similar to: bug in predict.nnet?

Full_Name: Jeff Schwarz Version: R1.3.1 OS: Windows 2000 Submission from: (NULL) (129.22.170.115) Error message (using predict and predict.nnet) > predict (smalltest, smallx[-jj,]) Error in matrix(NA, length(keep), nout, dimnames = list(rn, dimnames(object$fitted)[[2]])) : length of dimnames[1] not equal to array extent *** all relevant code and data source is given below *** I

Hi all I have a dataset whose response is a categorical variable of the ordinal scale type (6 levels). I'm interested in building classification models, and I'm wondering if there is something implemented in R (or its packages) that I'm not aware of, to treat the ordinal scale measurements straightforwardly. I can think of the alternative of building a conditional (hierarchical)

Hello I have several replicates (30) of a -- short -- time series (19 obs) and I'd like to estimate the first few lags autocorrelation function. Is there an immediate way of handling multiple realizations when computing the ACF in R? I'm using R-1.4.1 on a Linux Debian box. Thanks in advance claudia ------------------------------------------------------------------------- claudia

Dear R-helpers, I'm plotting geophysical data in the form of contours using "filled.contour". The display would be much more effective if the areas with negative values could be color coded by -- say -- "cold colors" in the range of blue to green, and conversely the areas with positive values got plotted with "warm colors", from yellow to red. Right now if I use

Hello I'm upgrading from R-1.4.1 to R-1.5.1, locally on my Linux PC running Debian. after running 'configure' this is what I get (just in case it helps figuring out what follows later...) R is now configured for i686-pc-linux-gnu Source directory: . Installation directory: /usr/local C compiler: gcc -D__NO_MATH_INLINES -mieee-fp -g -O2 C++

I'm running R-1.4.1 on a linux debian (version 2.2) box. I've tried to upgrade to R-1.6.0, and the execution of make halts with the following error messages: make[4]: Entering directory `/home/tebaldi/R/R-1.6.0/src/library/methods' dumping R code in package 'methods' Error in testRversion(descfile) : This package has not been installed properly See the Note in ?library

There is probably an elegant and efficient way to do this -- perhaps already implemented, and some of you knows it. Myself, I'm getting stuck in long loops ... Here is the question: I have a long (80,000 + obs) binary time series. For each of its elements I'd like to compute and save in a vector "the number of steps 'til the first switch of state happens". For example,

Hi all I'm running lm()/leaps() kind of computations within a loop, and I submitted the entire thing by sourcing a file of R code in interactive mode. I'd like to interrupt the evaluation of this (it's taking way too long) but I am not able to. I'm running R-1.3.0 from an emacs window (ESS-5.0) and I'm on a linux box, (Debian). I tried Control-c Control-c (nothing happens),

I'd like to apply the ideas of Pinheiro et al. (1997): "Robust Estimation in LInear Mixed Effects Models using the Multivariate t-Distribution" and I'm wondering if the EM-based estimation described in that paper has been implemented by somebody in R already (I'm assuming it's not yet available within the package nlme -- whose subtitle is "Fit and compare

Hi I'm trying to print several plots to a multi-page postscript file. Some plots are output of basic commands, some are produced by "trellis" commands (from the package lattice). Right now I'm not able to get a straightforward black and white color scheme for the latter kind: when I open the postscript file the "trellis" plots are invisible, or parts of them are

Here is the answer to an old question of mine, regarding maps in R Hope this help. claudia ---------- Forwarded message ---------- Date: Mon, 25 Jun 2001 13:48:05 -0400 From: Kieran Healy <kjhealy at Princeton.EDU> To: Claudia Tebaldi <tebaldi at ucar.edu> Subject: Re: [R] maps in R Hi Claudia -- you can get the maps() library from here:

Hello Is there a simple way of fitting a 'smooth curve' to a time series with the additional constraint that the first derivative be not negative? I know that if I choose a "large enough" window for a moving average sort of filter on my data I end up with a non-decreasing curve anyway, but I'm wondering if there is a pre-built method that can include the constraint from

Hi all Just this morning I upgraded to R 2.12.0 (for Mac OS X 10.6.4). All went well until I needed to run a help() or help.search() in my session, which I'm running within Emacs (ESS 5.3.7). Say I need help with the command 'density'. When I type help(density) or ?density the ESS help buffer opens, it is titled *help[R](density)* but it contains only a couple of lines saying,

Dear R-helpers, I have lat/lon coordinates of regularly spaced grid points, about 4Km apart, covering the entire US continental region. I would like to mask this rectangular grid in order to extract all and only the grid points within a specific region. Today I want to extract Montana, say, from this grid, and I am hoping to somehow use the returned value of the function

HI, Dear R community, My data set has 2409 variables, the last one is response variable. I have used the nnet after feature selection and works. But this time, I am using nnet to fit a model without feature selection. I got the following error information: > dim(train) [1] 1827 2409 nnet.fit<-nnet(as.factor(out) ~ ., data=train, size=3, rang=0.3, decay=5e-4, maxit=500) # model

Hello I'm having trouble figuring out how to use the output of "segmented()" with a new set of predictor values. Using the example of the help file: ??set.seed(12) xx<-1:100 zz<-runif(100) yy<-2+1.5*pmax(xx-35,0)-1.5*pmax(xx-70,0)+15*pmax(zz-.5,0)+rnorm(100,0,2) dati<-data.frame(x=xx,y=yy,z=zz) out.lm<-lm(y~x,data=dati) o<-## S3

Hi, I just want to point out a discrepancy between the documentation of predict.nnet & the function definition. >?predict.nnet => predict.nnet package:nnet R Documentation Predict New Examples by a Trained Neural Net Description: Predict new examples by a trained neural net. Usage: predict.nnet(object, x,

Greetings, I am having trouble calculating artificial neural network misclassification errors using errorest() from the ipred package. I have had no problems estimating the values with randomForest() or svm(), but can't seem to get it to work with nnet(). I believe this is due to the output of the predict.nnet() function within cv.factor(). Below is a quick example of the problem I'm

Dear All, After followed the help of nnet, I could get the networks trained and, excitedly, get the prediction for other samples. It is a two classes data set, I used "N" and "P" to label the two. My question is, how do I get the predicted numerical value for each sample? Not just give me the label(either "N" or "P")? Thanks! FYI: The nnet example I

Hi, I have a problem with predict.nnet when I try to use it. It crashes R with a memory access violation. platform Windows arch x86 os Win32 system x86, Win32 status major 1 minor 1.1 year 2000 month August day 15 language R I admit the data set is quite large ~ [3000, 101] and its a 3 class problem. I know that it works fine when there is a single target. I wonder