search

USGS-R / river-dl

Deep learning model for predicting environmental variables on river systems
Creative Commons Zero v1.0 Universal
21 stars 15 forks source link

use (normalized) nash-sutcliffe as loss function #82

Closed jsadler2 closed 3 years ago

jsadler2 commented 3 years ago

try using normalized nash-sutcliffe in loss function. I think this is a good idea to try b/c it is what I'm usually reporting for flow performance and it is the standard in the hydrology community (much more than rmse).

jsadler2 commented 3 years ago

I will use normalized Nash-sutcliffe since it bounded: image https://en.wikipedia.org/wiki/Nash–Sutcliffe_model_efficiency_coefficient