datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
nbm-conus-analysis
NOAA NBM CONUS Daily Analysis (Zarr)
Daily best-estimate analysis derived from NOAA NBM (National Blend of Models)
CONUS forecasts, on the native ~2.5 km Lambert conformal grid (2345 x 1597).
Built by nbm-to-zarr, dynamical.org-style.
Variables: tmean / tmax / tmin (degC), precip (mm), srad (MJ/m2/day)
Construction: best estimate for day D = lead-day 1 of that day's 00z NBM init
Coverage: rolling backfill from 2020-10-01 (AWS NBM archive floor) to present
Layout: one standalone… See the full description on the dataset page: https://huggingface.co/datasets/nakas/nbm-conus-analysis.conus
Decompression
If you need to decompress the files, please see the main README at the github repo.
If you want to use them directly from the parquet files, the original .tif/.nc files were read into the rows as binary file data sources
Contiguous United States AOI
We provide '.csv' files with predefined train (blue), test (orange) and validation (green) splits that can be used for repeatability and comparability of experiments.
60% of tiles are allocated for training… See the full description on the dataset page: https://huggingface.co/datasets/M3LEO/conus.nbm-conus-forecastconus
Contiguous UNited States AOI
Randomly sampled 5000 data tiles (3000 for training, and 1000 for validation and testing).
Occasionally, not all tiles were available for all datasets. We include a detailed breakdown of the number of chips per dataset below.
Datakind
Chips
s1grd-2020
5000
gssic
5000
gunw_2020-04-01_2020-06-30
4138
gunw_2020-07-01_2020-09-30
4171
gunw_2020-08-01_2020-10-31
4166
gunw_2020-01-01_2020-03-31
4097
s2rgbm-2020
5000
biomass-2020
5000… See the full description on the dataset page: https://huggingface.co/datasets/M3LEO-miniset/conus.conus-2020-blocks-aefbsrn-nsrdb-conus
BSRN NSRDB PSM v4 conus
Point-extracted NSRDB (National Solar Radiation Database) PSM v4 (conus variant) data for Baseline Surface Radiation Network (BSRN) stations.
Dataset Description
Each file contains 5-minute NSRDB variables at a single BSRN station location. Extraction is performed via the NREL NSRDB API (NLR). The conus variant (v4.0.0) provides 2 km spatial resolution and covers the Contiguous United States using GOES-East and GOES-West satellite data from 2018 to… See the full description on the dataset page: https://huggingface.co/datasets/dazhiyang/bsrn-nsrdb-conus.conus-h3-land-cover
CONUS H3 Land Cover Density (NLCD 2024)
A continental United States land cover dataset aggregated to Uber's H3 hexagonal grid at resolution 10 (~120m cell edge length). Each cell contains continuous per-channel coverage fractions derived from the USGS National Land Cover Database (NLCD) 2024. Unlike NLCD's native discrete classification, this dataset expresses land cover as continuous multi-channel density values — a cell at a forest-wetland boundary is represented as 0.6 forest and… See the full description on the dataset page: https://huggingface.co/datasets/Timmahw/conus-h3-land-cover.NOAA-HRRR-CONUS-ImageCaptionconusacademic_conuseling
Dataset Card for "academic_conuseling"
More Information needed
conus-crest-hydroacademic_conuseling
Dataset Card for "academic_conuseling"
More Information needed