The Satellite Precipitation Retrieval (SatRain) benchmark dataset develped by the International Precipitation Working Group (IPWG) is a benchmark dataset for machine-learning-based satellite precipitation retrievals, i.e., algorithms for detecting and quantifying precipitation from satellite imagery.
- Training, validation, and test splits derived from four years of overpasses of passive-microwave sensors over the conterminous united states (CONUS)
- Multi-spectral input observations (microwave, infrared, and visible) from low-earth and geostationary platforms
- Comprehensive ancillary data covering atmospheric parameters and surface properties
- Independent evaluation datasets dervied from independent time periods, regions, and measurement techniques
- Flexible evaluation framework
For instructions on how to get starting using the dataset refer to the documentation available here.