Kazakh and Tatar/Testvoc
Jump to navigation
Jump to search
[22:53:43] <spectie> go to [22:53:44] <spectie> LEXICON Root [22:53:52] <spectie> comment out everything except Pronouns [22:54:26] *** Quits: jacobEo (~Jacob@85.233.243.23) (Ping timeout: 265 seconds) [22:55:24] <selimcan> spectie, and then run testvoc.sh? [22:55:36] <spectie> yeah [22:55:42] <spectie> non o [22:55:43] <spectie> wait [22:56:10] <spectie> we can probably get away withou doing that [22:57:50] <spectie> $ hfst-fst2strings kaz-tat.automorf.hfst | grep '<prn' > /tmp/pronouns.exp [22:57:54] <spectie> $ wc -l /tmp/pronouns.exp [22:57:54] <spectie> 57444 /tmp/pronouns.exp [22:59:28] <selimcan> hmm, too much I guess ? :) [22:59:33] <spectie> no [22:59:35] <spectie> certainly not [23:01:11] <spectie> cat /tmp/pronouns.exp | cut -f2 -d':' | sed 's/^/^/g' | sed 's/$/$/g' | apertium-pretransfer | lt-proc -b kaz-tat.autobil.bin | apertium-transfer -b apertium-kaz-tat.kaz-tat.t1x kaz-tat.t1x.bin | apertium-transfer -n apertium-kaz-tat.kaz-tat.t2x kaz-tat.t2x.bin > /tmp/pronouns.kaz-tat.exp [23:07:26] <selimcan> So this is basically how I should testvoc certain POS's? [23:07:51] <spectie> $ cat /tmp/pronouns.kaz-tat.exp | hfst-proc -d kaz-tat.autogen.hfst > /tmp/pronouns.tat.exp [23:07:51] <spectie> Violació de segment [23:07:57] <spectie> [23:07:57] <spectie> $ wc -l /tmp/pronouns.* [23:07:58] <spectie> 57444 /tmp/pronouns.exp [23:07:58] <spectie> 57444 /tmp/pronouns.kaz-tat.exp [23:07:58] <spectie> 12461 /tmp/pronouns.tat.exp [23:08:13] <spectie> [23:08:14] <spectie> ~#без\<prn\>\<pers\>\<p1\>\<pl\>\<px\>\<ins\> [23:08:39] <spectie> [23:08:39] <spectie> $ head -12461 /tmp/pronouns.kaz-tat.exp | tail -2 [23:08:40] <spectie> ^без<prn><pers><p1><pl><sim>+бит<mod_ass>$ [23:08:40] <spectie> ~^без<prn><pers><p1><pl><px><ins>$ [23:08:40] <spectie> [23:09:13] <spectie> no <ins> tag in tatar lexc [23:09:20] <selimcan> yeah, instrumental case, commiting in a min [23:09:44] <spectie> but you see how to do it [23:10:52] <selimcan> no, I'll do it tomorrow, have to get some sleep now [23:11:04] <spectie> ok [23:11:07] <spectie> i mean [23:11:11] <spectie> you can see the principle [23:12:30] <selimcan> Yeah, I see now how I can testvoc stuff, thanx
