Table 8

Evaluation of ML methods

ModelMERMSEPBIAS %NSEcpR2VE
SVR with log of response and interactions* −15.5 90.2 −18.6 0.813 0.840 0.835 0.656 
Small DLNN −6.9 90.3 −8.3 0.813 0.840 0.814 0.552 
DLNN – bigger, using regularization 9.8 71.4 13.3 0.834 0.863 0.897 0.607 
ModelMERMSEPBIAS %NSEcpR2VE
SVR with log of response and interactions* −15.5 90.2 −18.6 0.813 0.840 0.835 0.656 
Small DLNN −6.9 90.3 −8.3 0.813 0.840 0.814 0.552 
DLNN – bigger, using regularization 9.8 71.4 13.3 0.834 0.863 0.897 0.607 

*C = 0.1, linear kernel.

Close Modal

or Create an Account

Close Modal
Close Modal