Classification reports in TensorFlow - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the function that generates a classification report.

TensorFlow

from sklearn.metrics import [1]

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

Aclassification_report

Baccuracy_score

Cconfusion_matrix

Droc_auc_score

Attempts:

3 left

💡 Hint

Common Mistakes

Importing confusion_matrix instead of classification_report

Using accuracy_score which only gives one metric

The classification_report function from sklearn.metrics provides detailed metrics for classification models.

2fill in blank

medium

Complete the code to generate predictions from the model on test data.

TensorFlow

y_pred = model.[1](X_test)

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

Aevaluate

Bpredict

Cfit

Dtransform

Attempts:

3 left

💡 Hint

Common Mistakes

Using evaluate which returns loss and metrics

Using fit which trains the model

The predict method generates predicted labels or probabilities from the model.

3fill in blank

hard

Fix the error in the code to correctly generate a classification report for true and predicted labels.

TensorFlow

report = classification_report(y_true, [1])
print(report)

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

Ay_pred

Bmodel.predict(X_test)

Cy_true

DX_test

Attempts:

3 left

💡 Hint

Common Mistakes

Passing y_true twice

Passing input features X_test instead of predictions

The classification report requires the true labels and predicted labels as inputs. Here, y_pred holds the predicted labels.

4fill in blank

hard

Fill both blanks to create a classification report with target names for classes 'cat' and 'dog'.

TensorFlow

report = classification_report(y_true, y_pred, target_names=[1])
print([2])

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

A['cat', 'dog']

Bclassification_report

Creport

D['dog', 'cat']

Attempts:

3 left

💡 Hint

Common Mistakes

Passing the function name instead of the report variable to print

Swapping class names order incorrectly

The target_names parameter expects a list of class names. The variable report holds the generated report to print.

5fill in blank

hard

Fill all three blanks to generate predictions, convert probabilities to class labels, and print the classification report.

TensorFlow

y_prob = model.[1](X_test)
y_pred = y_prob.argmax(axis=[2])
print(classification_report(y_true, [3]))

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

Apredict

B1

Cy_pred

Dpredict_proba

Attempts:

3 left

💡 Hint

Common Mistakes

Using predict instead of predict_proba for probabilities

Using wrong axis in argmax

Printing classification_report with wrong variable

Use predict_proba to get probabilities, then use argmax along axis 1 to get class labels, and finally print the report with predicted labels.