Marta-GPT v0.0.1

A basic personal assistant/chatbot that can instantly answer your questions and complete tasks within your documents, It was developed using Large Language Models (LLMs), LangChain, and Retrieval-Augmented Generation (RAG).

Langchain Docs RAG Architecture:

💻Installation guide:

Clone the repository

git@github.com:cousintiz/Marta-GPT-v0.0.1.git

If you have python3.10 intalled run this shell script otherwise follow the steps below:

./setup.sh

Create a virtual environment for the project's dependencies:

python3 -m venv <environment-name>

Ativacte the virtual environment:

<environment-name>\Scripts\Activate.ps1

Install the packages on the requirements.txt:

pip3 install -r requirements.txt

Usage

After all the steps above just run the App:

./app.sh

• Get an OpenAI API key here
• Submit an text file document with the relevant information to feed into the model once you run the app.

This window window will be opened on your browser

⚒️Development Stack

• Langchain
• Streamlit
• GPT 3.5-Turbo

⚙️Contribution

You are allowed use any part of the software or ideas found on this repository for anything, feel free to contribute, with improvements, issues or whatever you find useful...