TinyGPT

A lightweight AI text generator based on distilgpt2. You can generate text using a pre-trained model and fine-tune it on your own dataset.

---

Installation

Clone the Repository

git clone https://github.com/yourusername/TinyGPT.git
cd TinyGPT

Create and Activate a Virtual Environment

python3 -m venv ai_env
source ai_env/bin/activate  # on mac/linux
ai_env\Scripts\activate    # on windows powershell

Install Dependencies

pip install -r requirements.txt

---

Usage

Generating Text

Run TinyGPT with a prompt:

python src/tinygpt.py

Modify the prompt inside src/tinygpt.py to experiment with different outputs.

Fine-Tuning the Model

If you want to fine-tune the model on your own dataset (data/data.txt):

python src/fine_tune.py

This will train the model and save it in models/fine_tuned_model/

Running Fine-Tuned Model

After fine-tuning, update tinygpt.py to load the fine-tuned model:

# change model path to fine-tuned version
generator = pipeline("text-generation", model="../models/fine_tuned_model")

Then, run:

python src/tinygpt.py

---

Notes

• The data/data.txt file should contain training examples in plain text format.
• Model weights are not included here but will be downloaded automatically.