avik-jain/100-days-of-ml-code

100 Days of ML Code is a public learning journal where the author documents their daily progress learning machine learning over one hundred consecutive days. Each entry covers a topic studied that day, links to code implementations, and often includes a hand-drawn infographic that summarizes the concept visually. The repository serves as both a personal accountability log and a reference for others starting their own machine learning journey. The topics covered follow a structured progression from foundational concepts to more advanced ones. Early days focus on data preprocessing, which means cleaning and preparing raw data before feeding it to an algorithm, then move into core algorithms like linear regression, which predicts numeric values from input data, logistic regression for classification tasks, K-Nearest Neighbours, Support Vector Machines, and decision trees. Later entries cover deep learning basics, with the author working through Coursera's deep learning specialization, and also brush up on the underlying mathematics including linear algebra and calculus, which are the foundations most machine learning algorithms rely on. Code implementations are written in Python using the scikit-learn library. You would find this repository useful if you are just starting to learn machine learning and want a roadmap of topics to study in a logical order, or if you want simple Python code examples for classic algorithms alongside visual explainers. It is not a polished course or textbook, but rather an honest day-by-day record of one person learning. There is no single tech stack for the repo itself since it is primarily documentation and Jupyter notebooks, but the implementations use Python, scikit-learn, and standard data science libraries.