Just Mix, Match, and Layer! It’s not your Wardrobe, it’s your Style.

Neural Networks Made Easy : A Hands-On Introduction Part 3

“Give a girl the right shoes, and she can conquer the world.” — Marilyn Monroe. I’m going to start off Part 3 with that. How this quote and the title relate to an article on neural networks will be obvious by the end of it. I do hope that the person reading this has already read parts 1 and 2 of my deep learning series. The dreary basics are finally over so we’ll be touching the iceberg. And as for shoes, I always wear the right ones.

Photo by MARIOLA GROBELSKA on Unsplash

We talked about weights, and we’re going to talk more about weights.

1. Weight in a Simple Linear Regression

Suppose we have one feature (e.g., house size in square feet) and we want to predict house price. Then we use a single equation to do so:

• X is the input (house size)
• W is the weight (a number that determines how much house size affects price)
• B is the bias (a fixed number)
• y is the predicted output (house price)

2. Weights in a Multiple Linear Regression