similar to: setting weights for such a two-class problem in nnet andsvm

Dear All, I have such a two-class problem, one class is very large(~98% of total), and the other is just 2%. According to manual of nnet, I need setup "weights", so I intend to set 1 for class one, 49 for class 2. How do I do that? Just weights=49? Meanwhile I'd like to try svm(e1071), again, how do I setup "class.weights"? Thanks. BTW: Many thanks to Jake and Uwe for

One possibility is to look at predict.nnet(), and - Write an R function that write out parts of an nnet object that are needed by predict.nnet() to an external file. - Re-write predict.nnet() in C, reading the model information from the external file. Obviously you'll also need the C source for the code that predict.nnet() calls, and modify those as needed to strip out dependency on R, if

Dear All, I'm learning to train a neural network with my training data by using nnet package, then evaluate it with a evaluation set. My problem here is that, I need the trained network to be used in future, so, what should I store? and How? Any other options other than nnet package? Any example will be highly appreciated! Best, Baoqiang Cao

Hi There, I got some results from using nnet on a two-class problem, and I'd like to hear your comments to understand well about the algorithm. In the training set, the ratio of class 1 to class 2 is about 23:77. I did a 5-fold cross validation. The networks were trained twice, one with 'weights=1', one with 'weights=ifelse(species=="class1", 77/33, 1)'(pointed

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

Dear All, I have a data, suppose it is an N*M matrix data. All I want is to classify it into, let see, 3 classes. Which method(s) do you think is(are) appropriate for this purpose? Any reference will be welcome! Thanks! Best, Baoqiang Cao

Dear All, I am producing a figure with many curves on it. How do I make the legends for all those curves smaller so that it can fit the figure itself? The commands I used for ploting are: plot(x1,y1,col=1,lty=1) lines(x2,y2,col=2,lty=2) ... legend(0.3,0.4,c("name1","name2",...),col=1:20,lty=1:20) Any tips for making the legend fit the figure will very welcome! Thanks! Best,

Dear All, I tried to plot a variety of lines(curves) on same figure. What I did is, plot(x=x1,y=y1) lines(x=x2,y=y2) lines(x=x3,y=y3) ... In my data, the maximum of y1 is much smaller than those maximums of other y vectors. So, in the figure I got, there are some curves which are not complete, I mean, they were cut off at the maximum of y1 at the y axis. Could anybody point out some right

Dear All, I am using wilcox.test to test two samples, data_a and data_b, earch sample has 3 replicates, suppose data_a and data_b are 20*3 matrix. Then I used the following to test the null hypothesis (they are from same distribution.): wilcox.test(x=data_a, y=data_b, alternative="g") I got pvalue = 1.90806170863311e-09. When I switched data_a and data_b by doing the following:

I am using library(nnet) to train up an ANN with what I believe is a moderately sized dataset, but R is complaining about too many weights: --- > nn.1 <- nnet(t(data), targets, size = 4, rang = 0.1, decay = 5e-4, maxit = 200) Error in nnet.default(t(data), targets, size = 4, rang = 0.1, decay = 5e-04, : Too many (1614) weights > dim(targets) [1] 146 2 > dim(data) ## Note

Dear R-users I use nnet for a classification (2 classes) problem. I use the code CVnn1, CVnn2 as described in V&R. The thing I changed to the code is: I define the (class) weight for each observation in each cv 'bag' and give the vector of weights as parameter of nnet(..weights = weight.vector...) Unfortunately I get an error during some (but not all!) inner-fold cv runs:

Hi Everyone, I'm trying to use cutree to get the clusters after hclust. What I used is: mycluster<-cutree(cnclust,h=0.5) Now, my problem is, how can I get the actual clusters? Thanks! Best, Baoqiang Cao

Dear R-users, dear Prof. Ripley as package maintainer I tried to investigate the odd error, when I call nnet together with a 'weights' parameter, using the 'fgl' example in V&R p 348 The error I get is: Error in eval(expr, envir, enclos) : Object "w" not found I think it is a kind of scoping problem, but I really cannot see, what the problem exactly is. and

Dear All, Just start to use the long expected R, my focus will be doing clustering on microarray data, just wonder, anyone can show me any references to conquer the steep learning curve? Thanks! Best regards, Baoqiang Cao

Hi there, I really couldn't find out how to plot histogram with point/line instead of rectangle for each bin? Any help please? Thanks! Best, -Cao

The TITLE for the nnet package is garbled: it comes out as nnet Feed-forward neural networks and multinomial log-linear nnet Feed-forward neural networks and multinomial log-linear models The problem is in maketitle: auk% cat DESCRIPTION Bundle: VR Version: 6.1-9 Date: 2000/07/11 Depends: R (>= 1.1) Author: S original by Venables & Ripley. R port by Brian Ripley

Hi, I am trying to run a nnet algorithm but when I try to use the predict function with type='class', it gives the following error: fit <- nnet(y~., size = 1, data = train.set, rang = 0.5, maxit=200, decay = 0) predict<-predict(fit,test.set,type='class') Error in predict.nnet(fit, test.set, type = "class") : inappropriate fit for class I couldn't figure

Dearest all, Objective: I am now learning neural networks. I want to see how well can train an artificial neural network model to discriminate between the two files I am attaching with this message. http://r.789695.n4.nabble.com/file/n2240582/3dMaskDump.txt 3dMaskDump.txt http://r.789695.n4.nabble.com/file/n2240582/test_vowels.txt test_vowels.txt Question: when I am attempting to run

I hope this question is adequate for this list I use the nnet code from V&R p. 348: The very nice and general function CVnn2() to choose the number of hidden units and the amount of weight decay by an inner cross-validation- with a slight modification to use it for classification (see below). My data has 2 classes with unequal size: 45 observations for classI and 116 obs. for classII With

Dear All, I was trying to format a numeric vector (100*1) by using outd <- format(x=m, sci=F, digits=2) > outd[1:10] [1] " 0.01787758" "-0.14760306" "-0.45806041" "-0.67858525" "-0.64591748" [6] "-0.05918100" "-0.25632276" "-0.15980138" "-0.08359873" "-0.37866688" >m[1:10] [1]