Function to calculate e- or softmax scaled membership values
Calculates the e- or softmax scaled membership values of an argmax based classification rule.
e.scal(x, k = 1, tc = NULL)
x |
matrix of membership values |
k |
parameter for e-scaling (1 for softmax) |
tc |
vector of true classes (required if |
For any membership vector y exp(y*k) / sum(exp(y*k) is calculated.
If k=1, the classical softmax scaling is used. If the true classes are given, k is optimized
so that the apparent error rate is minimized.
A list containing elements
sv |
Scaled values |
k |
Optimal |
Karsten Luebke, karsten.luebke@fom.de
Garczarek, Ursula Maria (2002): Classification rules in standardized partition spaces. Dissertation, University of Dortmund. URL http://hdl.handle.net/2003/2789
library(MASS) data(iris) ldaobj <- lda(Species ~ ., data = iris) ldapred <- predict(ldaobj)$posterior e.scal(ldapred) e.scal(ldapred, tc = iris$Species)
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.