Machine Learning Teach by Doing: Day 9

Today, I recorded a 1 hour lecture on running your first ML algorithm: the Random linear classifier in Python.

We do this project right from scratch. It is meant for beginners. We take our learnings from the foundational lectures, and write code in Python.?

It is very satisfying to see theory come to life through code. I believe this combination of theory + code is the best way to teach ML.

I do not leave any concept unexplained. I write down equations wherever it is needed.?

Here is what we learn in this lecture:

(1) How to define the cats-dogs dataset in Python?

(2) How to visualise the dataset?

(3) How to write the function for the random linear classifier algorithm?

(4) How to visualise our results?

(5) How to split the data into training and testing?

(6) How to evaluate our hypothesis based on testing data?

(7) How to implement cross validation in Python?

(8) How to use cross validation to evaluate our learning algorithm?

(9) Comparison results with and without cross validation

(10) Best way to learn and master any coding language (including Python)

Below is the video I recorded:

This is the most comprehensive lecture so far and is packed with useful content. I had a lot of fun recording this.

Hope you have fun watching it too!

My Lecture Notes and code files have been added in the Youtube video information.

Stay tuned for Day 10!