DEPLOY.md

🚀 Deployment Guide

Deploy to Hugging Face Spaces

Prerequisites

1. Install Hugging Face CLI:

pip install huggingface_hub

2. Login to Hugging Face:

huggingface-cli login

Create and Deploy Space

1. Create a new Space on Hugging Face Hub:

huggingface-cli repo create --type space --space_sdk gradio your-username/one-pager-generator

2. Clone and set up the repository:

git clone https://huggingface.co/spaces/your-username/one-pager-generator
cd one-pager-generator

3. Copy files to the Space repository:

cp ../one-pager/* .

4. Add, commit and push:

git add .
git commit -m "Initial commit: AI One-Pager Generator"
git push

Alternative: Direct CLI Upload

You can also use the HF CLI to upload files directly:

huggingface-cli upload your-username/one-pager-generator . --repo-type=space

Files Required for Deployment

• app.py - Main application file
• requirements.txt - Python dependencies
• config.yaml - Space configuration
• README.md - Documentation
• .gitignore - Git ignore patterns

Configuration Notes

• The app uses distilgpt2 model for better compatibility
• CPU-only inference for free tier compatibility
• Fallback template system ensures reliable output
• Gradio interface optimized for Spaces

Post-Deployment

After deployment, your Space will be available at: https://huggingface.co/spaces/your-username/one-pager-generator

The app will automatically:

1. Install dependencies from requirements.txt
2. Load the AI model
3. Launch the Gradio interface
4. Be accessible via the web

Troubleshooting

• Model loading issues: The app falls back to structured templates
• Memory issues: Using smaller DistilGPT2 model for efficiency
• Timeout issues: CPU inference may be slower but more reliable