An AI-powered web application built with Streamlit that recognizes handwritten digits (0–9) using a Convolutional Neural Network (CNN) / Neural Network model trained on the MNIST dataset.
Handwritten-Prediction.webm
- Recognizes handwritten digits (0–9).
- Interactive drawing canvas to write digits directly.
- Option to upload digit images for prediction.
- Powered by a trained MNIST deep learning model.
- User-friendly Streamlit interface.
- Includes an About Me sidebar with portfolio links.
- Open the app in your browser.
- Draw a digit (0–9) in the canvas OR upload a digit image.
- Click Predict.
- The model will display the recognized digit.
The model is trained on the famous MNIST dataset:
-
Training Data: 60,000 images
-
Test Data: 10,000 images
-
Image Dimensions: 28×28 pixels, grayscale
-
Classes (Labels):
- Digits:
0, 1, 2, 3, 4, 5, 6, 7, 8, 9
- Digits:
- Python 3.9+
- Streamlit (Frontend Web App)
- NumPy & Pandas (Data Processing)
- Matplotlib & Seaborn (Visualization & Confusion Matrix)
- OpenCV (Image Processing)
- TensorFlow / Keras (Deep Learning Model)
Mirza Yasir Abdullah Baig
This project is for educational purposes only.
It is not intended for commercial use but as a demonstration of deep learning in computer vision.