Displaying 12 results from an estimated 12 matches for "kpar".
Did you mean:
gpar
2012 Apr 27
2
Where would i put feature requests for a library?
...he code of one of the libraries (not core), or, in my current case, would wish something minor changed for convenience, where can i get contact? Can i put it in the "official" bug repository?
(Problem discription for anyone interested:
Why call the default function kpca for a matrix with kpar=list(sigma=0.2), instead of putting this default sigma into the rbfkernel and using kpar=list()? Anytime i call kpca with a kernel without sigma, i have to supply kpar=list() or get an error.
)
2012 Jul 31
1
kernlab kpca predict
...ats wrong at the moment. I haven't any predict functions written by myself in the workspace either. I've tested it with using the matrix version and the dataframe/formula version on kpca.
data(iris)
test <- sample(1:150,20)
kpc <- kpca(~.,data=iris[-test,-5],kernel="rbfdot",kpar=list(sigma=0.2),features=2)
emb <- predict(kpc,iris[test,-5])
Error in UseMethod("predict") :
no applicable method for 'predict' applied to an object of class "c('kpca', 'prc')"
str(kpc)
Formal class 'kpca' [package "kernlab"] wit...
2010 Sep 24
0
kernlab:ksvm:eps-svr: bug?
...scal <- 1
844 error(ret) <- drop((scal^2)*crossprod(fitted(ret) - y)/m)
845 }
846 }
C. Finally, an example (taken from ?ksvm):
require (kernlab)
seed (1234)
x <- seq(-20,20,0.1); x <- x[x != 0]
y <- sin(x)/x + rnorm(400,sd=0.03)
regm <- ksvm(x,y,epsilon=0.01,kpar=list(sigma=16),cross=3)
te <- crossprod (fitted(regm)-y)/400
s <- (scaling(regm)$y.scale[["scaled:scale"]])^2
error (regm) # 0.03891344
te # 0.0008958718
te * s # 6.37252e-05
te / s # 0.01259449
These numbers can also be seen in the trace_output_concise.t...
2009 Dec 25
2
Help with SVM package Kernlab
...C/G/T" . There are 200 seq's for
training (100 + and - each). this is very similar to the "promotergene" data
set included as example with the package.
The model that I have generated is as follows:
modelforSVM <- ksvm(Class ~ ., data = train500, kernel = "rbfdot", kpar =
"automatic", C = 60, cross = 3, prob.model = TRUE)
The testSeq is a vector of 500 characters casted as a data.frame. I tried
adding the Class column as well later to the testSeq data frame but got the
same error.
I am using R with windows, 32 bit, version 2.9.0
Any help that I can ge...
2007 Sep 12
0
one-class SVM in kernlab
...m() in the kernlab package to fit a one-class
SVC, but I get a strage result on the cross-validation error estimate.
For example, consider this code:
data(spam)
classifier <- ksvm(type~.,data=spam[which(spam[,'type']=='spam'),],
type="one-svc",kernel="rbfdot",kpar=list(sigma=0.1),nu=0.05,cross=10)
what I get is:
> classifier
Support Vector Machine object of class "ksvm"
SV type: one-svc (novelty detection)
parameter : nu = 0.05
Gaussian Radial Basis kernel function.
Hyperparameter : sigma = 0.1
Number of Support Vectors : 660
Objective...
2012 Nov 15
1
Can't see what i did wrong..
with
pred.pca<-predict(splits[[i]]$pca,trainingData at samples)[,1:nPCs]
dframe<-as.data.frame(cbind(pred.pca,class=isExplosive(trainingData,2)));
results[[i]]$classifier<-ksvm(class~.,data=dframe,scaled=T,kernel="polydot",type="C-svc",
C=C,kpar=list(degree=degree,scale=scale,offset=offset),prob.model=T)
and a degree of 5 i get an error of 0 reported by the ksvm object. But when doing
pred.pca<-predict(splits[[i]]$pca,trainingData at samples)[,1:nPCs]
pred.svm<-kernlab::predict(results[[i]]$classifier,pred.pca,type="probabilit...
2007 Aug 14
0
kernlab ksvm() cross-validation prediction response vector
...e support vector classification function ksvm()
the response values stored in object at ymatrix are cross validated outputs/predictions:
Example code from package kernlab, function ksvm:
library(kernlab)
## train a support vector machine
filter <- ksvm(type~.,data=spam,kernel="rbfdot",kpar=list(sigma=0.05),C=5,cross=3)
filter
filter at ymatrix
if not:
what is the easiest way to obtain x-fold cross validated
predicted values of the instances of a data set ?
Does it have to be implemented by oneself?
thanks a lot again
best regards
Bj?rn
2008 Sep 14
0
ksvm accessing the slots of S4 object
...param" "scaling" "coef" "alphaindex" "b"
[6] "obj" "SVindex" "nSV" "prior" "prob.model"
[11] "alpha" "type" "kernelf" "kpar" "xmatrix"
[16] "ymatrix" "fitted" "lev" "nclass" "error"
[21] "cross" "n.action" "terms" "kcall" "class"
>ksvm.model
Support Vector Machi...
2009 Dec 24
0
Error with Package "Kernlab" for SVM prediction
...s: Col 1 is "Class" which
is "+" or "-" and Cols V1 to V500 are "A/C/G/T" . There are 200 seq's for
training (100 + and - each)
The model that I have generated is as follows:
modelforSVM <- ksvm(Class ~ ., data = train500, kernel = "rbfdot", kpar =
"automatic", C = 60, cross = 3, prob.model = TRUE)
The testSeq is a vector of 500 characters casted as a data.frame. I tried
adding the Class column as well later to the testSeq data frame but got the
same error.
I am using R with windows, 32 bit, version 2.9.0
Any help is much appre...
2010 Aug 29
0
SVM comparison
...;m trying to run an epsilon regression model, and am comparing the results
between e1071 and kernlab. I believe that I'm calling the ksvm and svm
functions the same way but I'm getting different results:
library(e1071); library(kernlab)
ksvm(x=1:100, y=(1:100)/5, type="eps-svr", kpar=list(sigma=1))
svm(x=1:100, y=(1:100)/5, type="eps-regression", gamma=0.5)
I get a different number of support vectors and different fitted values.
Am I doing something wrong?
Thanks,
Sean
[[alternative HTML version deleted]]
2009 Sep 06
2
Regarding SVM using R
Hi Abbas,
Before I try to give you answers, I just want to mention that you
should send R related reqests to the R-help list, and not me
personally because (i) there's a greater likelihood that it will get
answered in a timely manner, and (ii) people who might have a similar
problem down the road might benefit from any answer via searching the
list archives ... anyway:
On Sep 5, 2009, at
2010 Oct 31
1
R-help Digest, Vol 92, Issue 31
...bject: [R] ksvm problem
Message-ID: <1288363149599-3019212.post@n4.nabble.com>
Content-Type: text/plain; charset=UTF-8
Hi to all!!!!!
When I use the example from kernlab::ksvm this works fine.. Give me the
result?
> filter <-
> ksvm(type~.,data=spamtrain,kernel="rbfdot",kpar=list(sigma=0.05),C=5,cross=3)
But as soon as I change the type data as follows
> type_train<-spamtrain[,ncol(spamtrain)]
> filter <-
>ksvm(type_train,data=spamtrain,kernel="rbfdot",kpar=list(sigma=0.05),C=5,cross=3)
>)
Error: evaluation nested too deeply: infinite recu...