Class expansion

Tanel Alumäe tanel.alumae at aqris.com
Thu Feb 19 07:13:56 PST 2004


Hello,

I'm trying to to convert a class bigram to its equivalent word n-gram,
using the "ngram" tool with the -expand-classes option. The class model
has 1000 classes, and there are 60000 words. I use the following command
line:

ngram -lm <classmodel> -classes <classesfile> -expand-classes 2
-write-lm <outputmodel>

The process runs about 15 minutes using over 700M of RAM, and then gets
killed by the OS (I'm using Linux), probably when it asked even more
memory that the OS didn't have (I have 512M of main memory).

Is it normal that the class expansion takes that much RAM? Is there a
way around it?

Thanks and regards,

-- 
Tanel Alumäe <tanel.alumae at aqris.com>




More information about the SRILM-User mailing list