PyTorch MNIST Neuronal Network Classifier

A project demonstrating how to train a Neuronal Network for MNIST digit recognition and optimize its hyperparameters using PyTorch and Optuna.

Table of Contents

1. About The Project
   • Built With
2. Getting Started
3. Usage
4. Contributing
5. License

About The Project

This project provides a comprehensive walkthrough for training a Neural Network on the classic MNIST handwritten digit dataset. The entire workflow is self-contained in a Google Colab notebook for easy access and execution.

The primary focus is not only on building a model with PyTorch but also on leveraging Optuna to systematically and efficiently search for the best hyperparameters to maximize model accuracy. It serves as a practical template for applying automated hyperparameter tuning to deep learning projects.

Built With

This project was built with the following technologies:

• Python

• PyTorch

• Optuna

• Google Colab

Getting Started

This project is designed to be run entirely in Google Colab, so there is no local setup required.

Usage

After running the notebook, you will see the following process unfold:

1. Dependency Installation: The notebook will first install pytorch and optuna.

2. Data Preparation: The MNIST dataset is downloaded and prepared for training and validation.

3. Optuna Study: The hyperparameter search begins. Optuna will run multiple trials, each time training the CNN with a different set of hyperparameters (learning rate, optimizer, etc.).

4. Live Results: The output of each trial, including the validation accuracy, will be printed as the study progresses.

5. Final Report: At the end of the study, Optuna will report the best combination of hyperparameters found and the highest accuracy achieved.

This provides a hands-on example of how to automate the tedious process of hyperparameter tuning.

Contributing

Contributions are what make the open source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.

If you have a suggestion that would make this better, please fork the repo and create a pull request. You can also simply open an issue with the tag "enhancement". Don't forget to give the project a star! Thanks again!

1. Fork the Project

2. Create your Feature Branch (git checkout -b feature/AmazingFeature)

3. Commit your Changes (git commit -m 'Add some AmazingFeature')

4. Push to the Branch (git push origin feature/AmazingFeature)

5. Open a Pull Request

License

This project does not have a license. Feel free to fork, modify, and use it as you see fit.