Skip to main content

Table 2 Summary statistics of final tuned models in the test set (20% of the data). R2 (R squared), and root mean square error (RMSE)

From: Proposal of a new equation for estimating resting energy expenditure of acute kidney injury patients on dialysis: a machine learning approach

Models

R2

RMSE

Linear model with stepwise

0.28

569.24

Regularized linear model

0.26

578.44

SVM

0.38

530.05

Random forest

0.48

490.05

R part

0.36

566.61

GBM

0.36

535.88

XGBoost

0.45

499.75