Approximate AUC
AUC can be computed exactly by sorting the fitted values, which is often computationally slow. Instead, we can approximate the AUC numerically using monte carlo.
approxAUC(y, yhat, n = 1000)
y |
the actual class labels [0-1] |
yhat |
the predicted probabilities |
n |
number of samples to draw |
erik, Neal Fultz
g <- glm(y~x,data=data.frame(x=1:10,y=1:10)) classMethods(g)
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.