MiniChatGPT

MiniChatGPT is a simple chatbot application built with Flask and Ollama, designed to simulate a chat interface similar to ChatGPT. The application includes a web interface for user interaction and a backend API to handle chat messages.

---

Features

• Chat Interface: A web-based chat interface built with HTML, CSS, and Bootstrap.
• API Integration: Communicates with an API to generate chat responses.
• Testing: Comprehensive tests using pytest to ensure the functionality of the chat endpoints.
• CI/CD Pipeline: Automated linting, testing, and Docker image building using GitHub Actions.

Project Structure

• web/: Contains the Flask application code.
  • api.py: The main API endpoints for the chat application.
  • templates/: HTML templates for the web interface.
  • static/: Static files like CSS and JavaScript.
• tests/: Contains test cases for the application.
  • test_app.py: Test cases for the Flask application endpoints.
• .github/workflows/ci.yml: GitHub Actions configuration for CI/CD pipeline.
• requirements.txt: Python dependencies for the project.
• README.md: Project documentation.

Requirements

• Docker and Docker Compose
• Python venv
• Ubuntu 24.04.1 LTS

Installation

1. Clone the repository:

git clone git@github.com:MlCHAL-S/MiniChatGPT.git MiniChatGPT
cd MiniChatGPT

2. Create a virtual environment and activate it:

python -m venv .venv
source .venv/bin/activate

3. Install the dependencies:

pip install -r requirements.txt

Usage

Within the root directory run:

docker-compose up --build

Then visit http://localhost:5000/ in your browser.

Run tests

pytest