Understanding Machine Learning: A Guide for data engineer

Photo by Denis Volkov on Unsplash

Machine learning sometime feels like magic. How can a computer recognize objects in an image or a car drive itself?

These capabilities can baffle not only the layman but also many seasoned software developers.

Despite years of coding experience, machine learning (ML) remains a foreign, intimidating, and intriguing field for many.

This guide aims to demystify ML for developers, presenting the fundamentals in a language they understand.

Machine learning is a vast field, and a single start can’t cover everything. This guide focuses on three key areas:

1. Supervised Learning
2. Neural Networks
3. Deep Learning

Part I: Supervised Learning

Supervised learning is a prevalent type of ML. Unlike other ML flavors, it involves training a model on labeled data. For example, teaching a computer to recognize handwritten digits using images labeled with the correct digit.

When to Use:

• Classification tasks: Identifying spam emails, recognizing faces, diagnosing diseases from medical images.
• Regression tasks: Predicting house prices, forecasting stock prices, estimating life…