Machine Learning

Heart Disease Prediction Model

I built a simple yet powerful application that predicts the likelihood of heart disease based on user health parameters. The app is powered by a trained ML model and provides an easy-to-use interface for quick predictions.

Features

• Interactive form for patient metrics (age, sex, blood pressure, cholesterol, etc.)
• Client-side and server-side input validation
• Offline-trained ML model for probability/class risk prediction
• Responsive UI for desktop and mobile
• Prediction details with recommended next steps

Tech Stack

• Frontend: HTML, CSS, JavaScript
• Backend: Python (Flask)
• Machine Learning: scikit-learn, joblib
• Deployment: Render

Deep Learning

AI Text Summarizer

A simple AI-powered web application that generates concise summaries from long text inputs.

Built with Streamlit and my trained model on Hugging Face.

Key Features

• Summarize long text into paragraphs or bullet points
• Choose custom number of bullet points
• Adjust summary length with a slider
• Loading spinner for real-time feedback
• Clean, responsive UI powered by Streamlit

Tech Stack

• Frontend: Streamlit
• Backend/ML Model: Hugging Face Transformers, PyTorch
• Data Handling: NumPy, Pandas
• Deployment: Streamlit Cloud