by the total number of true labels, which is len(y_true).

3fill in blank

hard

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

Computer Vision

cm = confusion_matrix([1], y_pred)

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

Apredictions

By_pred

Clabels

Dy_true

Attempts:

3 left

💡 Hint

Common Mistakes

Swapping true and predicted labels

Passing undefined variable names

The first argument to confusion_matrix must be the true labels y_true, followed by predicted labels.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that counts true positives for each class.

Computer Vision

true_positives = {cls: sum((y_true == cls) & ([1] == cls)) for cls in [2]

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

Ay_pred

By_true

Cclasses

Dlabels

Attempts:

3 left

💡 Hint

Common Mistakes

Using y_true instead of y_pred in the condition

Using undefined variable names for classes

We compare predicted labels y_pred to each class and iterate over all classes stored in classes to count true positives.

5fill in blank

hard

Fill all three blanks to compute precision for each class from the confusion matrix.

Computer Vision

precision = {j: cm[[1], [2]] / sum(cm[:, [3]]) for j in classes}

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

Ai

Bj

Attempts:

3 left

💡 Hint

Common Mistakes

Mixing row and column indices

Dividing by sum over rows instead of columns

Precision for each class is true positives cm[j, j] divided by the sum of predicted positives sum(cm[:, j]). Here, j represents the class index.