Makemore_Shakespeare

This repository contains an implementation of a character-level language model inspired by Andrej Karpathy's YouTube series. The model is trained to generate text in the style of Shakespeare using the tiny_shakespeare.txt dataset.

Contents

• Mini_GPT.ipynb: A Jupyter notebook containing the complete implementation of the character-level language model. This notebook includes the training process, model architecture, and text generation.

• tiny_shakespeare.txt: The dataset used for training, containing a collection of Shakespeare's works in plain text.

Requirements

To run the notebook, you will need the following dependencies:

• Python 3.x
• PyTorch

Getting Started

1. Clone the repository:

   git clone https://github.com/Daimon5/Mini_GPT.git

2. Run the notebook cells to train the model and generate text.

How to Use

• Training the Model:

  • Follow the steps in the notebook to train the character-level language model on the tiny_shakespeare.txt dataset.
  • You can adjust hyperparameters within the notebook for different results.

• Generating Text:

  • After training, use the notebook to generate Shakespeare-like text.
  • The generated text will imitate the style of Shakespeare but may contain imperfections due to the simplicity of the model.

References

• Andrej Karpathy's YouTube Series
• Original Char-RNN Paper
• PyTorch Documentation

License

This project is licensed under the MIT License - see the LICENSE file for details.