Chatbot

This repo consist of RAG implementaion to OpenAI model for your private chatbot. The pdf files are stored in Azure storage which are retrived by this app and stored as vector embedding in Pinecone database.

Instruction for use

• clone this repo

  git clone https://github.com/sangamdeuja/RAG-LLMchatbot.git
  

• create.env file inside root folder and provide the following credentials

  STORAGE_ACCOUNT_NAME=your_azure_storage_account_name
  

  STORAGE_CONTAINER_NAME=your_container
  

  CONN_STRING=your_storage_connection_string
  

  PINECONE_API_KEY=your_pinecone_api_key
  

  INDEX_NAME=your_index_name
  

  OPENAI_API_KEY=your_openai_api_key
  

• create pinecone account and index in pincone.io and provide detail in above .env file
• create azure storage and container and provide and provide detail in above .env file
• create openai api key and and provide detail in above .env file
• (Optional) If you want to run without docker create virtual env, install packages from requrirements.txt and run the following

  streamlit run main.py
  

• To run in docker, run the following code

  docker build -t chatbot .
  

  docker run -p 8501:8501 chatbot