Precision-recall curves in TensorFlow - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the function that computes precision and recall.

TensorFlow

from sklearn.metrics import [1]

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

Aprecision_recall_curve

Broc_curve

Caccuracy_score

Dconfusion_matrix

Attempts:

3 left

💡 Hint

Common Mistakes

Importing roc_curve instead, which is for ROC curves.

Importing accuracy_score, which does not provide precision or recall values.

The precision_recall_curve function computes precision and recall values for different thresholds.

2fill in blank

medium

Complete the code to compute precision and recall from true labels and predicted scores.

TensorFlow

precision, recall, thresholds = precision_recall_curve(y_true, [1])

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

Ay_labels

By_pred

Cy_scores

Dy_prob

Attempts:

3 left

💡 Hint

Common Mistakes

Passing binary predictions instead of scores.

Passing true labels again instead of predicted scores.

The precision_recall_curve function requires the true labels and the predicted scores or probabilities, here named y_scores.

3fill in blank

hard

Fix the error in the code to plot the precision-recall curve correctly.

TensorFlow

plt.plot(recall, [1])
plt.xlabel('Recall')
plt.ylabel('Precision')
plt.title('Precision-Recall Curve')
plt.show()

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

Ay_scores

Bthresholds

Cy_true

Dprecision

Attempts:

3 left

💡 Hint

Common Mistakes

Plotting thresholds instead of precision.

Plotting true labels or scores instead of precision.

Precision values should be plotted on the y-axis against recall on the x-axis.

4fill in blank

hard

Fill both blanks to compute average precision score and print it.

TensorFlow

from sklearn.metrics import [1]
avg_precision = [2](y_true, y_scores)
print(f'Average Precision: {avg_precision:.2f}')

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

Aaverage_precision_score

Bprecision_recall_curve

Droc_auc_score

Attempts:

3 left

💡 Hint

Common Mistakes

Importing roc_auc_score instead, which is for ROC AUC.

Using precision_recall_curve instead of average_precision_score.

The average_precision_score function computes the average precision from true labels and predicted scores.

5fill in blank

hard

Fill all three blanks to create a dictionary of precision and recall values for thresholds above 0.5.

TensorFlow

pr_dict = {threshold: ([1], [2]) for precision, recall, threshold in zip(precision, recall, [3]) if threshold > 0.5}

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

Aprecision

Brecall

Cthresholds

Dy_scores

Attempts:

3 left

💡 Hint

Common Mistakes

Using y_scores instead of thresholds in the zip.

Swapping precision and recall order in the tuple.

This dictionary comprehension pairs precision and recall values for thresholds greater than 0.5 using the thresholds array.