From dfd3b65e37ea626fe92868f39ac019635438defa Mon Sep 17 00:00:00 2001 From: MORIOKA Tomohiko <tomo.git@chise.org> Date: Sun, 24 Aug 2014 20:08:06 +0900 Subject: [PATCH] - Generate corpus.non-KT.mc, corpus.mixed.mc and corpus.all.mc. (PATH): Use /usr/local/Cellar/mecab/0.996/libexec/mecab instead of /usr/local/Cellar/mecab/0.994/libexec/mecab. (CORPUS_LIST): Add "KT0", "KT2", "non-KT", "mixed" and "all". --- seed/system-eval.sh | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) diff --git a/seed/system-eval.sh b/seed/system-eval.sh index 08d06b1c2..2e77c4473 100755 --- a/seed/system-eval.sh +++ b/seed/system-eval.sh @@ -1,20 +1,28 @@ #!/bin/sh -PATH="/usr/local/bin:/usr/local/libexec/mecab:/bin:/usr/lib/mecab:/usr/local/Cellar/mecab/0.994/libexec/mecab" +PATH="/usr/local/bin:/usr/local/libexec/mecab:/bin:/usr/lib/mecab:/usr/local/Cellar/mecab/0.996/libexec/mecab" export PATH export LANG=ja_JP.UTF-8 +cat corpus.misc.mc corpus.kanjikai.mc corpus.ryomou.mc \ + > corpus.non-KT.mc + +cat corpus.KT2.mc corpus.misc.mc corpus.kanjikai.mc corpus.ryomou.mc \ + > corpus.mixed.mc + +cat corpus.KT0.mc corpus.misc.mc corpus.kanjikai.mc corpus.ryomou.mc \ + > corpus.all.mc + #CORPUS_LIST="misc kanjikai ryomou jts-wa jts-JP" -CORPUS_LIST="misc kanjikai ryomou" +CORPUS_LIST="KT0 KT2 misc kanjikai ryomou non-KT mixed all" for i in $CORPUS_LIST do mecab-test-gen < "corpus.$i.mc" > "test.$i" done -#cat corpus \ -#cat corpus.*.txt \ +#cat corpus.*.mc \ #| grep -v EOS | sed 's/ /,0,0,0,/' | sort | uniq > misc.corpus.csv -- GitLab