classifOutput-class {MLInterfaces} | R Documentation |
container with uniform contents for classifier outputs
Objects can be created by calls of the form new("classifOutput", ...)
.
These objects include label vectors for test set elements, scores on labeling
events, call history information, native classifier object access,
and a distance matrix for test set feature vectors.
method
:"character"
tag identifying
classification method in use predLabels
:"MLLabel"
vector
of labels predicted for test cases predScores
:"MLScore"
structure
of scores on classification events (for example, vote proportions
or fitted probabilities of class membership
call
:"call"
call engendering the
classifOutput object classifRobject
:"ANY"
fitted model
object distMat
:"dist"
distance matrix
of test set feature vectors signature(object = "classifOutput")
: ... signature(out = "classifOutput")
: ... VJ Carey <stvjc@channing.harvard.edu>
# access and trim an ExpressionSet library(golubEsets) data(Golub_Merge) smallG <- Golub_Merge[1:60,] # set a PRNG seed for reproducibilitiy set.seed(1234) # needed for nnet initialization # now run the classifiers f1 = knnB( smallG, "ALL.AML", 1:40 ) f1 confuMat(f1)