Automated evaluation metrics in Prompt Engineering / GenAI - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to calculate accuracy from predictions and labels.

Prompt Engineering / GenAI

accuracy = sum(predictions == [1]) / len(predictions)

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

Atargets

Bpredictions

Coutputs

Dlabels

Attempts:

3 left

💡 Hint

Common Mistakes

Using predictions instead of labels for comparison

Dividing by wrong length

Accuracy compares predicted labels to the true labels.

2fill in blank

medium

Complete the code to compute precision score using sklearn.

Prompt Engineering / GenAI

precision = precision_score([1], predictions)

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

Atargets

Blabels

Coutputs

Dpredictions

Attempts:

3 left

💡 Hint

Common Mistakes

Swapping predictions and labels

Using wrong variable names

Precision compares true labels (first argument) to predictions.

3fill in blank

hard

Fix the error in computing F1 score by filling the missing argument.

Prompt Engineering / GenAI

f1 = f1_score(labels, predictions, average=[1])

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

A'binary'

B'micro'

C'weighted'

D'macro'

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'macro' or 'micro' for binary tasks

Omitting the average argument

For binary classification, average='binary' is required.

4fill in blank

hard

Fill both blanks to create a dictionary of recall scores for each class.

Prompt Engineering / GenAI

recall_scores = {cls: recall_score(labels, predictions, average=[1], labels=[cls]) for cls in [2]

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

A'binary'

B'macro'

C'weighted'

Dclasses

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'macro' average which averages all classes

Not iterating over class labels

Recall per class uses average='binary' and iterates over classes.

5fill in blank

hard

Fill all three blanks to compute a confusion matrix and extract true positives.

Prompt Engineering / GenAI

cm = confusion_matrix([1], [2])
true_positives = cm[[3], [3]]

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

Alabels

Bpredictions

C1

D0

Attempts:

3 left

💡 Hint

Common Mistakes

Swapping labels and predictions

Using wrong index for true positives

The confusion matrix compares true labels and predictions. True positives for class 0 are at cm[0, 0].