5M+ views | Data Scientist & Top Tech Writer | YouTube: www.youtube.com/c/BetterDataScience
Photo by Sammy Williams on Unsplash

Machine learning is boring. Sure, there’s a lot of hype around it, but once you’ve trained dozens of models you’ll stop looking at machine learning through rose-tinted glasses. Building models is tedious and repetitive, at least for practitioners.

Does this ring a bell? You try out a bunch of models…

Photo by Luca Onniboni on Unsplash

There have never been more job openings for data science positions, but yet, it’s never been harder to get hired as a data scientist.

Why? The answer is simple — smaller companies usually don’t need an entire data science department, so they end up hiring a couple of seniors. That…

Photo by Jem Sahagun on Unsplash

The previous TensorFlow article showed you how to write convolutions from scratch in Numpy. Now it’s time to discuss pooling, a downscaling operation that usually follows a convolutional layer. You want to know a secret? It’s not rocket science to implement from scratch.

After reading, you’ll know what pooling and…

Photo by Grzegorz Walczak on Unsplash

Python’s equivalent of a switch statement is finally here in Python 3.10. It hides behind a fancy name of Structural Pattern Matching, and does what you’d expect and more. It packs support for working with primitive data types, sequences, and even classes and enums.

Today I’ll show you how to…

Photo by Ochir-Erdene Oyunmedeg on Unsplash

Convolutional networks are fun. You saw last week how they improve model performance when compared to vanilla artificial neural networks. But what a convolution actually does to an image? That’s what you’ll learn today.

After reading, you’ll know how to write your convolution function from scratch with Numpy. You’ll apply…

Photo by SpaceX on Unsplash

If I had to pick one Python library I couldn’t live without, it would be Pandas. Nothing else comes close. Does that mean Pandas is without flaws? Well, no — it’s terribly slow for processing large datasets. Also, working with larger-than-memory datasets is a challenge of its own. …

Photo by Markus Spiske on Unsplash

You saw last week that vanilla Artificial neural networks are terrible for classifying images. And that’s expected, as they have no idea about 2D relationships between pixels. That’s where convolutions come in — a go-to approach for finding patterns in image data.

Want to hear the good news? Today you’ll…

Photo by Anthony Riera on Unsplash

Managing data science workflows isn’t fun. There are usually dozens of virtual environments involved, and the whole thing is a mess. Seriously, how many times did the execution fail just because your Python environment was missing a dependency?

Picture this — you’ve created a new virtual environment for your project…

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store