Automated testing for ML code in MLOps - Mini Project: Build & Apply

Automated Testing for ML Code

📖 Scenario: You are working on a machine learning project. To keep your code reliable, you want to add automated tests that check if your ML functions work correctly.This helps catch mistakes early, just like checking your homework before submitting it.

🎯 Goal: Build a simple automated test for a machine learning function that calculates accuracy.You will create test data, set expected results, write the test logic, and print the test outcome.

📋 What You'll Learn

Create a list of true labels and predicted labels

Set an expected accuracy value

Write a function to calculate accuracy

Write a test that compares calculated accuracy to expected accuracy

Print the test result as 'Test passed' or 'Test failed'

💡 Why This Matters

🌍 Real World

Automated tests help catch errors in ML code early, saving time and improving reliability before deploying models.

💼 Career

ML engineers and MLOps specialists use automated testing to ensure their models and code work correctly in production environments.

1

Create test data for labels

Create two lists called true_labels and predicted_labels with these exact values: true_labels = [1, 0, 1, 1, 0] and predicted_labels = [1, 0, 0, 1, 0].

MLOps

# Create the true_labels and predicted_labels lists
# Your code here

Need a hint?

Use square brackets to create lists with the exact numbers given.

2

Set expected accuracy value

Create a variable called expected_accuracy and set it to the float value 0.8.

MLOps

true_labels = [1, 0, 1, 1, 0]
predicted_labels = [1, 0, 0, 1, 0]
# Set expected_accuracy to 0.8
# Your code here

Need a hint?

Use a variable name exactly as expected_accuracy and assign it the value 0.8.

3

Write accuracy calculation function

Define a function called calculate_accuracy that takes two parameters: true_labels and predicted_labels. Inside, calculate accuracy as the number of matching labels divided by total labels, and return the accuracy as a float.

MLOps

true_labels = [1, 0, 1, 1, 0]
predicted_labels = [1, 0, 0, 1, 0]
expected_accuracy = 0.8
# Define calculate_accuracy function below
# Your code here

Need a hint?

Use a for loop with zip to compare pairs of labels.

4

Write test and print result

Call calculate_accuracy(true_labels, predicted_labels) and store the result in actual_accuracy. Then write an if statement that prints 'Test passed' if actual_accuracy equals expected_accuracy, otherwise print 'Test failed'.

MLOps

true_labels = [1, 0, 1, 1, 0]
predicted_labels = [1, 0, 0, 1, 0]
expected_accuracy = 0.8
def calculate_accuracy(true_labels, predicted_labels):
    correct = 0
    for t, p in zip(true_labels, predicted_labels):
        if t == p:
            correct += 1
    return correct / len(true_labels)

# Call the function and print test result
# Your code here

Need a hint?

Use an if statement to compare actual_accuracy and expected_accuracy.