Sign inGet started
Dataset · Hugging Face

OliverLee520/WeatherForecast

WeatherForecast English · 简体中文 Hourly ERA5-derived weather fields for lower-cost forecasting experiments. Both subsets store unnormalized float32 values and continuous sequences, not pre-extracted windows. Datasets TyphoonTracking GlobalForecast Related experiment Met2Net Table 5 Met2Net Table 4 Time between consecutive frames 1 hour (source setting) 1 hour (UTC timestamps included) Grid 32 × 32, approximately 1° 32 × 64, 5.625° spacing Coverage… See the full description on the dataset page: https://huggingface.co/datasets/OliverLee520/WeatherForecast.

Author / organizationOliverLee520
License (as declared)other
Downloads0 (as reported by Hugging Face)
Likes3
Created23 October 2026
Last updated at source23 October 2026
weather-forecastingglobal-weather-forecastingweatherbenchera5meteorologytropical-cyclonespatiotemporal-predictionnumpymet2netgeospatialtimeseries
Embed this entry
<iframe src="https://silytics.si/embed/item/6344dba9-6f67-4748-837c-7122c92aecd4" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Recent changes

  • 23 Oct 2026 Description: WeatherForecast English · 简体中文 Hourly ERA5-derived weather fields for lower-cost forecasting experiments. Both subsets store unnormalized float32 values and continuous sequences, not pre-extracted windows. Datasets TyphoonTracking GlobalFor → WeatherForecast English · 简体中文 Hourly ERA5-derived weather fields for lower-cost forecasting experiments. Both subsets store unnormalized float32 values and continuous sequences, not pre-extracted windows. Datasets TyphoonTracking GlobalFor
  • 23 Oct 2026 Description: WeatherForecast English · 简体中文 Two low-resolution ERA5-derived datasets for weather forecasting research. Both contain hourly, multivariable weather fields in NumPy format, with fixed train/validation/test splits and training-set normalizat → WeatherForecast English · 简体中文 Hourly ERA5-derived weather fields for lower-cost forecasting experiments. Both subsets store unnormalized float32 values and continuous sequences, not pre-extracted windows. Datasets TyphoonTracking GlobalFor
  • 23 Oct 2026 Description: WeatherForecast 用于气象时空预测研究的ERA5派生降采样数据,现包含两个子集:TyphoonTracking为Met2Net表5相关的32×32区域气象场;GlobalForecast为Table4相关的32×64全球气象场。两者用于低计算成本实验,文件格式、轴顺序和归一化方法不同。 这是一份重新处理的低分辨率数据,不是Met2Net论文Table4/5的原始分辨率版本。文件包含连续气象场,不含台风中心经纬度、最佳路径标签或模型预测结果;TyphoonTrac → WeatherForecast English · 简体中文 Two low-resolution ERA5-derived datasets for weather forecasting research. Both contain hourly, multivariable weather fields in NumPy format, with fixed train/validation/test splits and training-set normalizat
  • 23 Oct 2026 Topics added geospatial, timeseries
  • 23 Oct 2026 Description: WeatherForecast 用于气象时空预测研究的降采样数据。当前发布TyphoonTracking子集:基于ERA5区域气象场,将Met2Net表5对应的128×128数据降采样到32×32,便于开展低计算成本实验。 这是一份重新处理的低分辨率数据,不是Met2Net论文表5的原始分辨率版本。文件包含连续气象场,不含台风中心经纬度、最佳路径标签或模型预测结果;TyphoonTracking是任务分类目录名,并不表示数据只包含台风发生时段。 目录 WeatherForec → WeatherForecast 用于气象时空预测研究的ERA5派生降采样数据,现包含两个子集:TyphoonTracking为Met2Net表5相关的32×32区域气象场;GlobalForecast为Table4相关的32×64全球气象场。两者用于低计算成本实验,文件格式、轴顺序和归一化方法不同。 这是一份重新处理的低分辨率数据,不是Met2Net论文Table4/5的原始分辨率版本。文件包含连续气象场,不含台风中心经纬度、最佳路径标签或模型预测结果;TyphoonTrac
  • 23 Oct 2026 Topics added global-weather-forecasting, weatherbench
  • 23 Oct 2026 License: mit → other
  • 23 Oct 2026 Description: none → WeatherForecast 用于气象时空预测研究的降采样数据。当前发布TyphoonTracking子集:基于ERA5区域气象场,将Met2Net表5对应的128×128数据降采样到32×32,便于开展低计算成本实验。 这是一份重新处理的低分辨率数据,不是Met2Net论文表5的原始分辨率版本。文件包含连续气象场,不含台风中心经纬度、最佳路径标签或模型预测结果;TyphoonTracking是任务分类目录名,并不表示数据只包含台风发生时段。 目录 WeatherForec
  • 23 Oct 2026 Topics added weather-forecasting, era5, meteorology, tropical-cyclone, spatiotemporal-prediction, numpy, met2net

Changes detected by Silytics between refreshes of the source.

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 23 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.