[R] LDA for topic modeling in R

[davyfeng]

Hi, All,

I am using the supervised lda function (slda) from 'lda' package in R for
topic modeling (http://cran.r-project.org/web/packages/lda/index.html), my
data is  a collection of documents, and within which each doc has a label.
There are about 97 different categories and 18K documents  in total. I tried
to use the slda for training a model from the data (the whole dataset and
sub_dataset), but failed with some strange problems during the procedure:
Iteration 0
Iteration 1
Iteration 2
Iteration 3
Error in structure(.Call("collapsedGibbsSampler", documents, as.integer(K), 
: 
  Numerical problems (-789.682, 0.0454545).

I am using R 2.15.0 and lda 1.3.1. Any one has any idea?  Thanks very much!





--
View this message in context: http://r.789695.n4.nabble.com/LDA-for-topic-modeling-in-R-tp4639263.html
Sent from the R help mailing list archive at Nabble.com.