Dimensions of the training dataset.
Hi all, when I load the ecmwf hindcast data as part of the training dataset, the dimensions of the xarray dataset looks good but I am confused about the dimensions of the two variables loaded. For t2m and tp, the third dimension representing "forecast_time" is 12. Does anyone have any explanation for why this dimension is 12 instead of 1060 for the hindcast? Why the third dimension for these two parameters is 12 instead of 53, for the forecast data of 2020?. Thanks, Muhammad.