This project implements a simple digit recognition application using TensorFlow/Keras for the machine learning model and Tkinter for the graphical user interface. It allows users to draw digits on a canvas, and the application predicts the drawn digit using a pre-trained model.

• Python 3.x (recommended 3.7 or higher)
• pip (Python package installer)

1. Create a virtual environment (recommended):

   python3 -m venv venv  # Creates a virtual environment named "venv"

2. Activate the virtual environment:

   • Linux/macOS:

     source venv/bin/activate

   • Windows:

     venv\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

1. Run the training script:

   python train.py

   This script (train.py) will:

   • Load the MNIST dataset.
   • Define and train a Convolutional Neural Network (CNN) model.
   • Save the trained model to a file named mnist_cnn.keras.

1. Run the application script:

   python app.py

   This will launch the Digit Recognizer application.

1. Drawing: Use the mouse to draw a digit on the white canvas. Click and drag to draw.
2. Prediction: The application will predict the digit you've drawn and display the prediction and confidence level below the canvas.
3. Clearing: Click the "Clear" button to erase the canvas and start drawing a new digit.

Digit-Recognizer/
├── app.py        # The main application script (Tkinter GUI)
├── train.py      # The model training script
├── mnist_cnn.keras # The trained model file (saved by train.py)
├── venv/         # The virtual environment (if you created one)
└── README.md     # This file

The model used in this application is a simple CNN trained on the MNIST dataset. It consists of convolutional layers, max pooling layers, a flatten layer, and fully connected layers. The model is trained using the Adam optimizer and categorical cross-entropy loss.

• GPU Acceleration: If you have a compatible NVIDIA GPU, install the CUDA toolkit and the GPU version of TensorFlow (tensorflow) for faster training and prediction.
• Model Architecture: Experiment with different CNN architectures to potentially improve accuracy.
• Data Augmentation: Implement data augmentation during training to make the model more robust to variations in handwriting.
• User Interface Enhancements: Add more features to the UI, such as different brush sizes, colors, or the ability to load and save images.
• Deployment: Package the application for distribution to other users.

• This project uses the MNIST dataset, which is a widely used dataset for digit recognition.
• The TensorFlow/Keras library is used for building and training the machine learning model.
• The Tkinter library is used for creating the graphical user interface.
• The Pillow and OpenCV libraries are used for image manipulation and processing.