SuSastho.AI Chatbot

A trusted companion for better health!

About The Project

Welcome to the official repository for our Adolescent Health Chatbot, a cutting-edge tool designed to support adolescents in navigating the complexities of Sexual, Reproductive, and Mental Health (SRMH). This chatbot, developed using the latest advancements in AI and natural language processing, aims to provide timely, accurate, and confidential assistance.

Our mission is to leverage the power of artificial intelligence to bridge the gap in adolescent health education and services, particularly in areas often hindered by cultural and societal barriers. We are committed to improving health literacy and empowering adolescents to make informed decisions about their health.

Getting Started

To run the chatbot in you local computer please follow the steps below. We use GPU to run the embedding model. If you want to run it with CPU please replace "cuda" with "cpu" in code. We provided a subset of knowledge base of the chatbot in src/data folder. To get full data please contact us.

Prerequisites

Please install the following prerequisites before proceeding.

• Pytorch
• CUDA
• Nvidia GPU

Installation

Please follow the instructions below to run SuSastho.AI Chatbot.

1. Clone the repo

   https://github.com/BanglaGPT/SuSastho.AI.git

2. Install required packages

   pip install -r requirements.txt

3. Deploy Llama3 adapter weight provided in src/models/llama3-instruct-lora-adapter on Fireworks. Visit https://docs.fireworks.ai for more info

4. Get Fireworks API Key at https://fireworks.ai

5. Enter your API Key and Model Path in src/.env file

   FIREWORKS_MODEL_PATH = "<Your Deployed Model Path>"
   FIREWORKS_API_KEY = "<Your API Key>"

6. Run Backend Server

   cd src
   python app.py

7. Run frontend UI

   cd src
   streamlit run ui.py

License

Distributed under the MIT License. See LICENSE.txt for more information.