An open API service indexing awesome lists of open source software.

https://github.com/bounebrayan/luca

An interactive chatbot application that utilizes transfer learning on GPT-2, designed for engaging and amusing conversations.
https://github.com/bounebrayan/luca

ai chatbot conversation fine-tuning gpt-2 machine-learning python web-interface

Last synced: about 2 months ago
JSON representation

An interactive chatbot application that utilizes transfer learning on GPT-2, designed for engaging and amusing conversations.

Awesome Lists containing this project

README

          

# Luca

Luca is a chatbot powered by a GPT-2 machine learning model. Designed to entertain and engage, ChatMind features both frontend and backend components, enabling a seamless conversational experience.

## Features

- **Engaging and really amusing Conversations**: AI-driven responses provide an interactive chat experience.
- **GPT-2 Integration**: Built using transfer learning to enhance GPT-2 capabilities for conversations.
- **Seamless Functionality**: Fully integrated frontend and backend for a complete application.*

## Screenshots

![shot6](https://i.ibb.co/HpJq50p/241837883-8aaae84b-2df2-455f-8c0f-d10fda3d3685.png)
![shot7](https://i.ibb.co/RBLnWhW/241837917-68379919-2d85-4019-8b3a-531cd6ada584.png)

_Above: A glimpse of the funny and engaging responses I got while chatting with the model_

### Prerequisites
Ensure you have the following installed:
- **Node.js** and **npm** for the frontend.
- **Python** and **pip** for the backend.

## Backend Setup

-Navigate to the backend directory of the project and install the required **transformers** dependency
```bash
pip install transformers
```

- Download the GPT-2 model file from [this link](https://drive.google.com/file/d/1XYYWyIYbZfPRpO5FsCGwKL-HMp5arU01/view?usp=share_link) and place it in the backend folder with app.py.

## Running the Application
1- Start the frontend server

2- Start the backend server in a separate terminal

3-Open http://localhost:3000 in your browser to access the chatbot.