https://www.shivasud.github.io/Shiva Sud's personal website for projects and blog posts. 2025-09-09T10:39:26+01:00 Shiva Sud https://www.shivasud.github.io/ Jekyll © 2025 Shiva Sud /assets/img/favicons/favicon.ico /assets/img/favicons/favicon-96x96.png 2025-08-09T00:00:00+01:00 2025-08-09T00:00:00+01:00 https://www.shivasud.github.io/posts/Spin-vs-Pace/ Shiva Sud

Theory In this article I want to create an index for batters to see how they perform against spin bowling and pace bowling, to see which batsmen perform the best. I will use different variables to quantify a batsman’s performance, and scale this value into a value between 0 and 100, which will be the batsman’s score. The first variable I will use is the batsman’s run rate, while facing pace v...

2025-08-09T00:00:00+01:00 2025-08-09T00:00:00+01:00 https://www.shivasud.github.io/posts/k-nearest-neighbours/ Shiva Sud

Theory Overview In this blog, I will be explaining the K Nearest Neighbours algorithm. This is a classification algorithm that can predict which class a new point belongs to. The way it works is by finding the Euclidean distance from the new point to each other points in the dataset. Then after taking the k closest points, you need to rank all of the labels of those points. The label that is...

2025-08-08T00:00:00+01:00 2025-08-08T00:00:00+01:00 https://www.shivasud.github.io/posts/perceptron-algorithm/ Shiva Sud

Theory Overview The perceptron algorithm, is a learning algorithm that separates two classes of data with a line such that it can classify a new piece of data into a class accordingly. Essentially, if we have the same equation as in the previous blog: \[z = \theta_1x_1 + \theta_2x_2 + ... + \theta_nx_n + c\] \[z = \vec{\theta}^\top\vec{x_i}+c\] then if after plugging in the input features,...

2025-08-07T00:00:00+01:00 2025-08-10T16:11:05+01:00 https://www.shivasud.github.io/posts/logistic-regression/ Shiva Sud

Theory Overview Logistic regression is a machine learning algorithm, which is used for binary classification, for example seeing if a student passed or failed an exam, given the amount of hours they spent studying for it. The reason why logistic regression has regression in the name, is because rather than predicting the class (whether the student will pass or not), it predicts the probability...

2025-07-29T00:00:00+01:00 2025-08-07T16:14:40+01:00 https://www.shivasud.github.io/posts/linear-regression/ Shiva Sud

This is the first blog of a long series, where I will explain how common machine learning algorithms work intuitively from scratch. I hope you enjoy! Theory Overview Linear regression is a type of supervised learning algorithm that models the relationship between one or more input variables (features) and a target variable, by calculating the line of best fit during the training process. This...