{"repo":"joanna-lgmt/Swasthyabot","free":true,"listed":false,"github":"https://github.com/joanna-lgmt/Swasthyabot","clone":"git clone https://github.com/joanna-lgmt/Swasthyabot.git","description":"A simple and responsive login page built with HTML, CSS, and JavaScript. Includes username and password fields with basic client-side validation. Clean and user-friendly design suitable as a starter template for web applications or learning front-end development.","language":"HTML","stars":19,"topics":[],"license":"EPL-2.0","category":"databases-storage","readme_excerpt":"Swasthyabot: AI Medical Help Agent 📄 Overview Swasthyabot is a Retrieval-Augmented Generation (RAG) pipeline designed as a medical help AI agent. It allows users to upload medical documents (PDF/DOCX), retrieve relevant information, and get AI-generated answers grounded in the uploaded content. 🚀 Features - Upload PDF or DOCX medical documents - Extract and chunk text into manageable pieces - Generate embeddings using nomic-embed-text via Ollama - Store and search chunks with FAISS vector database - Query documents in natural language and get context-aware answers - Interactive Streamlit web UI for easy usage 🛠️ Installation & Setup 1. Install dependencies: bash pip install streamlit pyngrok python-docx PyPDF2 faiss-cpu requests tqdm Install Ollama: bash Copy code curl -fsSL https://ollama.com/install.sh sh Start Ollama: bash Copy code ollama serve Pull required models: bash Copy code ollama pull nomic-embed-text ollama pull gemma3:1b Run Streamlit app: bash Copy code streamlit run app.py --server.port 8501 --server.headless true (Optional: use cloudflared to expose the app publicly.) 📂 Project Structure RAG.ipynb – Setup notebook app.py – Streamlit web application faiss index.index – Saved FAISS index metadatas.pkl – Metadata for text chunks 🖥️ Usage Upload medical PDF or DOCX documents in the Streamlit UI. Click \"Process & Build DB\" to create the FAISS index. Enter a medical question in natural language. The app will: Retrieve relevant chunks Generate context-aware ans","default_branch":null,"files":null,"tree":[],"storefront":"/r/joanna-lgmt","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/joanna-lgmt/Swasthyabot/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}