Its truly helpful that EF51.2.3 have done better job on the prediction of flood!
However, there are some issues like my observed value of output(csv) all become nan, and the discharge of prediction are normal. How did this happen? Is this a new bug of it?
I'm worried about this issue will disturb the calibration...
Hope i can get your reply soon, thanks!