A simple AI model that recognizes handwritten digits (0β9) using deep learning. Trained on the MNIST dataset, this model is a classic example of computer vision and image classification using neural networks.
- Model Type: Neural Network
- Dataset: MNIST (1,000 train / 50 test)
- Accuracy: ~90%
- Input: 28x28 grayscale digit image
- Output: Digit class (0β9)
- Python
- Numpy
- MNIST Dataset
train.csvβ Train the modelrecognizer.ipynbβ Saved trained modeltest.csvβ Optional UI (Streamlit/Flask)
MIT License β Free to use and modify.