get_predict_table() - messy output table

[mueblacker]

I first tested the get_predict_table() on a very small basin with a handful of variables and it worked perfectly fine. But when I run it with many variables for the subcIDs of the Danube basin the output table I get is a mess. It has almost double of the rows than it should have. There are multiple rows with NAs and multiple duplicated subcIDs. For the duplicated subcIDs, I even have one variable where the values are different.

[JaimeRi]

We have experienced this issue when the tables of different variables (for the same tile) do not have the same number of rows (although they should have). This means that the some variable tables have not been calculated right (i.e., some sub-catchments are missing) and we need to re-run the calculations. We want to add to the function this internal checking at the beginning and provide a warning message instead of letting the calculations run and get a messy-non-sense table. Could you check the number of rows of the different tables for the same tile and let us know which ones are different?