Matryoshka in-the-wild query tables (openly licensed subset: Page Views, Energy, Fire Incidents)

Openly licensed in-the-wild (ITW) query tables of the PVLDB 2026 paper Matryoshka, each with its splits.json (features, target, target type, query column): Page Views (pageviews.csv, GitTables lake): Pantheon 1.0 (Yu et al., Scientific Data 2016), Harvard Dataverse doi:10.7910/DVN/28201, CC0 1.0. Energy (energy.csv, NYC lake): NYC Municipal Building Energy Benchmarking Results, City of New York via Kaggle, CC0. Fire Incidents (fire.csv, NYC lake, from the lake): NYC Open Data tm6d-hbzd (Incidents Responded to by Fire Companies), NYC Administrative Code § 23-502(d). Street Trees (NYC Open Data kyad-zm4j) is also openly licensed and will be added once its raw query table is restored. Query tables with unclear or restrictive licences (Jobs, IMDB, VG Sales, Real Estate, Arrest Records, Hospital) are not redistributed here; see the companion code for download and preparation scripts. Food Facts is in a separate ODbL record.