fastapi-app-with-langchain

This project is purposed to create RAG(retriever-augmented-generation) chatbot application using fastapi and docker.

If you upload .pdf file, RAG chatbot performs Q&A based on your document.

This project heavily referenced @AshishSinha5’s rag_api project.

Table of Content

• Getting Started
• Usage
  • Starting the local server
  • Deploying the server

Getting Started

1. Clone this repository to your local machine.

   git clone https://github.com/jodog0412/langchain-app-with-fastapi.git
   cd langchain-app-with-fastapi

2. Create a virtual environment in your local directory and activate it.

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

3. Install Python packages in your virtual environment.

   pip install -r requirements.txt
   

4. Set OPEN_AI_API_KEY on .env file in the app directory.

   # app/.env
   OPENAI_API_KEY =
   

Usage

Starting the local server

cd app
uvicorn main:app --reload

Deploying the server

1. Set OPEN_AI_API_KEY on Dockerfile.

   FROM python:3.10
   
   WORKDIR /code
   
   COPY . /code
   
   RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
   
   ENV OPENAI_API_KEY "ENTER YOUR OPENAI API KEY"
   
   WORKDIR /code/app
   
   CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"]
   

2. Build docker-image.

   docker build --pull --rm -f "Dockerfile" -t langchainappwithfastapi:latest "." 

3. Create docker-container.

   docker run -d --name container -p 8000:8000 langchainappwithfastapi

Features/Update

• build history-aware RAG chatbot
• implement client-server RESTful-API
• implement websocket connection
• deploy the app
• bug fixes(deployed app has some errors)
• support local LLM(only supports GPT model in current version)