Dear All ,
I have the same problem with IRSTLM that , With the large data it fails . I tried to build 1G Words Language Model using IRSTLM , the IARPA ( * .ilm ) was generated successfully but when trying to compile it to the binary format or the ARPA text format using compile-lm script it fails due to the file size. IRSTLM was compiled on 64-bit machine with Open SUSE 64-bit Linux as an operating system . I appreciate your help. Thanks, Ibrahim Ahmed Mohammedy Senior Software Developer Sakhr Company cid:[email protected]
<<image001.jpg>>
<<image002.gif>>
_______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
