Download as txt, pdf, or txt
Download as txt, pdf, or txt
You are on page 1of 1

Hello everyone, hope you all are doing well.

This series will cover some machine


learning
algorithms from scratch to understand how things are working under the hood when we
simply call an ML algorithm API
from a library.

Why ML from scratch?


1 - This way we can do more novel things or get more better results even with more
basic models
2 - When solve famous ML problems often it is seen that we just import existing
libraries and use preplanned
networks/algorithms which we know works for specific dataset. But that means
you probably fall flat on face when you
try to solve something which someone hasn't already solved for you.
3 - If you really want to know what happens with data as it comes into your
training phase or while moving through neurons
in case of neural networks, how loss is calculated, how optimizers fit in and
most importantly how to build models
that do new or novel things.

What - in this series and after that ?


1 - We are going to cover some important concepts in machine learning and how
different elements work together to give
solve ML problem and give better results
2 - We would explain and implement these machine learning algorithms from scratch.
They will be implemented using python.
Only numpy would be used for numerical computations.
3 - After this series, a similar series would be done for deep learning as well,
where we would implement neural networks
from scratch.

Pre requisites -
1 - Basic understanding of Python.

Contents -
1 - What is Machine learning and why do we use it?
2 - Parametric and non-parametric ML algorithms
3 - Bias, Variance and the trade of
4 - Linear regression algorithm from scratch
5 - Logistic regression algorithm from scratch
6 - Linear discriminant analysis algorithm from scratch
7 - Classification and Regression trees from scratch
8 - Naive Bayes algorithm from scratch
9 - K-Nearest Neighbours algorithm from scratch
10 - Support vector machine algorithm from scratch
11 - Bagging and random forest algorithm from scratch
12 - Boosting and AdaBoost from scratch

So let us meet at the first section of this series. Thank you.

You might also like