Medical Chatbot

Installation

Installation of Dependencies

pip install -r requirements.txt

Installing the model

Download GGML format model files for Meta Llama 2's 8-bit quantized Llama 2 7B Chat model from here. Keep the filename as lama-2-7b-chat.ggmlv3.q8_0.bin and save at the root of this directory.

Customizing the database

Inside the 'data/' folder, put in the desired PDF(s). These will serve as your datasets.

Generation of Vector Database

python ingest.py

Running this script will generate a vector database

Execution

chainlit run model.py -w

References

[1] https://github.com/AIAnytime/Chat-with-PDF-Chatbot

[2] https://github.com/AIAnytime/Llama2-Medical-Chatbot