Hi, can you take a look at the files corpus.lowercased.* ? There seems to be a mismatch between those files and the GIZA++ output.
-phi On Mon, Nov 12, 2012 at 6:58 AM, Cuong Hoang <[email protected]> wrote: > I attach the output when I run MOSES. > I want to make a note that the quality of my toolkit is good since I > reference GIZA++ > from every aspect before coding. > To besides, the format of the word alignment output is correct. > However, I stuck with the problem here! > > On Mon, Nov 12, 2012 at 10:42 PM, Cuong Hoang <[email protected]> > wrote: >> >> Hi all, >> These days I've been stuck with a very fuzzy error. >> I've been coding an IBM models training toolkit (1-3) and now finishing >> IBM Model 4. >> The output of my toolkit is exactly as the MOSES specification as >> described: http://www.statmt.org/moses/?n=FactoredTraining.RunGIZA >> For example: >> >> # Sentence pair (1) source length 4 target length 3 alignment score : >> 0.00643931 >> wiederaufnahme der sitzungsperiode >> NULL ({ }) resumption ({ 1 }) of ({ }) the ({ 2 }) session ({ 3 }) >> # Sentence pair (2) source length 17 target length 18 alignment score : >> 1.74092e-26 >> ich erklaere die am donnerstag , den 28. maerz 1996 unterbrochene >> sitzungsperiode >> des europaeischen parlaments fuer wiederaufgenommen . >> >> >> >> >> Now, I've been stuck with the very strange error alignment point out of >> range. For example: >> >> !alignment point (42,23) out of range (0-31,0-25) in line 1, ignoring >> alignment point (33,24) out of range (0-31,0-25) in line 1, ignoring >> alignment point (28,26) out of range (0-31,0-25) in line 1, ignoring >> alignment point (30,28) out of range (0-31,0-25) in line 1, ignoring >> alignment point (29,29) out of range (0-31,0-25) in line 1, ignoring >> alignment point (32,29) out of range (0-31,0-25) in line 1, ignoring >> alignment point (31,30) out of range (0-31,0-25) in line 1, ignoring >> alignment point (34,31) out of range (0-31,0-25) in line 1, ignoring >> alignment point (37,34) out of range (0-31,0-25) in line 1, ignoring >> alignment point (42,35) out of range (0-31,0-25) in line 1, ignoring >> alignment point (41,36) out of range (0-31,0-25) in line 1, ignoring >> alignment point (35,37) out of range (0-31,0-25) in line 1, ignoring >> alignment point (42,37) out of range (0-31,0-25) in line 1, ignoring >> alignment point (36,38) out of range (0-31,0-25) in line 1, ignoring >> alignment point (43,39) out of range (0-31,0-25) in line 1, ignoring >> >> However, the outputs of my toolkit of the pair 1 are: >> From English-French: >> # Sentence pair (1) source length 40 target length 44 alignment score : >> 1.0 >> à partir de la fin du xixe siècle , la découverte du spectre >> électromagnétique et du monde de l 'atome va aussi mener à l 'apparition d >> 'une nouvelle branche de l 'astronomie , la plus importante de nos jours : l >> 'astrophysique . >> NULL ({ }) from ({ 1 }) the ({ 4 }) 19th ({ 41 }) century ({ 8 }) onwards >> ({ 2 5 7 }) , ({ 9 }) the ({ 6 }) discovery ({ 11 }) of ({ 12 }) the ({ 10 >> }) electromagnetic ({ 14 }) spectrum ({ 13 }) and ({ 15 }) the ({ 16 }) >> world ({ 17 }) of ({ 18 }) the ({ 19 }) atom ({ 20 39 }) spurred ({ 21 22 40 >> }) on ({ 24 }) the ({ 25 }) development ({ }) of ({ 3 }) astrophysics ({ 23 >> 26 43 }) , ({ 34 }) a ({ 27 28 }) new ({ 29 }) discipline ({ }) in ({ 31 }) >> astronomy ({ 30 33 }) that ({ 32 }) is ({ 35 }) now ({ }) considered ({ }) >> to ({ 38 }) be ({ }) the ({ 42 }) most ({ 36 }) important ({ 37 }) . ({ 44 >> }) >> >> From French-English >> # Sentence pair (1) source length 44 target length 40 alignment score : >> 1.0 >> from the 19th century onwards , the discovery of the electromagnetic >> spectrum and the world of the atom spurred on the development of >> astrophysics , a new discipline in astronomy that is now considered to be >> the most important . >> NULL ({ }) à ({ }) partir ({ 1 }) de ({ }) la ({ 2 }) fin ({ }) du ({ 7 }) >> xixe ({ 3 5 }) siècle ({ 4 }) , ({ 6 }) la ({ 10 }) découverte ({ 8 }) du ({ >> 9 }) spectre ({ 12 }) électromagnétique ({ 11 34 }) et ({ 13 }) du ({ 14 }) >> monde ({ 15 }) de ({ 16 }) l ({ 17 }) 'atome ({ 18 19 33 }) va ({ 20 32 }) >> aussi ({ }) mener ({ }) à ({ }) l ({ 21 }) 'apparition ({ 22 }) d ({ 23 }) >> 'une ({ 26 }) nouvelle ({ 27 }) branche ({ }) de ({ 29 }) l ({ }) >> 'astronomie ({ 28 30 }) , ({ 25 }) la ({ 31 }) plus ({ }) importante ({ 39 >> }) de ({ 35 }) nos ({ }) jours ({ }) : ({ }) l ({ 37 }) 'astrophysique ({ 24 >> 36 38 }) . ({ 40 }) >> >> So, I just wonder what's the exactly problem here since the output of word >> alignment is very normal? >> Thanks, >> best regards, >> C. Hoang >> >> -- >> Hoàng Cường >> SMTNerd >> > > > > -- > Hoàng Cường > SMTNerd > > > _______________________________________________ > Moses-support mailing list > [email protected] > http://mailman.mit.edu/mailman/listinfo/moses-support > _______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
