Skip to content

DEPLOYMENT.md

Latest commit

 

History

History
251 lines (183 loc) · 5.81 KB

DEPLOYMENT.md

File metadata and controls

251 lines (183 loc) · 5.81 KB
Raw

Dirghayu Deployment Guide

📍 Current Status

🚀 Deployment Options

Option 1: HuggingFace Spaces (RECOMMENDED) 🤗

Best for: Gradio web app with file upload
Cost: FREE
URL: https://huggingface.co/spaces/YOUR_USERNAME/dirghayu

Steps:

  1. Create HuggingFace Account: https://huggingface.co/join

  2. Create New Space:

  3. Push Code:

cd c:\Projects\open-source\dirghayu

# Add HuggingFace remote
git remote add hf https://huggingface.co/spaces/YOUR_USERNAME/dirghayu

# Copy simplified app
cp app_hf.py app.py
cp requirements-hf.txt requirements.txt

# Commit and push
git add app.py requirements.txt
git commit -m "Add HuggingFace deployment"
git push hf main
  1. Done! Your app will be live at: https://huggingface.co/spaces/YOUR_USERNAME/dirghayu

Option 2: Streamlit Cloud ☁️

Best for: Simple data apps
Cost: FREE
URL: https://YOUR_USERNAME-dirghayu.streamlit.app

Create Streamlit App:

# streamlit_app.py
import streamlit as st
from data import parse_vcf_file

st.title("🧬 Dirghayu")
st.caption("India-First Longevity Genomics")

uploaded_file = st.file_uploader("Upload VCF", type=['vcf'])

if uploaded_file:
    variants_df = parse_vcf_file(uploaded_file)
    st.success(f"✅ {len(variants_df)} variants analyzed")
    st.dataframe(variants_df)

Deploy:

  1. Go to: https://share.streamlit.io/
  2. Connect GitHub repo: VedantMadane/dirghayu
  3. Main file: streamlit_app.py
  4. Deploy!

Option 3: Render (FastAPI) 🔧

Best for: Production API server
Cost: FREE tier available
URL: https://dirghayu.onrender.com

Steps:

  1. Create render.yaml:
services:
  - type: web
    name: dirghayu-api
    env: python
    buildCommand: "pip install -r requirements.txt"
    startCommand: "uvicorn src.api.server:app --host 0.0.0.0 --port $PORT"
    envVars:
      - key: PYTHON_VERSION
        value: 3.11
  1. Deploy:
    • Go to: https://render.com
    • New → Web Service
    • Connect repo: VedantMadane/dirghayu
    • Auto-deploy!

Option 4: GitHub Pages (Static Reports) 📄

Best for: Static HTML demo reports
Cost: FREE
URL: https://VedantMadane.github.io/dirghayu/

Steps:

# Generate sample report
python web_demo.py data/clinvar_sample.vcf

# Rename for GitHub Pages
mv dirghayu_report.html index.html

# Create docs folder
mkdir docs
mv index.html docs/

# Push to GitHub
git add docs/
git commit -m "Add demo report for GitHub Pages"
git push origin main

# Enable GitHub Pages
gh repo edit --enable-pages --pages-branch main --pages-path docs

Live Demo: https://VedantMadane.github.io/dirghayu/

Option 5: Railway 🚂

Best for: Full-stack deployment
Cost: $5/month credit FREE
URL: https://dirghayu.up.railway.app

Steps:

  1. Go to: https://railway.app
  2. New Project → Deploy from GitHub
  3. Select: VedantMadane/dirghayu
  4. Add start command: python app.py
  5. Deploy!

🎯 Recommended Setup

For Maximum Reach:

  1. HuggingFace Spaces - Interactive web app

    • URL: https://huggingface.co/spaces/YOU/dirghayu
    • Users can upload VCF files

  2. GitHub Pages - Static demo

    • URL: https://VedantMadane.github.io/dirghayu/
    • Show sample analysis report

  3. FastAPI on Render - Production API

    • URL: https://dirghayu.onrender.com/docs
    • OpenAPI documentation

📊 Cost Comparison

Platform Free Tier Best For URL Format
HuggingFace ✅ Unlimited Gradio apps hf.co/spaces/USER/APP
Streamlit ✅ Unlimited Data apps USER-APP.streamlit.app
GitHub Pages ✅ Unlimited Static sites USER.github.io/REPO
Render ✅ 750 hrs/mo APIs APP.onrender.com
Railway ✅ $5 credit Full-stack APP.railway.app
Vercel ✅ Unlimited Static/Next.js APP.vercel.app

🚀 Quick Deploy Commands

Deploy to HuggingFace:

cd c:\Projects\open-source\dirghayu
git remote add hf https://huggingface.co/spaces/YOUR_USERNAME/dirghayu
git push hf main

Deploy GitHub Pages:

python web_demo.py data/clinvar_sample.vcf
mkdir docs
mv dirghayu_report.html docs/index.html
git add docs/ && git commit -m "Add demo" && git push
gh repo edit --enable-pages --pages-branch main --pages-path docs

Deploy Streamlit:

# Just connect your GitHub repo at https://share.streamlit.io/
# No commands needed - click deploy!

🌐 Expected URLs After Deployment

Once deployed, your project will be available at:

  • 🤗 HuggingFace: https://huggingface.co/spaces/YOUR_USERNAME/dirghayu
  • 📄 GitHub Pages: https://VedantMadane.github.io/dirghayu/
  • ☁️ Streamlit: https://YOUR_USERNAME-dirghayu.streamlit.app
  • 🔧 Render API: https://dirghayu.onrender.com

🎯 Next Steps

1. Choose Your Platform

Pick from: HuggingFace, Streamlit, GitHub Pages, or Render

2. Follow the Steps Above

Each platform has simple deployment instructions

3. Share Your Live Demo

Once deployed, share the URL!

📝 Notes

  • HuggingFace Spaces - Best for interactive demos
  • GitHub Pages - Best for showcasing results
  • Streamlit Cloud - Easiest to deploy
  • Render/Railway - Best for production APIs

🆘 Need Help?

  1. HuggingFace Docs: https://huggingface.co/docs/hub/spaces
  2. Streamlit Docs: https://docs.streamlit.io/deploy
  3. Render Docs: https://render.com/docs
  4. GitHub Pages: https://pages.github.com/

Ready to deploy? Let's get your genomics platform live! 🚀