CNN Sudoku Workshop

Welcome to the CNN Sudoku Workshop ! This repository is designed for ML enthusiasts looking to learn Convolutional Neural Networks (CNNs) through Sudoku . While Sudoku can be easily solved using automated reasoning approaches (e.g., SAT solvers), it also provides a great playground for exploring CNNs in action . Dive in, experiment, and have fun!

📌 Dataset

The dataset used for training and testing the CNN model is available on Kaggle: Sudoku CSV Dataset

---

📚 Modules

Module 1: Training and Visualizing

In this module, you will learn how to:

• Load data to train a Convolutional Neural Network (CNN) in TensorFlow to solve Sudoku puzzles
• Split the data into training and testing datasets
• Create a CNN model using TensorFlow
• Implement an early stopping callback to stop training when a threshold is reached
• Visualize accuracy and loss during training
• Use the model to run predictions on the test dataset
• Visualize predictions

---

Module 2: Optimization

In Module 1 , you built a CNN to solve Sudoku puzzles. However, the model's performance wasn't the best. Now, it's time to improve!

In this module, you will learn:

• Different strategies to optimize the CNN model from Module 1 to improve accuracy
• How to clean up the notebook by moving utility functions out of the way

---

🛠 Requirements

Make sure you have the following dependencies installed:

pip install tensorflow numpy pandas matplotlib seaborn

Alternatively, you can use the requirements.txt file:

pip install -r requirements.txt

🚀 Getting Started

Clone the repository and navigate to the project directory:

git clone https://github.com/marcilio/cnn_sudoku_workshop.git
cd cnn_sudoku_workshop

Start with Module 1:

cd module1
jupyter notebook

🤝 Contributing

Contributions are welcome! Feel free to open an issue or submit a pull request.

⭐ Like this repo?

If you found this repository useful, give it a star ⭐ ! Your support is appreciated!

Happy coding! 🎉