[R] Train error:: subscript out of bonds

[mxkuhn]

You should try different tuning parameters; the defaults are not likely to work for many datasets. I don't use the polynomial kernel too much but scale parameter values that are really of could cause this. Unlike the rbf, I don't know of any good techniques for estimating this.

Max

On Jan 25, 2011, at 11:47 AM, Neeti <nikkihathi at gmail.com> wrote:

> 
> after using options(error=utils::recover) option, following is the output. if
> i am correct this means that in ksvm there is some problem, but really could
> not understand. could anyone please tell me what is wrong...
> 
> any help will be great ....
> 
> thank you so much..
> 
> 
> Enter a frame number, or 0 to exit   
> 
> 1: train(train1, as.factor(trainset[, ncol(trainset)]), "svmpoly",
> trControl = trainControl((method = "cv"), 10,
> 2: train.default(train1, as.factor(trainset[, ncol(trainset)]), "svmpoly",
> trControl = trainControl((method = "c
> 3: do.call(trControl$computeFunction, argList)
> 4: function (X, FUN, ...) 
> 5: FUN(X[[25]], ...)
> 6: do.call(createModel, args)
> 7: function (data, method, tuneValue, obsLevels, pp = NULL, ...) 
> 8: ksvm(as.matrix(trainX), trainY, kernel = polydot(degree =
> tuneValue$.degree, scale = tuneValue$.scale, offset
> 9: ksvm(as.matrix(trainX), trainY, kernel = polydot(degree =
> tuneValue$.degree, scale = tuneValue$.scale, offset
> 10: .local(x, ...)
> 11: ksvm(x[c(indexes[[i]], indexes[[j]]), , drop = FALSE][cind, ], yd[cind],
> type = type(ret), kernel = kernel, k
> 12: ksvm(x[c(indexes[[i]], indexes[[j]]), , drop = FALSE][cind, ], yd[cind],
> type = type(ret), kernel = kernel, k
> 13: .local(x, ...)
> 
> -- 
> View this message in context: http://r.789695.n4.nabble.com/Train-error-subscript-out-of-bonds-tp3234510p3236494.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.