/NowCasting-nets

NowCasting-Nets: Representation Learning to Mitigate Latency Gap of Satellite Precipitation Products Using Convolutional and Recurrent Neural Networks

Primary LanguagePythonMIT LicenseMIT

NowCasting-nets

This repository contains codes to our proposed Nowcasting-Nets models in the NowCasting-nets: Deep Neural Network Structures for Precipitation Nowcasting Using IMERG paper. The Models.py file contains the details of our proposed models and the rest of the files are used to properly read the data, generate hyperparameter tuning experiments, and create, train and save the models.

Dataset

The clean and preprocessed dataset is available here in a h5 file.