Heart Disease Prediction: Logistic Regression & ADAM from Scratch

Reasearch wtih EDA

This project is an educational implementation of logistic regression (binary and multi-class) and the ADAM optimization algorithm from scratch, without using machine learning libraries like Scikit-Learn or TensorFlow. The goal is to understand the fundamental mathematics and coding required to build these models from the ground up.

Features

• Binary Logistic Regression: Classifies whether a patient has heart disease or not.
• Multi-Class Logistic Regression: Extends logistic regression to handle multiple classes.
• ADAM Optimization: Implements the ADAM optimizer for efficient gradient-based optimization.
• Dataset Handling: Prepares and processes heart disease data for training and evaluation.