[SRILM User List] gtmin and kndiscount

Mohammed Mediani mohammed.mediani at kit.edu
Fri Dec 14 13:21:59 PST 2012


Could anybody please tell me how the discounting parameters for  
modified kneser-ney smoothing (D1, D2, D3+) are computed in case we  
have gtmin parameter greater than 1.
In such case, the corresponding ni would be zero, and we eventually  
have to divide by this ni to get one of the Di's.
Many thanks,


More information about the SRILM-User mailing list