Data for the manuscript "DEUCE v1.0: A neural network for probabilistic precipitation nowcasting with aleatoric and epistemic uncertainties" by Harnist et al. (2023)

Beskrivning

# File descriptions: ## `verif_inputs.zip` and `verif_case_inputs.zip` Contain folders of raw PGM composites, containing needed input data for verification experiments, that is for the creation of baseline nowcast as well as observation HDF5 files. ## `deuce_inputs.zip` Contains HDF5 archives of the composite crops used for the training of DEUCE, derived from raw PGM composites using the `create_hdf5_dataset.py` script given. ## `deuce_model_checkpoints.py` Contains the two intermediate pytorch lightning checkpoints of DEUCE, as the final model checkpoint used for making predictions. ## `metrics.zip` Contains the netcdf files of verification metric values computed from DEUCE and baseline model predictions compared to observations.
Visa mer

Publiceringsår

2023

Typ av data

Upphovspersoner

Harnist, Bent - Upphovsperson

Mäkinen, Terhi - Upphovsperson

Pulkkinen, Seppo - Upphovsperson

Projekt

Övriga uppgifter

Vetenskapsområden

Geovetenskaper

Språk

engelska

Öppen tillgång

Öppet

Licens

Creative Commons Attribution 4.0 International (CC BY 4.0)

Nyckelord

precipitation, INSPIRE theme: climatologyMeteorologyAtmosphere, weather radar, Bayesian, convolutional, neural network, nowcasting, probabilistic, uncertainty quantification

Ämnesord

Temporal täckning

undefined

Relaterade till denna forskningsdata