Displaying 4 results from an estimated 4 matches for "alphaindex".
2010 Sep 24
0
kernlab:ksvm:eps-svr: bug?
...he),
719 as.double(tol),
720 as.integer(shrinking),
721 PACKAGE="kernlab")
722 tmpres <- resv[c(-(m+1),-(m+2))]
723 alpha(ret) <- coef(ret) <- tmpres[tmpres != 0]
724 svindex <- alphaindex(ret) <- which(tmpres != 0)
725 xmatrix(ret) <- x[svindex, ,drop=FALSE]
726 b(ret) <- resv[(m+1)]
727 obj(ret) <- resv[(m+2)]
728 param(ret)$epsilon <- epsilon
729 param(ret)$C <- C
730 }
731
732 ## nu regression
733...
2008 Sep 14
0
ksvm accessing the slots of S4 object
...tors for each of model I am building and store it in a vector.
>ksvm.model <- ksvm(Class ~ ., data = somedata,kernel = "vanilladot", cross = 10, type ="C-svc")
>names(attributes(ksvm.model))
[1] "param" "scaling" "coef" "alphaindex" "b"
[6] "obj" "SVindex" "nSV" "prior" "prob.model"
[11] "alpha" "type" "kernelf" "kpar" "xmatrix"
[16] "ymatrix" "fitted&q...
2011 May 26
0
R svm prediction kernlab
...n't know if its possible or not, how do
classify new data using predict method? Is it possible to build a new ksvm
object using new weights?
Weight calculation is as follows:
svp <- ksvm(x,y,type="C-svc", kernel="vanilladot",C=1)
w <- colSums(coef(svp)[[j]] * x[unlist(alphaindex(svp)[[j]]),])
b <- b(svp)[[j]]
for all j = 1:N(N-1)/2 where N is number of classes
Alternately, I implemented the majority voting myself to perform the
classification , but I am getting slightly different results from
predict.svm method for a case where I am not tuning the weights. I am not
su...
2012 Sep 13
0
I need help for svm package kernlab in R
I use the svm package kernlab .I have two question.
In R
library(kernlab)
m=ksvm(xtrain,ytrain,type="C-svc",kernel=custom function, C=10)
alpha(m)
alphaindex(m)
I can get alpha value and alpha index about package.
1.
Assumption that number of sample are 20.
number of support vectors are 15.
then rest 5`s alphas are 0?
2. I want use kernelMatrix
xtrain=as.matrix(xtrain)
k=KernelMatrix(custom function, xtrain)
it is k equal kernel matrix?
th...