On Oct 27, 2009, at 6:02 AM, Grzes wrote:
>
> Hi everybody!
>
> I want to find a closer neighbourins observation. This is my code:
> ##########################
> library(klaR)
> library(ipred)
> library(mlbench)
> data(PimaIndiansDiabetes2)
> dane=na.omit(PimaIndiansDiabetes2)[,c(2,5,9)]
> dane[,2]=log(dane[,2])
> dane[,1:2]=scale(dane[,1:2])
> zbior.uczacy=sample(1:nrow(dane),nrow(dane)/2,F)
>
> klasyfikatorKNN=ipredknn(diabetes~glucose
> +insulin,data=dane,subset=zbior.uczacy,k=3)
>
> oceny=predict(klasyfikatorKNN,dane[-zbior.uczacy,],"class")
>
> #data frames with my result from klasyfikatorKNN
> df=data.frame(glucose=c(klasyfikatorKNN$learn$X[,
> 1]),insulin=klasyfikatorKNN$learn$X[,2],diabetes=c(klasyfikatorKNN
> $learn$y))
> #And picture
> drawparti(as.factor(df$diabetes), df$glucose, df$insulin, method =
> "sknn",
> prec = 100, xlab = NULL, ylab = NULL)
I get an error: Error: could not find function "drawparti"
>
> ##########################
> My question is: How or where may I find correct or wrong values
> which were
> drawn (found,classification) in this picture?
No picture resulted.
> It means I'm looking for x, y
> values.
Not sure exactly what you are asking. Does this modification to df and
fairly obvious the cross table help?
> df=data.frame(glucose=c(klasyfikatorKNN$learn$X[,
1]),insulin=klasyfikatorKNN$learn$X[,2],pred.diabetes=klasyfikatorKNN
$learn$y, trueDiab=dane[,3])
Warning message:
In data.frame(glucose = c(klasyfikatorKNN$learn$X[, 1]), insulin =
klasyfikatorKNN$learn$X[, :
row names were found from a short variable and have been discarded
> with( df, table(pred.diabetes, trueDiab))
trueDiab
pred.diabetes neg pos
neg 174 86
pos 88 44
>
>
> --
> View this message in context:
http://www.nabble.com/%22ipredknn%22---How-may-I-find-values--tp26074994p26074994.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at 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.
David Winsemius, MD
Heritage Laboratories
West Hartford, CT