Diabetes and Heart Disease Risk Prediction using Logistic Regression from Scratch

This project implements logistic regression from scratch using NumPy to predict diabetes and 10-year risk of heart disease based on patient health data.

---

Dataset

• File: diabData.csv
• Columns include:
  • male – Gender (0 = Female, 1 = Male)
  • age – Age in years
  • education – Education level
  • currentSmoker – Whether currently smoking (0/1)
  • cigsPerDay – Number of cigarettes per day
  • BPMeds – On blood pressure medication (0/1)
  • prevalentStroke – History of stroke (0/1)
  • prevalentHyp – History of hypertension (0/1)
  • diabetes – Diabetes status (0/1)
  • totChol – Total cholesterol
  • sysBP – Systolic blood pressure
  • diaBP – Diastolic blood pressure
  • BMI – Body mass index
  • heartRate – Heart rate
  • glucose – Glucose level
  • TenYearCHD – 10-year risk of coronary heart disease (0/1)

---

Project Overview

• Implemented gradient descent manually to train logistic regression.
• Computes weights and bias without using sklearn’s LogisticRegression.
• Includes:
  • Sigmoid function
  • Gradient calculation
  • Accuracy computation
  • Predicting for new user input

---

How to Run

1. Clone or download this repository.
2. Open diabetes.ipynb in Jupyter Notebook.
3. Run all cells in order:
   • Load and scale data
   • Train the model
   • Test accuracy
   • Predict new input

---