Syed Saadullah Shah

No project description available.

No project description available.

An end-to-end Computer Vision application that trains a custom PyTorch Neural Network (MLP) on the MNIST dataset and deploys it as a live, interactive web app via Streamlit. Users can draw digits (0–9) on a digital canvas, and the model processes the image matrix in real-time to generate instant predictions with class confidence percentages.

I just deployed an end-to-end Machine Learning web app that predicts whether a breast tumor is Malignant or Benign with 94-96% accuracy! Instead of forcing users to input 30 complicated medical values, I optimized the model down to the top 5 core features (like size and texture). This made the app fast and simple to use without losing accuracy.

Heart Stroke Risk Prediction System: An end-to-end Machine Learning web application utilizing Logistic Regression to assess clinical stroke probability. Developed with Scikit-Learn and deployed via Streamlit.

A Machine Learning Web App that predicts the winner of Asia cup cricket matches using historical data. Built with "Random Forest Classifier", Python and deployed with Streamlit