Monitoring model performance in ML Python - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to calculate the accuracy of a model's predictions.

ML Python

accuracy = [1](y_true, y_pred)

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

Amean_squared_error

Baccuracy_score

Clog_loss

Droc_auc_score

Attempts:

3 left

💡 Hint

Common Mistakes

Using mean_squared_error for classification accuracy

Confusing accuracy_score with log_loss

The accuracy_score function calculates the fraction of correct predictions.

2fill in blank

medium

Complete the code to plot the model's loss over epochs using matplotlib.

ML Python

plt.plot(history.history['[1]'])

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

Aval_accuracy

Baccuracy

Closs

Dval_loss

Attempts:

3 left

💡 Hint

Common Mistakes

Plotting accuracy instead of loss

Using validation keys when plotting training loss

The loss key contains the training loss values per epoch.

3fill in blank

hard

Fix the error in the code to compute the F1 score for binary classification.

ML Python

f1 = f1_score(y_true, [1], average='binary')

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

Ay_pred

By_true

Cy_score

Dy_prob

Attempts:

3 left

💡 Hint

Common Mistakes

Passing true labels twice

Using predicted probabilities instead of labels

The f1_score function requires the predicted labels as the second argument.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that tracks precision and recall for each class.

ML Python

metrics = {cls: {'precision': precision_score(y_true, y_pred, pos_label=cls), 'recall': [1](y_true, y_pred, pos_label=[2])} for cls in classes}

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

Arecall_score

Bprecision_score

Ccls

Dy_pred

Attempts:

3 left

💡 Hint

Common Mistakes

Using precision_score twice

Passing predicted labels as pos_label

Use recall_score for recall and cls as the positive label for each class.

5fill in blank

hard

Fill all three blanks to create a filtered dictionary of metrics where F1 score is above 0.7.

ML Python

filtered_metrics = {cls: metrics[cls] for cls in metrics if metrics[cls]['[1]'] [2] [3]

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

Af1_score

B>

C0.7

Dprecision

Attempts:

3 left

💡 Hint

Common Mistakes

Using precision instead of f1_score

Using wrong comparison operator

Using string '0.7' instead of number

Filter classes where the f1_score is greater than 0.7.