Codemium

Description

This system leverages a Large Language Model (LLM) to take a code repository into its context. It can understand and analyze the repository's structure, code, and documentation, providing accurate and relevant answers to technical questions. Whether you need help understanding a particular function, debugging an issue, or getting insights into the overall architecture, this tool is designed to assist developers in navigating complex codebases efficiently.

Getting Started

Dependencies

Python 3.10 or higher
Langchain
FAISS
Boto3 (for Amazon Web Services)
streamlit
Aws bedrock

Please refer to the requirements.txt file for a comprehensive list of dependencies.

Installing

Clone the repository and install the dependencies using pip:

git clone https://github.com/tawhidii/codemium.git
cd codemium
pip install -r requirements.txt

Project Structure

The project is structured as follows:

codemium/
│
├── .DS_Store
├── requirements.txt
├── README.md
├── .gitignore
├── .env.example
├── main.py
│
├── core/
│   ├── __init__.py
│   ├── codemium_engine.py
│   └── __pycache__/
│       ├── __init__.cpython-312.pyc
│       └── codemium_engine.cpython-312.pyc
│
└── utils/
    ├── ui.py
    ├── __init__.py
    └── __pycache__/
        ├── ui.cpython-312.pyc
        └── __init__.cpython-312.pyc

Executing program

Run the main program:

streamlit run main.py