This project implements a Convolutional Neural Network (CNN) using PyTorch to classify handwritten digits from the MNIST dataset. The model achieves ~99% accuracy on the test set and can be used to predict digits from custom images.

• Installation
• Project Structure
• Results
• License

1. Clone the repository:

   git clone https://github.com/your-username/mnist-digit-recognition.git cd mnist-digit-recognition 

2. Install dependencies:

   pip install -r requirements.txt 

There might be some unnecessary dependencies (I forgot to to create Virtual Envinronment)

mnist-digit-recognition/
├── .gitignore # Specifies files to ignore
├── LICENSE # MIT License
├── README.md # Project overview
├── requirements.txt # Python dependencies
├── images/ # Example images
├── errors/ # Misclassified images
├── notebooks/ # The Notebook file
├── model.py # The current model
├── predict.py # Predicts given image of a digit
├── test.py # Tests the model accuracy
├── utils.py # For image presentation
├── train.py # The code of training the model

• Test Accuracy: 98.2%

This project is licensed under the MIT License. See LICENSE for details.

• The MNIST dataset for providing the handwritten digit images.

• PyTorch for the deep learning framework.

For questions or feedback, feel free to reach out:

• Diyorbek Umaraliyev: d.umaraliyev@newuu.uz
• GitHub: @umaraliyev0101
• Telegram: @umaraliyev0101

1. Clear Structure: Added a table of contents for easy navigation.
2. Visuals: Included placeholders for images (e.g., MNIST examples, prediction results).
3. Detailed Usage: Explained how to train, test, and predict with the model.
4. Project Structure: Added a folder structure overview.
5. Acknowledgments: Gave credit to the MNIST dataset and PyTorch.
6. Contact Info: Added a section for questions or feedback.