conjugateprior/austin

Do things with words. Scale them, mostly.

austin's Issues

Problem in installing austin

Hi. I am trying to install "austin" package to R 3.4.2, but I receive the following message:

Warning in install.packages :
package ‘austin’ is not available (for R version 3.4.2)

Is the package available only on very old versions of R?
I think it is not given the recent postings on this git hub.
Can you please help me with this problem shooting?
Thank you!

Run out of memory

When importing a documents of 40'000 row, the system crash when running wordfish.

Error: vector memory exhausted (limit reached?)

svd error when trying to run wordfish

I have a 2x900 csv with doc_id and text and try to run wordfish on it to see if I can identify some specific term.

# Load data
data = read.csv("all.mbox.csv", stringsAsFactors = FALSE)
meta = read.csv("all.mbox.meta.csv", stringsAsFactors = FALSE)

mycorpus = Corpus(DataframeSource(data))

# Prepare data
mycorpus = tm_map(mycorpus, content_transformer(tolower))
mycorpus = tm_map(mycorpus, removeWords, stopwords("english"))
mycorpus = tm_map(mycorpus, stemDocument)
mycorpus = tm_map(mycorpus, removePunctuation)
mycorpus = tm_map(mycorpus, removeNumbers)

# Make a document term matrix
dtm = DocumentTermMatrix(mycorpus)
# remove rare words
dtm = removeSparseTerms(dtm, sparse = 0.4)
inspect(dtm)

# Run word fish
dtm$dimnames$Docs
wfm = as.wfm(dtm)
wf = wordfish(wfm, dir=c(2, 34), control = list(tol = .00003))

I run in this error with the last line:

Error in svd(ystarm, nu = 1) : infinite or missing values in 'x'