Linear Regression Introduction

In Day 21 of the ML: Teach by Doing Project, I revisited one of the foundational ML techniques: Regression.

In this lecture, we saw how to proceed with the 6 Steps for any ML framework, and how they relate to Regression:

(1) Collect Data

(2) Generate Hypothesis

(3) Define loss function

(4) Find learning algorithm through optimization

(5) Run the optimization and find best hypothesis

(6) Validate results

We finished steps 1, 2 and 3 in this lecture.?

The lecture also covers a very interesting section on a live case study we did with school teachers to teach them regression.?

This case study demonstrates how humans can intuitively come up with linear regression frameworks for easy, single variable datasets.

This provides a strong intuition for regression and why we need to rely on AI for complex data with more variables.

I made a video to explain all my learnings here:

My Lecture notes code files can be accessed here: Link

Stay tuned for Day 22!