RTextTools on Windows?

[devmag]

Dear all,

I am using RTextTools package on Windows 7 Professional with R version
2.13.

I managed to create a corpus but I basically can not run any models.

If I run:

models <- train_models(corpus, algorithms=c("GLMNET","MAXENT","SVM"))

The error is:
Error in if (outlist$msg != "Unknown error") return(outlist) :
argument is of length zero

Only running
models <- train_models(corpus, algorithms=c("SVM"))

The error is:

Error in SparseM::t(SparseM::t(x[cret$index[1:cret$nr], ])) :
error in evaluating the argument 'x' in selecting a method for
function 't': Error in SparseM::t(x[cret$index[1:cret$nr], ]) :
error in evaluating the argument 'x' in selecting a method for
function 't': Error in `[.matrix.coo`(x, rw, cl) : Subscripts out of
bound

When running

models <- train_models(corpus, algorithms=c("MAXENT"))

most of the time the RGui just crashes with the following error:

Problem Event Name: APPCRASH
Application Name: Rgui.exe
Application Version: 2.130.55427.0
Application Timestamp: 4da5a51a
Fault Module Name: maxent.dll
Fault Module Version: 0.0.0.0
Fault Module Timestamp: 4ee2cc13
Exception Code: c0000005

Do you know what the problem is here? It seems to be loading all
required packages, though some are build under R 2.13.2 - is this the
reason why it wont work?

Thanks a lot for your help!

[Tim Jurka]

Hello,

Do you think you could show the full code you're running? It looks like something might be wrong with the way the data is partitioned.

Best,

Tim

[Brian Abelson]

I am having the same problem on Mac OSx

here is my code:

 library(RTextTools)

#import data

alltweets <- read.csv("~", header=T)

#separate text elements into a character matrix

#genderdata is a 4x53432 character matrix.

genderdata  <- cbind(as.character(alltweets$text),as.character(alltweets$handle),as.character(alltweets$name),alltweets$gender)

#separate out training and test data

train  <- genderdata[1:142,]

test  <- genderdata[143:52432,]

#create document matrix

train1  <- create_matrix(cbind(train[,1:4]), weighting=weightTfIdf)

#create corpus

corpus  <- create_corpus(train1, train1$gender, trainSize=c(1:50,100:120), testSize=c(51:99,121:142), virgin=FALSE) 

#generate models

maxent_model  <- train_model(corpus, "MAXENT")

svm_model  <- train_model(corpus, "SVM")