R help - Jun 2010 - BreastCancer Dataset for Classification in kknn

Dear All,

I'm getting a error while trying to apply the BreastCancer dataset
(package=mlbench) to kknn (package=kknn) that I don't understand as I'm
new
to R.
The codes are as follow:

rm = (list = ls())
library(mlbench)
data(BreastCancer)
library(kknn)

BCancer = na.omit(BreastCancer)
d      = dim(BCancer)[1]
i1 = seq(1, d, 2)
i2 = seq(2, d, 2)

t1     = BCancer[i1, ]
t2     = BCancer[i2, ]
y2  = BCancer[i2, 11]

x     = 10
k     = array(1:x, dim = c(x,1))
ker = array(c( "rectangular", "triangular",
"epanechnikov", "biweight",
                "triweight", "cos", "inv",
"gaussian"), dim = c(8,1))

f     = function(x, ker){

    BreastCancer.kknn  <-  kknn(Class~., train = t1, test = t2, k = x,
            kernel = ker, distance = 1)
    fit = fitted(BreastCancer.kknn)

    z <- (fit==y2)
    z.e <- (100 - (length(y2)-length(z[!z]))/length(y2)*100 )
}

err.k = function(ker){
    error.BreastCancer = apply(k,1,function(y) f(y, ker))
}

err.ker = apply(ker, 1, err.k)
colnames(err.ker) = c("rectangular", "triangular",
"epanechnikov",
"biweight",
        "triweight", "cos", "inv",
"gaussian")
print(err.ker)

It throws a error: Error in as.matrix(learn[, ind == i]) :
  (subscript) logical subscript too long
In addition: Warning messages:
1: In model.matrix.default(mt, mf) : variable 'Id' converted to a factor
2: In model.matrix.default(mt, test) : variable 'Id' converted to a
factor

I tried the codes with other datasets in mlbench package and most of them
working. That is the mistake here for this particular dataset and how can I
solve it?

Thanks
Nitin

	[[alternative HTML version deleted]]

Hi Nitin,

It can be solved by splitting your data a bit different. You need more
training data than you have evaluation data, eg :
i1 = 1:400
i2=401:d

Then it works on my computer. No clue as to where the error originates from
though.

Cheers
Joris

On Tue, Jun 1, 2010 at 4:27 PM, Nitin <nitin29@gmail.com> wrote:
> Dear All,
>
> I'm getting a error while trying to apply the BreastCancer dataset
> (package=mlbench) to kknn (package=kknn) that I don't understand as
I'm new
> to R.
> The codes are as follow:
>
> rm = (list = ls())
> library(mlbench)
> data(BreastCancer)
> library(kknn)
>
> BCancer = na.omit(BreastCancer)
> d      = dim(BCancer)[1]
> i1 = seq(1, d, 2)
> i2 = seq(2, d, 2)
>
> t1     = BCancer[i1, ]
> t2     = BCancer[i2, ]
> y2  = BCancer[i2, 11]
>
> x     = 10
> k     = array(1:x, dim = c(x,1))
> ker = array(c( "rectangular", "triangular",
"epanechnikov", "biweight",
>                "triweight", "cos", "inv",
"gaussian"), dim = c(8,1))
>
> f     = function(x, ker){
>
>    BreastCancer.kknn  <-  kknn(Class~., train = t1, test = t2, k = x,
>            kernel = ker, distance = 1)
>    fit = fitted(BreastCancer.kknn)
>
>    z <- (fit==y2)
>    z.e <- (100 - (length(y2)-length(z[!z]))/length(y2)*100 )
> }
>
> err.k = function(ker){
>    error.BreastCancer = apply(k,1,function(y) f(y, ker))
> }
>
> err.ker = apply(ker, 1, err.k)
> colnames(err.ker) = c("rectangular", "triangular",
"epanechnikov",
> "biweight",
>        "triweight", "cos", "inv",
"gaussian")
> print(err.ker)
>
> It throws a error: Error in as.matrix(learn[, ind == i]) :
>  (subscript) logical subscript too long
> In addition: Warning messages:
> 1: In model.matrix.default(mt, mf) : variable 'Id' converted to a
factor
> 2: In model.matrix.default(mt, test) : variable 'Id' converted to a
factor
>
> I tried the codes with other datasets in mlbench package and most of them
> working. That is the mistake here for this particular dataset and how can I
> solve it?
>
> Thanks
> Nitin
>
>        [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>


-- 
Joris Meys
Statistical Consultant

Ghent University
Faculty of Bioscience Engineering
Department of Applied mathematics, biometrics and process control

Coupure Links 653
B-9000 Gent

tel : +32 9 264 59 87
Joris.Meys@Ugent.be
-------------------------------
Disclaimer : http://helpdesk.ugent.be/e-maildisclaimer.php

	[[alternative HTML version deleted]]