[R] Different number of resamples error
PIKAL Petr
petr@p|k@| @end|ng |rom prechez@@cz
Fri Feb 21 10:06:01 CET 2020
Hi
your code is not reproducible.
I get error with
> setwd("C:/Users/PC/Documents")
Error in setwd("C:/Users/PC/Documents") : cannot change working directory
>
so probably any other line of your code gives me error too.
Use dput(d) or dput(head(d)) to provide your data
Cheers
Petr
> -----Original Message-----
> From: R-help <r-help-bounces using r-project.org> On Behalf Of javed khan
> Sent: Friday, February 21, 2020 10:00 AM
> To: Jeff Newmiller <jdnewmil using dcn.davis.ca.us>
> Cc: R-help <r-help using r-project.org>
> Subject: Re: [R] Different number of resamples error
>
> The whole code is as follows:
>
> library(caret)
> library(farff)
> library(gbm)
> library(nnet)
>
> setwd("C:/Users/PC/Documents")
> d=readARFF("myresults.arff")
>
> index <- createDataPartition(d$results, p = .70,list = FALSE) tr <-
d[index, ] ts <-
> d[-index, ]
> index_2 <- createFolds(tr$results, returnTrain = TRUE, list = TRUE) ctrl
<-
> trainControl(method = "repeatedcv", index = index_2)
>
> set.seed(30218)
> nnet1 <- train(results~ ., data = tr,
> method = "nnet",
>
> metric = "MAE",
> trControl = ctrl,
>
> preProc = c("center", "scale", "zv"),
> tuneGrid = data.frame(decay = (1),
> size = (1.3801517))) nnet1$results
>
> ///For SVM
>
> set.seed(30218)
> svm1 <- train(results ~ ., data = tr,
> method = "svmRadial",
>
> metric = "MAE",
> preProc = c("center", "scale", "zv"),
> trControl = ctrl,
> tuneGrid=expand.grid(sigma = (0.5),
> C = c(1.348657)))
> getTrainPerf(svm1)
> svm1$results
>
> //For GBM
>
> set.seed(30218)
> gbm <- train(results ~ ., data = tr,
> method = "gbm",
> preProc = c("center", "scale", "zv"),
> metric = "MAE",
>
>
> tuneGrid = data.frame(n.trees = (200.09633523),
interaction.depth =
> (1),
> shrinkage=(0.1), n.minobsinnode=(10)))
gbm$results
>
> //Then the boxplot
>
> rvalues=resamples(list(nnet=nnet1, svm=svm1, GBM=gbm))
>
> bwplot(rvalues, metric="MAE")
>
>
> On Fri, Feb 21, 2020 at 12:16 AM Jeff Newmiller <jdnewmil using dcn.davis.ca.us>
> wrote:
>
> > You are being way too cavalier about what packages you are using. Read
> > the Posting Guide about contributed packages... this list cannot
> > provide expert support for every package out there. This confusion is
> > why you should be providing a reproducible example when you ask for help
> about R.
> >
> > The caret package depends on lattice and provides some overloaded
> > versions of the bwplot function that do have a metric argument. I have
> > no expertise with caret myself... but recommend that you supply a
> > reproducible example for best luck in prompting someone to look closer.
> >
> > On February 20, 2020 1:31:30 PM PST, javed khan
> > <javedbtk111 using gmail.com>
> > wrote:
> > >Thanks for your reply.
> > >
> > >I am not using any specific package for bwplot. I just used caret,
> > >nnet and gbm packages.
> > >
> > >When I use resample (instead of resamples), it give me error message.
> > >
> > >metric=MAE gives the MAE values at x-axis when I used simple plots in
> > >the recent past.
> > >
> > >Best regards
> > >
> > >On Thu, Feb 20, 2020 at 10:29 PM Bert Gunter <bgunter.4567 using gmail.com>
> > >wrote:
> > >
> > >> cc the list!
> > >> (which I have done here)
> > >>
> > >> Bert Gunter
> > >>
> > >> "The trouble with having an open mind is that people keep coming
> > >along and
> > >> sticking things into it."
> > >> -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )
> > >>
> > >>
> > >> On Thu, Feb 20, 2020 at 1:20 PM javed khan <javedbtk111 using gmail.com>
> > >wrote:
> > >>
> > >>> Thanks for your reply.
> > >>>
> > >>> I am not using any specific package for bwplot. I just used caret,
> > >nnet
> > >>> and gbm packages.
> > >>>
> > >>> When I use resample (instead of resamples), it give me error
> > >message.
> > >>>
> > >>> metric=MAE gives the MAE values at x-axis when I used simple plots
> > >in the
> > >>> recent past.
> > >>>
> > >>> Best regards
> > >>>
> > >>> On Thu, Feb 20, 2020 at 10:15 PM Bert Gunter
> > ><bgunter.4567 using gmail.com>
> > >>> wrote:
> > >>>
> > >>>> ??
> > >>>> Isn't is resample() not resamples()?
> > >>>> From what package?
> > >>>> What package is bwplot from? lattice:::bwplot has no "metric"
> > >argument.
> > >>>>
> > >>>>
> > >>>>
> > >>>> Bert Gunter
> > >>>>
> > >>>> "The trouble with having an open mind is that people keep coming
> > >along
> > >>>> and sticking things into it."
> > >>>> -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip
> > >>>> )
> > >>>>
> > >>>>
> > >>>> On Thu, Feb 20, 2020 at 12:55 PM javed khan
> > >>>> <javedbtk111 using gmail.com>
> > >>>> wrote:
> > >>>>
> > >>>>> Hello to all
> > >>>>>
> > >>>>> I have different train functions for NN, SVM and GBM and when I
> > >combine
> > >>>>> the
> > >>>>> results using bwplot, it gives me the error " Different number
> > >>>>> of resamples in each model". It gives me the results (MAE
> > >>>>> values) but using the boxplot, it gives the error. The code is
> > >>>>> as follows:
> > >>>>>
> > >>>>> set.seed(30218)
> > >>>>> nnet1 <- train(results~ ., data = tr,
> > >>>>> method = "nnet",
> > >>>>>
> > >>>>> metric = "MAE",
> > >>>>> trControl = ctrl,
> > >>>>>
> > >>>>> preProc = c("center", "scale", "zv"),
> > >>>>> tuneGrid = data.frame(decay = (1),
> > >>>>> size = (1.3801517)))
> > >>>>> nnet1$results
> > >>>>>
> > >>>>> ///For SVM
> > >>>>>
> > >>>>> set.seed(30218)
> > >>>>> svm1 <- train(results ~ ., data = tr,
> > >>>>> method = "svmRadial",
> > >>>>>
> > >>>>> metric = "MAE",
> > >>>>> preProc = c("center", "scale", "zv"),
> > >>>>> trControl = ctrl,
> > >>>>> tuneGrid=expand.grid(sigma = (0.5),
> > >>>>> C =
> > >>>>> c(1.348657)))
> > >>>>> getTrainPerf(svm1)
> > >>>>> svm1$results
> > >>>>>
> > >>>>> //For GBM
> > >>>>>
> > >>>>> set.seed(30218)
> > >>>>> gbm <- train(results ~ ., data = tr,
> > >>>>> method = "gbm",
> > >>>>> preProc = c("center", "scale", "zv"),
> > >>>>> metric = "MAE",
> > >>>>>
> > >>>>>
> > >>>>> tuneGrid = data.frame(n.trees = (200.09633523),
> > >>>>> interaction.depth = (1),
> > >>>>> shrinkage=(0.1),
> > >>>>> n.minobsinnode=(10)))
> > >>>>> gbm$results
> > >>>>>
> > >>>>> //Then the boxplot
> > >>>>>
> > >>>>> rvalues=resamples(list(nnet=nnet1, svm=svm1, GBM=gbm))
> > >>>>>
> > >>>>> bwplot(rvalues, metric="MAE")
> > >>>>>
> > >>>>> [[alternative HTML version deleted]]
> > >>>>>
> > >>>>> ______________________________________________
> > >>>>> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more,
> > >>>>> see 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 using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> > >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.
> >
> > --
> > Sent from my phone. Please excuse my brevity.
> >
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.
More information about the R-help
mailing list