The dataset viewer is not available for this split.
Error code: TooManyColumnsError
Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
Archived analysis inputs for zhoism/protostar-climate-2026
The public files that the analysis in zhoism/protostar-climate-2026
reads (data joins, cooling gap, reuse example), archived so analysis/run_all.py --full still rebuilds if a source
moves or changes a file. None of this is a forecast input; the forecasts read only the challenge data.
analysis/fetch_external.py downloads from the original sources and falls back to this archive, at a pinned commit,
for any file that fails or no longer matches its recorded SHA-256; --mirror takes everything from here. Every file's
SHA-256 is in MANIFEST.json (also committed in the submission as analysis/integration/archive_manifest.json).
Files
- Files from
provenance.jsonare byte-identical copies, with the SHA-256 recorded there. - GHCN-Daily station files are cut at 2018-12-31: the header and every row dated on or before that day, byte for byte as NOAA served them on 2026-11-07; later rows removed. Every script reads them only up to 2018-12-31. NOAA rewrites these files in place, so the original hashes are not stable.
kilthub/USW00003856.csvis CMU's KiltHub v6 file as published (it runs to 2023); it is read only to locate a gap in the Huntsville record before 2018.
Licenses
U.S. Government works (Census Bureau, HUD, EIA, NOAA): public domain in the United States. CMU KiltHub file: CC0. Cite the original publishers: U.S. Census Bureau and HUD (American Housing Survey; SAIPE; Population Estimates; cartographic boundaries; CBSA delineation), U.S. EIA (RECS 2020), NOAA NCEI (GHCN-Daily; nClimDiv), and Lai & Dzombak (2019), doi:10.1184/R1/7890488.v6.
- Downloads last month
- 75