Why pipelines automate the ML workflow in MLOps - Test Your Understanding

Practice - 5 Tasks

1fill in blank

easy

Complete the code to define a pipeline step that preprocesses data.

MLOps

def preprocess_data(data):
    cleaned_data = data.[1]()
    return cleaned_data

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

Aremove

Bfilter

Cdropna

Dclean

Attempts:

3 left

💡 Hint

Common Mistakes

Using a method that does not exist on the data object.

Trying to filter instead of dropping missing values.

The dropna() method removes missing values, which is a common preprocessing step.

2fill in blank

medium

Complete the code to add a training step in the ML pipeline.

MLOps

def train_model(features, labels):
    model = SomeModel()
    model.[1](features, labels)
    return model

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

Apredict

Bfit_transform

Ctrain

Dfit

Attempts:

3 left

💡 Hint

Common Mistakes

Using predict instead of fit.

Using a method that transforms data instead of training.

The fit() method trains the model using features and labels.

3fill in blank

hard

Fix the error in the pipeline step that evaluates the model.

MLOps

def evaluate_model(model, test_features, test_labels):
    predictions = model.predict(test_features)
    accuracy = [1](test_labels, predictions)
    return accuracy

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

Aaccuracy_score

Bevaluate_accuracy

Cscore_accuracy

Dcalc_accuracy

Attempts:

3 left

💡 Hint

Common Mistakes

Using a non-existent function name.

Confusing evaluation function names.

The correct function to calculate accuracy is accuracy_score from sklearn.metrics.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that maps features to their importance if importance is above 0.1.

MLOps

important_features = {feature: importance for feature, importance in zip(model.feature_names_in_, model.[1]) if importance [2] 0.1}

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

Afeature_importances_

B>

C<

Dcoef_

Attempts:

3 left

💡 Hint

Common Mistakes

Using the wrong attribute for feature importance.

Using less than instead of greater than in the condition.

feature_importances_ holds feature importance values, and we filter those greater than 0.1.

5fill in blank

hard

Fill all three blanks to create a pipeline dictionary that stores model name, accuracy, and timestamp if accuracy is above 0.8.

MLOps

pipeline_result = {
    '[1]': model_name.upper(),
    'accuracy': accuracy,
    'timestamp': datetime.[2](),
}
if accuracy [3] 0.8:
    save_result(pipeline_result)

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

Amodel

Bnow

C>

Dname

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong dictionary keys.

Using incorrect datetime method.

Using wrong comparison operator.

Use 'name' as the key for model name, datetime.now() for current time, and > to check accuracy above 0.8.