added code for streamlit deployment

Author: RATED-R-SUNDRAM

FASTAPI-DEPLOYMENT/DEPLOYMENT_GUIDE.md

@@ -0,0 +1,267 @@
+# Complete Deployment Guide: FastAPI Backend + Streamlit Frontend
+
+## Architecture Overview
+- **FastAPI Backend** (`rhl_fastapi_deploy.py`): Provides `/chat` and `/chat-stream` API endpoints
+- **Streamlit Frontend** (`streamlit_stream_demo.py`): Makes HTTP requests to FastAPI backend
+
+Both components MUST be deployed separately and accessible via URLs.
+
+---
+
+## Option 1: Railway.app (Recommended - Easy)
+
+### Part A: Deploy FastAPI Backend
+
+1. **Install Railway CLI:**
+   ```bash
+   npm i -g @railway/cli
+   ```
+
+2. **Login to Railway:**
+   ```bash
+   railway login
+   ```
+
+3. **Initialize Railway project:**
+   ```bash
+   cd FASTAPI-DEPLOYMENT
+   railway init
+   ```
+
+4. **Set environment variables in Railway dashboard:**
+   - `OPENAI_API_KEY` = your OpenAI key
+   - `GOOGLE_API_KEY` = your Google API key
+
+5. **Deploy:**
+   ```bash
+   railway up
+   ```
+
+6. **Get your FastAPI URL:**
+   - Railway will provide a URL like: `https://your-app-name.railway.app`
+   - **Save this URL!** You'll need it for Streamlit
+
+---
+
+### Part B: Deploy Streamlit Frontend
+
+**Option B1: Streamlit Cloud (Free)**
+
+1. **Push your code to GitHub** (if not already)
+
+2. **Go to:** https://share.streamlit.io
+
+3. **Click "New app"**
+
+4. **Configure:**
+   - Repository: Your GitHub repo
+   - Branch: `main`
+   - Main file path: `FASTAPI-DEPLOYMENT/streamlit_stream_demo.py`
+
+5. **Advanced Settings → Secrets:**
+   ```
+   FASTAPI_URL=https://your-app-name.railway.app
+   ```
+   (Use the FastAPI URL from Part A)
+
+6. **Click "Deploy!"**
+
+7. **Share the Streamlit URL** with users!
+
+---
+
+**Option B2: Railway for Streamlit (Paid)**
+
+1. **Create new service in Railway**
+
+2. **Set environment variable:**
+   ```
+   FASTAPI_URL=https://your-fastapi-url.railway.app
+   ```
+
+3. **Start command:**
+   ```
+   streamlit run FASTAPI-DEPLOYMENT/streamlit_stream_demo.py --server.port $PORT --server.address 0.0.0.0
+   ```
+
+---
+
+## Option 2: Render.com (Free Tier)
+
+### Part A: Deploy FastAPI Backend
+
+1. **Create `render.yaml` in project root:**
+   ```yaml
+   services:
+     - type: web
+       name: medical-chatbot-api
+       env: python
+       buildCommand: pip install -r requirements.txt
+       startCommand: cd FASTAPI-DEPLOYMENT && uvicorn rhl_fastapi_deploy:app --host 0.0.0.0 --port $PORT
+       envVars:
+         - key: OPENAI_API_KEY
+           sync: false
+         - key: GOOGLE_API_KEY
+           sync: false
+   ```
+
+2. **Push to GitHub**
+
+3. **Go to render.com → New Web Service**
+
+4. **Connect repo → Render auto-detects `render.yaml`**
+
+5. **Get FastAPI URL:** `https://medical-chatbot-api.onrender.com`
+
+---
+
+### Part B: Deploy Streamlit on Streamlit Cloud
+
+1. **Deploy on Streamlit Cloud** (same as Option 1 Part B)
+
+2. **Set secret:**
+   ```
+   FASTAPI_URL=https://medical-chatbot-api.onrender.com
+   ```
+
+---
+
+## Option 3: Quick Testing with ngrok
+
+### Part A: Start FastAPI Locally
+
+1. **Terminal 1 - Start FastAPI:**
+   ```bash
+   cd FASTAPI-DEPLOYMENT
+   python rhl_fastapi_deploy.py
+   ```
+
+2. **Terminal 2 - Expose with ngrok:**
+   ```bash
+   ngrok http 8000
+   ```
+   - Copy the HTTPS URL: `https://abc123.ngrok.io`
+   - **This is your FastAPI URL!**
+
+---
+
+### Part B: Deploy Streamlit
+
+**Option 1: Streamlit Cloud**
+1. Deploy on Streamlit Cloud
+2. Set secret: `FASTAPI_URL=https://abc123.ngrok.io`
+
+**Option 2: Local + ngrok**
+1. **Terminal 3 - Start Streamlit:**
+   ```bash
+   cd FASTAPI-DEPLOYMENT
+   streamlit run streamlit_stream_demo.py
+   ```
+
+2. **Terminal 4 - Expose Streamlit:**
+   ```bash
+   ngrok http 8501
+   ```
+   - Copy URL: `https://xyz789.ngrok.io`
+   - **Share this with users!**
+   - Users set API URL in sidebar: `https://abc123.ngrok.io`
+
+---
+
+## Option 4: Local Network (Same WiFi)
+
+### Part A: Start FastAPI
+
+```bash
+cd FASTAPI-DEPLOYMENT
+python rhl_fastapi_deploy.py
+```
+
+- FastAPI runs on: `http://YOUR_IP:8000`
+
+---
+
+### Part B: Start Streamlit
+
+```bash
+cd FASTAPI-DEPLOYMENT
+streamlit run streamlit_stream_demo.py --server.address 0.0.0.0
+```
+
+- Streamlit runs on: `http://YOUR_IP:8501`
+- Share this URL with others on same WiFi
+- They set API URL in sidebar: `http://YOUR_IP:8000`
+
+---
+
+## Quick Reference
+
+### FastAPI Deployment Commands
+
+**Railway:**
+```bash
+railway login
+railway init
+railway up
+```
+
+**Render:**
+- Push to GitHub, connect to Render, auto-deploy
+
+**Local:**
+```bash
+python rhl_fastapi_deploy.py
+```
+
+---
+
+### Streamlit Deployment Commands
+
+**Streamlit Cloud:**
+- Push to GitHub → share.streamlit.io → Deploy
+
+**Local:**
+```bash
+streamlit run streamlit_stream_demo.py
+```
+
+**Local (accessible on network):**
+```bash
+streamlit run streamlit_stream_demo.py --server.address 0.0.0.0
+```
+
+---
+
+## Important Notes
+
+1. **FastAPI must be deployed FIRST** - Streamlit needs its URL
+2. **Set `FASTAPI_URL` environment variable** in Streamlit deployment
+3. **Or users can manually enter** FastAPI URL in Streamlit sidebar
+4. **Both services must be accessible** from the internet (or same network)
+
+---
+
+## Troubleshooting
+
+### "Connection refused" error in Streamlit
+- ✅ Check FastAPI is running
+- ✅ Verify FastAPI URL is correct
+- ✅ Test FastAPI endpoint directly: `https://your-api-url.com/test-debug`
+
+### Streamlit can't find FastAPI
+- ✅ Check `FASTAPI_URL` environment variable is set
+- ✅ Or enter FastAPI URL manually in Streamlit sidebar
+- ✅ Ensure FastAPI URL is accessible (no firewall blocking)
+
+---
+
+## Recommended Setup
+
+**For Quick Testing:**
+- FastAPI: Local + ngrok
+- Streamlit: Streamlit Cloud (set ngrok URL as secret)
+
+**For Production:**
+- FastAPI: Railway.app or Render.com
+- Streamlit: Streamlit Cloud (set production API URL as secret)
+