Interactive PyTorch Guide

A responsive, single-page learning hub covering PyTorch foundations, architectures, transfer learning strategy, engineering practices, and mastery. Built with semantic HTML, Tailwind via CDN, and Chart.js. Ready for GitHub Pages.

• Live URL (after publish): https://ashwani65.github.io/interactive-pytorch-guide/
• Entry point: index.html

Highlights

• Responsive layout with mobile drawer navigation
• Sectioned content: Foundations, Vision, Architectures, Strategy, Engineering, Mastery, Advanced, Pro Toolkit
• Chart.js comparison chart (Experiment Tracking tools)
• SEO meta tags and favicon
• GitHub Pages friendly: .nojekyll and 404.html included

Local Development

You can open index.html directly, but a tiny static server is recommended for consistent behavior:

# Python 3
python3 -m http.server 5500
# then open http://127.0.0.1:5500/

Or use VS Code “Live Server”.

Deploy to GitHub Pages

Choose one approach.

Option A — Personal site repo

If you use ashwani65.github.io:

# from personal site repo
rsync -a --delete /Users/ashwanisingh/Documents/github/interactive-pytorch-guide/ ./interactive-pytorch-guide/
git add .
git commit -m "Add interactive-pytorch-guide"
git push

Then in GitHub → Settings → Pages: Source = Deploy from branch (main), root.

URL: https://ashwani65.github.io/interactive-pytorch-guide/

Option B — Dedicated project repo

cd /Users/ashwanisingh/Documents/github/interactive-pytorch-guide
git init -b main
git add .
git commit -m "Initial commit: interactive PyTorch guide"
git remote add origin git@github.com:ashwani65/interactive-pytorch-guide.git
git push -u origin main

Then in GitHub → Settings → Pages: Source = Deploy from branch (main), root.

Optional: “Ask AI” (currently disabled)

The page includes an on-device AI chat integration (WebLLM) that is commented out to avoid blocking load:

• Requires a WebGPU-enabled browser (Chrome/Edge latest)
• First run downloads model weights (one-time)

To re-enable:

1. In index.html, uncomment the blocks labeled “Ask AI UI (temporarily disabled)”, the WebLLM script tag, and the Ask AI logic <script>.
2. Hard refresh the page. Use “Diagnose” in the Ask AI header if initialization stalls.

Structure

• index.html — app content and behavior
• favicon.svg — icon
• .nojekyll — bypass Jekyll on GitHub Pages
• 404.html — SPA-friendly redirect

License

Specify a license here if you intend to open source (e.g., MIT).

interactive-pytorch-guide