This repository publishes the NILMTK start page at https://nilmtk.github.io/. It answers four questions: which repository to use, how to install it, which paper to cite, and how a model reaches the benchmark leaderboard.
| Research task | Repository |
|---|---|
| Dataset conversion, meter access, preprocessing, and metrics | NILMTK core |
| Appliance taxonomy, synonyms, meter relationships, and dataset schema | NILM Metadata |
| Disaggregation model implementation and testing | nilmtk-contrib |
| Fixed T1/T2/T3 evaluation and published result bundles | NILMbench |
The site is deliberately dependency-free: semantic HTML, one stylesheet, and a
small progressive-enhancement script. Existing generated core API documentation
under nilmtk/ remains separate from the landing page.
python3 -m http.server 8000Then open http://localhost:8000/.
python3 scripts/check_site.pyThe check rejects divergent repository responsibilities, duplicate IDs, broken local assets and fragment links, missing tab panels, insecure public URLs, missing canonical repository or citation links, promotional copy that was intentionally removed, and accidental reintroduction of the retired Bootstrap/jQuery stack.
- Keep this page a router; detailed package APIs belong in their repositories.
- Keep one recommended Python path and one general onboarding image.
- Treat NILMbench's CPU and CUDA images as scientific runtimes, not per-model development images.
- Update citation guidance only when the canonical repository papers change.
- Verify commands against the target repository before publishing them here.