Medical-Chatbot-LLM

• git clone the repository
• create a new env

conda create -n chatbot-llm python=3.10 -y
conda activate chatbot-llm

• create template.py and run the file -> project structure

• define setup.py

• define requirements.txt and install required packages

• Data Downloaded

  • our data is in pdf format and stored in data directory

• we will be downloading Llama 2 model from huggingface to our local disk

  • Download Link
  • We will choose: llama-2-7b-chat.ggmlv3.q4_0.bin

• runing jypyer notebook for experimental purposes

• utils.py

  • create src/utils.py
  • add all the required functions
    • function to extract data from pdf files
    • function to create text chunks
    • function to downlaod embedding model

• define store_index.py

  • Pinecone Section

• Define Prompt Template

• Define project UI

  • add templates/chat.html
  • add static/style.css

• define app.py