Python Code Completion Model (GPT-2)

This project contains a web-based interface for a Python code completion model developed using HuggingFace. The model is based on GPT-2 and trained on the semeru/code-code-CodeCompletion-TokenLevel-Python dataset. Users can interact with the model via a Streamlit web interface.

Project Overview

This project aims to develop a model for automatic Python code completion. The GPT-2 model used here has been optimized to understand and complete Python code. The model analyzes the code written by the user and suggests relevant completions.

Features

Model: GPT-2 based, Python code completion
Dataset: semeru/code-code-CodeCompletion-TokenLevel-Python
Web Interface: Streamlit
Platform: HuggingFace
Code Completion: Automatic Python code suggestions

Installation

1.Clone the repository:

git clone https://github.com/your_username/code-completion-model.git cd code-completion-model

2.Install the required dependencies:

pip install -r requirements.txt

3.To upload your model to HuggingFace, you can use transformers-cli:

transformers-cli login transformers-cli upload ./path_to_your_model

4.Start the Streamlit app:

streamlit run app.py

Contributers

Talha Dağlayan - talhad1109@hotmail.com Ramiz Can Akbıyık - ramizcanakbiyik@gmail.com

TalhaDaglayan
/

CodeCompletionModel