datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
gfs-kerchunksevere-weather-env-gfs-forecast-v1
Severe Weather Environment GFS Forecast v1
This dataset contains compact GFS point-feature rows from severe-weather forecast-map experiments. It is meant for lead-time testing, forecast calibration, and medium-range SPC-style probability research.
This is a research dataset, not operational guidance. It is not an official NOAA, NWS, SPC, or GFS product, and it should not be used as a stand-alone warning or forecast system.
Files
data/gfs_point_forecast.parquet: primary… See the full description on the dataset page: https://huggingface.co/datasets/deepguess/severe-weather-env-gfs-forecast-v1.gfs_s3_gik_refstags:
zarr
AWS S3 GFS made into refrecnes and stored as paraquet file
Based on the methods discussed in https://github.com/icpac-igad/grib-index-kerchunk
Each Paraquet file is 00z run with 15 variables as used in the file
https://github.com/icpac-igad/grib-index-kerchunk/blob/main/run_day_gfs_gik.py
S.No
Description
Variable
1
Convective available potential energy
CAPE:surface
2
Convective precipitation (water)
ACPCP:surface
3
Medium cloud cover
MCDC:middle cloud… See the full description on the dataset page: https://huggingface.co/datasets/Nishadhka/gfs_s3_gik_refs.GFS_PCS_Datasets
[CVPR 2025] GFS-VL: Generalized Few-shot 3D Point Cloud Segmentation with Vision-Language Model
Overview
GFS-VL is a novel framework proposed in our CVPR 2025 paper: Generalized Few-shot 3D Point Cloud Segmentation with Vision-Language Model.
Our approach leverages the synergy between:
Dense but noisy pseudo-labels from 3D Vision-Language Models
Precise yet sparse few-shot samples
by maximizing the strengths of both data sources for effective generalized few-shot 3D… See the full description on the dataset page: https://huggingface.co/datasets/ZhaochongAn/GFS_PCS_Datasets.gfs-merra-1800-2400-forecast
GFS - MERRA - 1800 - 2400 - Forecasts
This dataset contains GFS CAPE forecasts and MERRA-2 AOD assimilations for the contiguous U.S. from 2008 - 2020.
Temporal Domain
January 1, 2008 to December 31, 2020
Spatial Domain
The contiguous U.S. defined as the region spanning longitudes 145.25°W to 54.75°W and latitudes 23.75°N to 62.25°N.
Features
Date
The date of the forecasts.
Spatial Field
The spatial field is a raster with eight… See the full description on the dataset page: https://huggingface.co/datasets/zj37/gfs-merra-1800-2400-forecast.gf_scanobject_ss_0.01gfsdgfsdggf_scanobjectnn_ss_0.05gfs-merra-1800-2400-0-3-6-forecastGF-Screencrazy-asian-gfscrazy-ex-gfsdataset_gfsqibgzgfsggwgfsrxvgfs3434gfshbshggFSeLKKqgfsdgrwgGFSGRER32gfsgfdhg54y6uyrtgfsdgsdt456546546gfsayhapifagfsgdsfggfsagwr