pamr.predict {pamr} | R Documentation |
A function producing predicted information, from a nearest shrunken centroid fit
pamr.predict(fit, newx, threshold, type= c("class", "posterior", "centroid", "nonzero"), prior = object$prior, threshold.scale = object$ threshold.scale)
fit |
The result of a call to pamr.train |
newx |
Input (feature) matrix for which predictions are desired |
threshold |
The desired threshold value |
type |
Type of prediction desired: class predictions, posterior probabilities, (unshrunken) class centroids, vector of genes surviving the threshold |
prior |
Prior probabilities for each class. Default is that specified in "fit" |
threshold.scale |
Additional scaling factors to be applied to the thresholds. Vector of length equal to the number of classes. Default is that specified in "fit". |
pamr.predict
Give a cross-tabulation of true versus
predicted classes for the fit returned by pamr.train or pamr.cv,
at the specified threshold
Trevor Hastie, Robert Tibshirani, Balasubramanian Narasimhan, and Gilbert Chu
set.seed(120) x <- matrix(rnorm(1000*20),ncol=20) y <- sample(c(1:4),size=20,replace=TRUE) mydata <- list(x=x,y=y) mytrain <- pamr.train(mydata) mycv <- pamr.cv(mytrain,mydata) pamr.predict(mytrain, mydata$x , threshold=1)