Food For Thought

AI chatbot for restaurants

Goal:

To create an AI Chatbot Application that uses Retrieval Augumented Generation (RAG) in Ollama Langflow (GUI implementation of Langchain) to manage restaurant FAQs

Setup/Installation:

Update python through homebrew (for macOS):

  Install home brew:

   /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install.sh)"

  Update homebrew:

   brew update

  Update python:

   brew upgrade python

Download langflow:

  python3 -m pip install langflow -U

Run langflow:

  python3 -m langflow run

Download Ollama (for macOS):

  https://ollama.com/

Start Ollama in the background:

  ollama serve

Use the latest version:

  Ollama pull llama2

Upload the Entity Relationship Diagram (ERD):

  Use the JSON file to create playground

Connect a vectorized database:

  Create an astradb database on datastax:

   https://www.datastax.com/products/datastax-astra

  Details:

   database type: serverless (vector)
   server: aws/gcp server
   name: langflow_db
   region: us-east-2/us-east1

Feeding data:

  Copy the API endpoint and application tokens and paste in the corresponding fields in ERD

Run the program:

  Use the play button to build the program
  Click on playground button to launch chatbot

Dependencies:

  Langflow
  Ollama

Python version:

  3.10

Next Steps:

  Implement program in langchain