first number: observed "normal" BLEU score second and third number: lower and upper bound of confidence interval at confidence level choses (.95 by default, I think)
fourth number median BLEU score Are the numbers you are giving actual numbers? They would mean that the observed BLEU score is an extreme outlier. This does not make sense to me. The first number should be close to the fourth. Was that be 21.26, maybe? - Uli On Thu, Jun 25, 2015 at 5:28 PM, Saeed Farzi <[email protected]> wrote: > Tnx Ulrich. > > Would please tell me about its output > > ex. 23.26 multi-ref [20.29-21.71; 21.27] > > On Tue, Jun 9, 2015 at 2:40 AM, Ulrich Germann <[email protected]> > wrote: > >> scripts/generic/bsbleu.py -h >> >> - Uli >> >> On Sun, Jun 7, 2015 at 2:40 PM, Saeed Farzi <[email protected]> wrote: >> >>> Dear all, >>> I need a script for testing the statistical significance using bootstrap >>> re-sampling . I heard that there is a tool in Mmoses toolkit for this. I >>> need some information about it and how to use it. >>> >>> tnx in advance, >>> bests, >>> -- >>> S.Farzi, Ph.D. Student >>> Natural Language Processing Lab, >>> School of Electrical and Computer Eng., >>> Tehran University >>> Tel: +9821-6111-9719 >>> >>> >>> >>> _______________________________________________ >>> Moses-support mailing list >>> [email protected] >>> http://mailman.mit.edu/mailman/listinfo/moses-support >>> >>> >> >> >> -- >> Ulrich Germann >> Senior Researcher >> School of Informatics >> University of Edinburgh >> > > > > -- > S.Farzi, Ph.D. Student > Natural Language Processing Lab, > School of Electrical and Computer Eng., > Tehran University > Tel: +9821-6111-9719 > > > -- Ulrich Germann Senior Researcher School of Informatics University of Edinburgh
_______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
