Website Update Commands

This guide contains all the commands you need to update your clean, optimized academic website.

Note: The codebase has been cleaned up - removed 20+ unnecessary template files, unused layouts, demo content, and duplicate files. See STRUCTURE.md for full details.

Quick Update (Recommended)

For a complete website refresh with latest publications and clean organization:

cd /Users/nesar/Projects/Misc/nesar.github.io
python3 scripts/cleanup_and_organize.py

This single command will:

✅ Refresh research images by extracting figures from papers
✅ Update publications page with single Paper link logic
✅ Analyze all 60+ publications and categorize them correctly
✅ Ensure exhaustive publication lists for each research area
✅ Select diverse figures from different papers (max 2 per category)
✅ Create clean portfolio pages without duplicates
✅ Generate a clean research overview page
✅ Fix any content issues automatically

Individual Update Commands

1. Refresh Research Images Only

If you just want to update the research figures:

# Auto-extract figures from all publications
python3 scripts/auto_extract_from_publications.py

# Extract figures from a specific PDF
python3 scripts/extract_figures.py /path/to/new_paper.pdf

2. Update Publications Data

To fetch latest publications from Google Scholar/arXiv:

python3 scripts/update_scholar_publications.py

3. Validate Website Structure

To check for issues without making changes:

# Check for duplicate content
grep -r "Machine Learning & AI" _portfolio/ _pages/

# Count publications per category
python3 -c "
import os, re
from collections import defaultdict

categories = defaultdict(int)
for f in os.listdir('_publications'):
    if f.endswith('.md'):
        with open(f'_publications/{f}', 'r') as file:
            content = file.read()
            if any(k in content.lower() for k in ['machine learning', 'deep learning', 'ai ']):
                categories['ML'] += 1
            elif any(k in content.lower() for k in ['dark matter', 'cosmic web', 'cosmology']):
                categories['Dark Matter'] += 1
            elif any(k in content.lower() for k in ['uncertainty', 'probabilistic']):
                categories['UQ'] += 1
            elif any(k in content.lower() for k in ['emulator', 'surrogate']):
                categories['Emulation'] += 1

for cat, count in categories.items():
    print(f'{cat}: {count} papers')
"

Testing Locally

Before deploying changes:

# Install Jekyll dependencies (one-time setup)
bundle install

# Serve locally
bundle exec jekyll serve

# View at: http://localhost:4000

Deployment

Your website auto-deploys via GitHub Pages when you push to master:

git add .
git commit -m "Update research content and figures"
git push origin master

File Structure Reference

_portfolio/
├── portfolio-1-machine-learning.md       # 24 ML papers
├── portfolio-2-dark-matter.md           # 10 Dark Matter papers  
├── portfolio-3-uncertainty-quantification.md  # 4 UQ papers
└── portfolio-4-statistical-emulation.md # 10 Statistical Emulation papers

_pages/
└── research.html                         # Clean overview page

images/research/figures/                  # Extracted figures
scripts/
├── cleanup_and_organize.py              # Main update script
├── extract_figures.py                   # Figure extraction
└── update_scholar_publications.py       # Publication updates

Troubleshooting

If figures aren’t showing:

# Check if figure files exist
ls -la images/research/figures/

# Regenerate figures
python3 scripts/extract_figures.py

If duplicates appear:

# Run cleanup (fixes all duplicates)
python3 scripts/cleanup_and_organize.py

If publication counts are wrong:

# Force refresh all publications
python3 scripts/cleanup_and_organize.py

Automation Schedule

For regular updates, you can set up a cron job:

# Edit crontab
crontab -e

# Add line for monthly updates (runs 1st of each month at 9 AM)
0 9 1 * * cd /Users/nesar/Projects/Misc/nesar.github.io && python3 scripts/cleanup_and_organize.py

Current Status

After running cleanup_and_organize.py, your website has:

✅ Foundation Models: LLM and foundation model research, 2 figures
✅ Machine Learning for Science: Non-LLM ML applications, 2 figures
✅ Dark Matter & Cosmology: Cosmological simulations, 2 figures
✅ Emulation & Inference: Combined UQ and emulation methods, 2 figures
✅ Clean research overview with no duplicates
✅ Single “Paper” links (published version or arXiv fallback)
✅ Publications properly categorized by research area

Support

If you encounter issues:

Check this file for relevant commands
Run python3 scripts/cleanup_and_organize.py to fix most problems
Check the console output for specific error messages

Nesar Ramachandra