Search⌘ K
AI Features
Home
Courses
Reliable Machine Learning

Why Machine Learning is Specific

Explore why machine learning software is unique in terms of testing challenges and defect detection. Understand common ML-specific bugs such as mismatched preprocessing, numerical errors, and data drift to improve the reliability and performance of your ML models.

We'll cover the following...

ML-specific software

In regular software, bugs tend to be more explicit. In ML software, defects are much easier to hide. Let’s consider the code snippet below, where we have a simple function from a computer vision pipeline.

Python 3.8
def extract_angle_from_warp_matrix(rotation_matrix: np.ndarray) -> np.float:
    # Extracts rotation angle (in degrees) from warp matrix
    return np.rad2deg(np.arctan2(-rotation_matrix[1, 0], rotation_matrix[1, 1]))

Even code authors (people who probably understand the domain) might have a hard time ...