Add deployment guide for Hugging Face Spaces and local setup

DEPLOYMENT.md

@@ -0,0 +1,190 @@
+# Deployment Guide
+
+## 🚀 Deploying to Hugging Face Spaces
+
+### Step 1: Create a Space
+
+1. Go to [Hugging Face Spaces](https://huggingface.co/spaces)
+2. Click "Create new Space"
+3. Fill in the details:
+   - **Owner**: Your username
+   - **Space name**: `PaperReview` (or your preferred name)
+   - **License**: MIT
+   - **SDK**: Gradio
+   - **Space hardware**: CPU basic (free tier works fine)
+
+### Step 2: Configure Secrets
+
+In your Space settings, add these secrets:
+
+#### Required:
+- `OPENAI_API_KEY`: Your OpenAI API key or compatible API key
+  - Get from: https://platform.openai.com/api-keys
+
+#### Optional but Recommended:
+- `SEMANTIC_SCHOLAR_API_KEY`: `0NK3h4KTzo1O1hXHsueIvwARh3L8JgE5IU0tLkGi`
+  - This provides higher rate limits for Semantic Scholar API
+  - Without it, you'll be limited to basic rate limits
+
+#### Optional:
+- `OPENAI_BASE_URL`: Custom API endpoint (if not using OpenAI)
+  - Default: `https://api.openai.com/v1`
+  - For Azure: `https://your-resource.openai.azure.com/`
+  - For local: `http://localhost:8000/v1`
+
+- `MODEL_NAME`: Model identifier
+  - Default: `gpt-3.5-turbo`
+  - Options: `gpt-4`, `gpt-4-turbo`, etc.
+
+### Step 3: Upload Files
+
+Upload these files to your Space:
+```
+├── app.py
+├── agents.py
+├── requirements.txt
+├── README.md
+├── .gitignore
+└── .env.example (optional)
+```
+
+**Important**: Do NOT upload `.env` file with actual secrets!
+
+### Step 4: Wait for Build
+
+Hugging Face will automatically:
+1. Install dependencies from `requirements.txt`
+2. Load secrets from Space settings
+3. Start the Gradio app
+
+Build time: ~3-5 minutes
+
+### Step 5: Test Your Space
+
+1. Open your Space URL
+2. Upload a test PDF
+3. Click "Review Paper"
+4. Wait 3-6 minutes for results
+
+## 🔧 Local Development
+
+### Setup
+
+```bash
+# Clone or navigate to project
+cd aireviewer
+
+# Create virtual environment
+python -m venv venv
+source venv/bin/activate  # On Windows: venv\Scripts\activate
+
+# Install dependencies
+pip install -r requirements.txt
+
+# Copy environment template
+cp .env.example .env
+
+# Edit .env with your API keys
+nano .env  # or use your preferred editor
+```
+
+### Required in .env:
+
+```bash
+OPENAI_API_KEY=sk-...your-key-here
+SEMANTIC_SCHOLAR_API_KEY=0NK3h4KTzo1O1hXHsueIvwARh3L8JgE5IU0tLkGi
+```
+
+### Run
+
+```bash
+python app.py
+```
+
+Open browser to: http://localhost:7860
+
+## 🐳 Docker Deployment
+
+### Build Image
+
+```bash
+docker build -t aireviewer .
+```
+
+### Run Container
+
+```bash
+docker run -p 7860:7860 \
+  -e OPENAI_API_KEY=your-key \
+  -e SEMANTIC_SCHOLAR_API_KEY=0NK3h4KTzo1O1hXHsueIvwARh3L8JgE5IU0tLkGi \
+  -e MODEL_NAME=gpt-4 \
+  aireviewer
+```
+
+Or use docker-compose:
+
+```yaml
+version: '3.8'
+services:
+  aireviewer:
+    build: .
+    ports:
+      - "7860:7860"
+    environment:
+      - OPENAI_API_KEY=${OPENAI_API_KEY}
+      - SEMANTIC_SCHOLAR_API_KEY=0NK3h4KTzo1O1hXHsueIvwARh3L8JgE5IU0tLkGi
+      - MODEL_NAME=gpt-4
+```
+
+## ⚠️ Important Notes
+
+### Rate Limiting
+- **LLM API**: 1 request per second (sequential processing)
+- **Semantic Scholar**: 1 request per second
+- Total processing time: 3-6 minutes per paper
+
+### API Costs
+- Using GPT-4: ~$0.10-0.50 per paper review (depending on paper length)
+- Using GPT-3.5-turbo: ~$0.01-0.05 per paper review
+
+### Concurrency
+- The app does NOT support concurrent reviews
+- Reviews are processed sequentially to avoid API rate limit issues
+- One paper at a time recommended
+
+### Security
+- Never commit `.env` file to git
+- Use Hugging Face Spaces secrets for production
+- API keys are loaded from environment variables only
+
+## 🔍 Troubleshooting
+
+### "Could not extract text from PDF"
+- Ensure PDF is not scanned/image-based
+- Try OCR preprocessing if needed
+
+### "API key error"
+- Check if secrets are properly set in Space settings
+- Verify API key is valid and has credits
+
+### "Rate limit exceeded"
+- Wait a few minutes and try again
+- The app has built-in rate limiting
+- Consider upgrading API tier
+
+### "Review taking too long"
+- Normal processing: 3-6 minutes
+- Check API status if longer
+- Review progress bar for current status
+
+## 📞 Support
+
+For issues:
+1. Check the logs in Hugging Face Space
+2. Verify all secrets are properly configured
+3. Test with a shorter paper first
+4. Open an issue on GitHub
+
+---
+
+**Live Demo**: https://huggingface.co/spaces/syaikhipin/PaperReview