Skip to content

This repository is full of handmade Machine Learning algorithms that I have coded out after understanding the intuition from the book from Jason Brownlee.

Notifications You must be signed in to change notification settings

AayushSameerShah/Manual-ML

Repository files navigation

Manual ML

I know it is a bit awkward to call it aslearn which is my name "Aayush Shah learn" and which also sounds like sklearn: a library of machine learning.

🎒 Background

So I completed learning Jake Venderplas's book: Python Data Science Handbook and also completed Jose's course on Data Analysis its when I implemented codes, ran models, did vizzes... but still there was something which was lacking; my understanding of models: How Do They Work.

Now of course, when I tried to think about the models' internals I freak out. I see the formulae, functions, derivations etc. As being a student of non-mathematical background it does. But still I had to learn stuff. I had to know why the model is working. What makes it to learn. I wanted to see the nuts and bolts of the model so it doesn't stay the black box for me. Then, I don't know somewhere while I was reading some article, I accidently reached to this amazing website which led me to read the book and start my data science journey with confidence it is Jason Brownlee's machinelearningmastery.com.

ℹ️ About

This repository is based on the book: Master Machine Learning Algorithms. This is truly amazing book which teaches you more than enough to get you started with the internals of the models. It covers various models which I have covered and implemented in this repository (yes, this one).

🔠 2 Words on Book

The book is focused more in explaining the theory first and then implement each model from scratch in Excel. Theory and implementation have their own separate chapters. As said, book implements the work in Excel sheet which is much abstract, so I had to figure out the "hows" in python.

📖 How to read

I have implemented many chapters, books, lectures, courses in my github and from there I have learnt a single most common thing from my mistakes—keep things simple, comprehensive and short.

This time, I have followed that strategy here. I have explained stuff where I needed to and had commented code for further understanding. The repository is divided into different kinds of models and then they are divided into subfolders accordingly trying to maximize the search time whenever you come in.

🎶 Tone

See, I am a learner. I make things for keeping me first. But I have tried to keep it for anyone. You can put yourself in the story and continue reading this book. Have kept the human language there (avoiding the jargons) and compiled into my own words, it shouldn't sound difficult to you.

Especially it is for me to refer some topics when I want to have a look in any point of time in the future the same thing you can as well! And the tone is friendly so don't worry about the some explicit words there!

Okay then, I hope you find my work useful. Much things are explained when you look inside. I would suggest you to take it as a journey and start with Welcome Me.md file and end with Amazing Journey.md file.

Have a Great Learning! Aayush Shah

About

This repository is full of handmade Machine Learning algorithms that I have coded out after understanding the intuition from the book from Jason Brownlee.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published