This is a plain HTML, CSS, and JavaScript starter for a dataset website.
- No build step
- Works well on GitHub Pages
- Content lives in one JSON file
- Easy to customise without learning a framework
index.htmlcontains the page structureassets/site.csscontains the stylingassets/site.jsloads and renders the contentdata/site.jsonstores the dataset contentassets/examples/contains example images.nojekyllprevents GitHub Pages from running Jekyll processing
- Copy these files into the root of
crowd-dataset.github.io - Replace the text in
data/site.json - Replace the sample images in
assets/examples/ - Commit and push to your default branch
- In GitHub, open Settings, then Pages, and publish from the branch root
- Add a teaser figure near the top
- Replace the benchmark table with your actual results
- Add download mirrors, licence, and terms of use
- Add a paper PDF thumbnail or poster section
- Add a challenge or leaderboard page if needed
If you later want markdown driven content, search, or blog style pages, move to a static site generator. For a research project page, this starter keeps things simpler.