Two-stage-CNN-LSTM: Learning the spatio-temporal relationship between wind and significant wave height

This is the code for the predition of significant wave height (Hs) using wind conditions. The method consists of two stages: a CNN and an LSTM stage (for more details on the method, see the paper: https://doi.org/10.1017/eds.2022.35 ). Data is available at: https://drive.google.com/drive/folders/1SIXYRXIpoegZ_bTLsvmr1g77GLMnMWy_?usp=sharing

Significant wave height data comes frome Homere hindcast database: https://marc.ifremer.fr/produits/rejeu_d_etats_de_mer_homere And wind data comes from the Climate Forecast System Reanalysis (CFSR) database: https://climatedataguide.ucar.edu/climate-data/climate-forecast-system-reanalysis-cfsr