This repo is a plain HTML/CSS site (no build step) designed to be served directly by GitHub Pages.
.
├── index.html # Shared landing page (institution, course, group,
│ AI disclosure, assignment index with deliverable links)
├── assignment1.html # Assignment 1 page
├── assignment2.html # Assignment 2 page (+ Dataset Proposal section)
├── assignment3.html # Assignment 3 page (+ Dataset Proposal + modality-interaction section)
├── assets/
│ ├── css/
│ │ └── style.css # Shared stylesheet for all pages
│ └── images/ # Put EDA plots, diagrams, screenshots here
├── code-repo-template/ # Template for the *source-code* repository
│ ├── README.md # install / data prep / train / evaluate / traceability
│ ├── AI_USAGE.md # repo-level AI usage disclosure
│ ├── configs/
│ │ └── config.example.yaml
│ ├── environment/
│ │ ├── requirements.txt
│ │ └── environment.yml
│ ├── src/ # model / data / training code goes here
│ ├── scripts/
│ │ ├── prepare_data.sh
│ │ ├── train.py
│ │ └── evaluate.py
│ ├── checkpoints/
│ │ └── README.md # download links or reconstruction instructions
│ ├── results/
│ │ └── experiment_log.csv # traceability: result → config → split → checkpoint → log ID → commit
│ └── .gitignore
└── README.md
code-repo-template/ is not part of the GitHub Pages site — it's a starting
point to copy into each assignment's actual source-code repository (one repo per
assignment, or one folder per assignment in a monorepo) so it satisfies the
Section 4 reproducibility requirements out of the box.
Every placeholder is wrapped like <span class="ph">[...]</span> or inside a
<div class="ph-block">...</div> and rendered with a dashed amber box so it's
easy to spot while editing. Search each HTML file for [ to find them quickly,
or search for class="ph".
Checklist per file:
- index.html — group name/ID, all members' names/IDs/roles/GitHub links, repo link, assignment titles, course-wide AI usage disclosure.
- assignment1.html / assignment2.html / assignment3.html — title, problem statement, dataset & EDA, methodology, experimental setup, results, comparison & discussion, error analysis, limitations & conclusion, source code / checkpoint / report-slides / YouTube links, and an assignment-specific AI usage disclosure. Assignment 3 additionally has a "Modality interaction" section (support vs. conflict cases).
Remember:
- Do not invent GitHub profile links — write "N/A" if a member has none.
- YouTube videos must be Public or Unlisted, titled
CO3133-Semester-261 – Group [ID] – Assignment [1/2/3], and verified before submission.
- Assignment page on GitHub Pages (
assignmentN.html) - Source-code repository (use
code-repo-template/as a starting point) - Report
- Presentation slides
- YouTube presentation video (Public or Unlisted, correct title format)
- AI Usage Disclosure in all four places: landing page, assignment page,
report, and
AI_USAGE.mdin the code repo - Checkpoint(s) or reconstruction instructions
- LMS submission per course rules
- Assignments 2 & 3 only: Dataset Proposal record + approval status
- Push this repo to GitHub (e.g.
group-XX-co3133). - Go to Settings → Pages.
- Under "Build and deployment", set Source to
Deploy from a branch. - Choose the branch (e.g.
main) and folder/ (root), then Save. - Your site will be published at:
https://<your-username>.github.io/<repo-name>/ - Put that URL (and the repo link) wherever your course submission requires it.
Save image files into assets/images/ and reference them in a page like:
<img src="assets/images/your-file.png" alt="Describe the image" style="max-width:100%;border-radius:6px;">Copy assignment1.html, rename it, and update the <title>, kicker,
<h1>, section content, and the AI-disclosure heading — keep the shared
<div class="topbar"> navigation identical across all pages (just add a new
link if you add a page) so navigation stays consistent site-wide.