Introduction to Testing with Doctest for Data Scientists¶

This guide introduces you to doctest, a simple and effective tool for testing and documenting your Python code. As a team of data scientists, you might not have extensive coding experience, but doctest provides a straightforward way to ensure your code works as expected while also serving as a form of documentation.

What is Doctest?¶

doctest is a module included in the Python standard library that allows you to test your code by running examples embedded in the documentation and verifying that they produce the expected results. This approach helps to ensure that your code works as documented.

Why Use Doctest?¶

• Ease of Use: doctest is simple to set up and use, making it an excellent starting point for those new to testing.
• Documentation: Tests written with doctest serve as both test cases and examples of how to use your functions, improving code readability and usability.
• Immediate Feedback: Running doctest provides immediate feedback on whether your code is working as expected.

Why Unit Testing is Important¶

Even though libraries like pandas, scikit-learn, and TensorFlow are well-tested and reliable, it’s crucial to unit test your custom code. In machine learning projects, custom code often includes data preprocessing, feature engineering, model evaluation, and utility functions. Ensuring the correctness of this code is vital for the accuracy and reliability of your models. Unit testing helps catch errors early, facilitates code maintenance, and promotes confidence in your code’s functionality.

As a data scientist, testing your code before opening a pull request (PR) and merging it into the production branch is critical. This practice is especially important in projects like Retrieval-Augmented Generation (RAG) systems, where data scientists are expected to contribute to production code from the very beginning. Unlike traditional machine learning projects, RAG systems require seamless integration of custom code into the production environment, making thorough testing essential to prevent potential issues and ensure smooth operation.

By incorporating unit testing into your workflow, you ensure that your contributions are reliable and maintain the integrity of the overall project. This not only enhances the quality of your code but also promotes collaborative development, where team members can confidently build upon each other's work.

Doctest for Simple Unit Tests¶

Here's an example of how to use doctest with Google-style docstrings:

def add(a, b):
    """
    Adds two numbers together.

    Args:
        a (int): The first number.
        b (int): The second number.

    Returns:
        int: The sum of the two numbers.

    Example:
        >>> add(2, 3)
        5
        >>> add(-1, 1)
        0
        >>> add(0, 0)
        0
    """
    return a + b

Running Doctests¶

To run the doctests in your module, use the following command:

python -m doctest -v your_module.py

Additional Resources¶

For further learning on testing in Python, including more advanced techniques, consider the following resources:

• Introduction to Testing in Python - DataCamp
• Python Testing - Real Python
• Python Unit Testing - Real Python
• Unit Testing in Python - DataCamp

By following this guide and incorporating doctest into your development workflow, you can ensure that your code is well-documented, tested, and reliable. This foundational practice will help you and your team produce high-quality, maintainable machine learning projects.