Spaces:
Runtime error
HealsMindAI: AI-Powered Wellness Advisor
Welcome to HealsMindAI, an AI-powered wellness advisor that provides personalized healthcare insights using the power of Natural Language Processing (NLP) and open-source data. This repository contains the source code and resources for the HealsMindAI project.
Overview
HealsMindAI is designed to offer users personalized healthcare information, focusing on topics like yoga, natural remedies, and holistic wellness. The project leverages the capabilities of fine-tuned Large Language Models (LLMs) to generate human-like responses and engage users in meaningful conversations about their health-related inquiries.
Features
- AI-Powered Conversations: Engage in informative and natural conversations with the HealsMindAI to get personalized healthcare insights.
- Open-Source Dataset: The project is built on a dataset sourced from Project Gutenberg, offering a foundation for comprehensive health-related information.
- User Customization: Integrate your own health-related documents into the system, allowing HealsMindAI to provide insights based on your unique information.
- Transparent AI: HealsMindAI focuses on transparency and explainability, ensuring that the AI-generated responses are understandable and informative.
Getting Started for deploying locally:
- git lfs install
- git clone 'https://huggingface.co/spaces/lavanjv/HealsmindAI`
- Download 'llama-2-7b-chat.ggmlv3.q8_0.bin' from https://huggingface.co/TheBloke/Llama-2-7B-Chat-GGML/tree/main
- Install the required dependencies:
pip install -r requirements.txt
- Run the Streamlit app:
streamlit run app.py -w
- Interact with HealsMindAI through the provided UI and explore its capabilities.
For fine tunning using custom pdf:
- Place the pdf files in data folder
- Run 'python ingest.py'
- Then run
streamlit run app.py -w
Contributing
We welcome contributions to enhance and expand HealsMindAI' knowledge base. If you have health-related documents or insights, feel free to contribute by submitting a pull request.
License
This project is licensed under the MIT License.
Contact
For questions or inquiries, please contact [email protected]. Created with Love by Lavan