Custom Chatbot using LangChain, OpenAI and Chainlit

You can read further here, here and here.

Overview

Overview

This repo contains the code for Scoopsie, a custom chatbot that answers ice-cream-related questions and fetches information from a fictional ice-cream store's API. Using LangChain and OpenAI's text model, alongside a Flask web service, Scoopsie can provide users with details on flavors, toppings, and store offers. Everything is put together with Chainlit for easy web application integration.

Getting Started

Prerequisites

• Python 3.8 or later
• An OpenAI API key

Installation

1. Clone the Repository

   git clone git@github.com:tahreemrasul/simple_chatbot_langchain.git
   cd ./simple_chatbot_langchain
   

2. Set Up a Conda Environment (Recommended)

• If you don't have Conda, install it first.
• Create a new Conda environment:

  conda create -n chatbot_langchain python=3.8

• Activate the environment:

  conda activate chatbot_langchain
  

3. Install Dependencies

• Install the required packages using the requirements.txt file:

  pip install -r requirements.txt
  

4. Set Up Your OpenAI API Key

• Create a .env file in the root directory of the project.
• Add your OpenAI API key to the .env file:

  OPENAI_API_KEY='Your-OpenAI-API-Key-Here'
  

Usage

To run the fictional store's API, execute the following command:

python ice_cream_store_app.py

The fictional store's API will be accessible at http://localhost:5000/{endpoint_name}

Make sure the fictional store's application is running before running the chatbot. To run Scoopsie, simply execute the chatbot.py script:

chainlit run chatbot.py -w --port 8000

To run the chatbot application, navigate to http://localhost:8000