similar to: Problems creating packages.

Hello! I have a function: zywnoscCalosc<- function( jedzenie, n1, n2, n3, n4, d1, d2, d3, d4 ) { ndf <- data.frame(nn1=n1,nn2=n2,nn3=n3,nn4=n4) ddf <- data.frame(dd1=d1,dd2=d2,dd3=d3,dd4=d4) for (i in 1:length(n1)){ wekt_n = ndf[i,] wekt_n_ok = wekt_n[!is.na(wekt_n)] dl_n = length(wekt_n_ok) wynik = (1*wekt_n_ok)/(1*dl_n) } } and I get an error like this: Error in 1 * wekt_n_ok :

Hello, I want to know what do you think about my function. I know that it isn't briliant :/ but what do you think? What I should do that my function will be better? (now is very slow and not ideal, sometimes I also get a mistake!) ########## My function ############################################# dzieci<-transform(dzieci, zywnosc=0) zywnoscCalosc<- function( jedzenie, sklep, n1, n2,

Dear all, Objective: I am trying to learn about neural networks. I want to see if i can train an artificial neural network model to discriminate between spam and nonspam emails. Problem: I created my own model (example 1 below) and got an error of about 7.7%. I created the same model using the Rattle package (example 2 below, based on rattles log script) and got a much better error of about

Hi all, I'm running a monte carlo test of a neural network tool I've developed, and it looks like it's going to take a very long time if I run it in R so I'm interested in translating my code (included below) into something faster like Fortran (which I'll have to learn from scratch). However, as you'll see my code loads the nnet library and uses it quite a bit, and I

Dear all I am trying to compare the performances of several methods using the AUC0.1 and not the whole AUC. (meaning I wanted to compare to AUC's whose x axis only goes to 0.1 not 1) I came to know about the Mcneil Hanley test from Bernardo Rangel Tura and I referred to the original paper for the calculation of "r" which is an argument of the function cROC. I can only find the

Hi, Dear all, I am just starting using R in my work and got some trouble to figure out some of the errors. Can anybody help me? The following is the script: read.csv('pupil.txt',header=TRUE,sep='\t')->pupil samp<-c(1:50, 112:162, 171:220, 228:278) pupil.nn2 <- nnet(Type ~ ., data = pupil, subset = samp, size = 2, rang = 0.1, decay = 5e-4, maxit = 200)

Dear all, I try to compare the performances of several parameters to diagnose lameness in dogs. I have several ROC curves from the same dataset. I plotted the ROC curves and calculated AUC with the ROCR package. I would like to compare the AUC. I used the following program I found on R-help archives : From: Bernardo Rangel Tura Date: Thu 16 Dec 2004 - 07:30:37 EST

I might be missing something but I thought that AUC was a measure for comparing ROC curves, so there is nothing else needed to "compare" them. The larger AUC is the higher correlation of 2 variables compared. No other measures or calculations are needed. Jarek Tuszynski -----Original Message----- From: r-help-bounces at stat.math.ethz.ch [mailto:r-help-bounces at stat.math.ethz.ch] On

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

Please help. I found the code in the archive. The author of this script says: "The first function (seROC) calculate the standard error of ROC curve, the second function (cROC) compare ROC curves." Can some one explain to me what are the na, nn and r parameters which are used as the input to the following two functions? Thanks much in advance. > From: Bernardo Rangel Tura >

I was wondering if there is a way in R to find k nearest neighbors of various orders, say order 2, 3, or 4. In otherwords neighbors of neighbors of neighbors. You get the idea. I know that I can use knearneigh(matrix.data, k) but this only gives me the k nearest neighbors and not of a particular order. Thanks in advance. -- View this message in context:

Hi All, I am trying to use a neural network for my work, but I am not sure about my approach to select a parsimonious model. In R with nnet, the IAC has not been defined for a feed-forward neural network with a single hidden layer. Is this because it does not make sens mathematically in this case? For example, is this pseudo code sensible? Thanks in advance for your help. I am sorry if this

Hello, I want to use nnet package in R, to train and simulate a NN and get the value of MSE. I am reading in a file which has 19 input variables and one output variable and has a total of 2000 observations. The first column in the file is a column just for giving the serial numbers of the observations. I have already read in the file and also extracted the different values into the matrices to

Dear all This is a follow-up to an earlier posting today regarding a regular expression question. In the meantime, this is the best approximation I could come up with and should give you a better idea what I am talking about. a<-c("<w AT0>a <w NN1>blockage <w CJC>and <w DT0>that<c PUN>.", "<w AT0>a <w NN1>blockage <w CJC>and

As a follow-up to my email below: The input data frame to nnet() has dimensions: > dim(coreaff.trn.nn) [1] 5088 8 And the predictions from the neural network (35 records are dropped - see email below for more details) has dimensions: > pred <- predict(coreaff.nn1) > dim(pred) [1] 5053 1 So, the following line of R code does not work as the dimensions are

Sorry for these multiple postings. I solved the problem using na.omit() to drop records with missing values for the time being. I will worry about imputation, etc. later. I calculated the sum of squared errors for 3 models, linear regression, neural networks, and support vector machines. This is the first run. Without doing any parameter tuning on the SVM or playing around with the number of

I am exploring neural networks (adding non-linearities) to see if I can get more predictive power than a linear regression model I built. I am using the function nnet and following the example of Venables and Ripley, in Modern Applied Statistics with S, on pages 246 to 249. I have standardized variables (z-scores) such as assets, age and tenure. I have other variables that are binary (0 or 1). In

Hi! I'm displaying a contingency table with heatmap(): > svm.predPix.tabla svm.predPix CC DD LL NN NN2 CC 22 0 3 8 3 DD 0 27 0 1 0 LL 1 1 90 3 7 NN 2 0 1 11 4 NN2 0 0 5 1 20 > heatmap(svm.predPix.tabla[5:1,], Rowv=NA, Colv=NA,col = rev(heat.colors(32)), scale="column", margins=c(5,10)) and I'm happy

"lpt1" through "lpt9" and "com1" through "com9" are in fact "special" file names under various versions of Microsoft Windows. However, "R Extensions" says only that "lpt1" through "lpt4" and "com1" through "com3" cannot be used as filenames, and "R CMD check" checks for just these

I would like to use samp as a part of a filename that I can change. My source files are .csv files with date as the file name, and I would like to be able to type in the date (later perhaps automate this using list.files) and then read the csv and write the pdf automatically. I have tried different combinations with "" and () around samp, but I keep getting the error "object