Good point. I'm not sure how I missed that.
This does lead to an additional question:
Is the "probability of the true label" the best prediction to feed to
the ROCR package, or is it better to use the "decision.value"
Anybody have any experience on this one?
Thanks!
-N
On 8/4/09 3:28 AM, Christian Schulz wrote:
> Hi,
>
> you need the score value , have a look at ?svm.predict and in the
> ROCR example.
>
> traindata <- as.data.frame(matrix(runif(1000),ncol=10))
> trainlabels <-
> as.factor(sample(c("win","lose"),nrow(data),replace=T,prob=c(0.5,0.5)))
>
> model <- svm(traindata,trainlabels, type="C-classification",
> kernel="radial", cost=10,
> class.weights=c("win"=3,"lose"=1), scale=FALSE, probability = TRUE)
>
> prediction <- predict(model, traindata, decision.values = TRUE,
> probability = TRUE)
> probs <- attr(prediction, "probabilities")[,1]
> pred <- prediction(probs,trainlabels)
>
> HTH Christian
>
>> Hello,
>>
>> I've come across a strange error...
>>
>>
>> Here is what happens:
>>
>> model <- svm(traindata,trainlabels, type="C-classification",
>> kernel="radial", cost=10, class.weights=c("win"=3,"lose"=1),
>> scale=FALSE, probability = TRUE)
>> predictions <- predict(model, traindata)
>> pred <- prediction(predictions, trainlabels)
>>
>>
>> This returns an error:
>> Error in prediction(predictions, trainlabels) :
>> Format of predictions is invalid.
>>
>> Yet my predictions is just a matrix of predicted labels. Nothing
>> fancy. (In fact, my step follow the exact example on the ROCR
>> homepage.)
>>
>> A search through google for "Format of predictions is invalid"
>> returns zero results.
>>
>> Can anyone suggest how I might fix this problem?
>>
>> Thank You,
>>
>>
>>
>>
>>
>> [[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.
>>
>
[[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.
