diff options
| author | eherbst <eherbst@1f5c12ca-751b-0410-a591-d2e778427230> | 2006-08-16 18:51:04 +0400 |
|---|---|---|
| committer | eherbst <eherbst@1f5c12ca-751b-0410-a591-d2e778427230> | 2006-08-16 18:51:04 +0400 |
| commit | 674c609fcd4200a6dfb9458d082dee227b48aa69 (patch) | |
| tree | c64fa898c565c844b4dc7604f73fb0c45908f507 /scripts | |
| parent | c34aca3053a675a56334db1a476ff3f67d9777fa (diff) | |
adding show-phrases-used
git-svn-id: https://mosesdecoder.svn.sourceforge.net/svnroot/mosesdecoder/trunk@768 1f5c12ca-751b-0410-a591-d2e778427230
Diffstat (limited to 'scripts')
| -rw-r--r-- | scripts/analysis/README | 6 |
1 files changed, 5 insertions, 1 deletions
diff --git a/scripts/analysis/README b/scripts/analysis/README index 177550e70..8eeab22f1 100644 --- a/scripts/analysis/README +++ b/scripts/analysis/README @@ -1,5 +1,9 @@ Put any scripts useful for human analysis of MT output here. sentence-by-sentence.pl [EVH]: show comparison of sentences in reference translation(s)/system output(s)/(truth) in colorful format --- shows all sentences given, with non-matching words in the system output marked, BLEU scores given by sentence, and matching n-grams shown in a table +-- show all sentences given, with non-matching words in the system output marked, BLEU scores given by sentence, and matching n-grams shown in a table -- requires all input files be utf8-encoded (you can convert a file with `cat FILE | perl -n -e 'binmode(STDOUT, ":utf8"); print;' > FILE.utf8`) + +show-phrases-used.pl [EVH]: draw colorful diagram of which source phrases map to which target phrases +-- show average length of source phrases used for each sentence and overall +-- command-line options -r for reference and -s for source; lone filenames are taken to be system outputs |