Skip to content

Commit

Permalink
latest models added
Browse files Browse the repository at this point in the history
  • Loading branch information
jorgtied committed Sep 19, 2022
1 parent 875e48b commit ea1ca55
Show file tree
Hide file tree
Showing 2,781 changed files with 78,593 additions and 59,729 deletions.
29 changes: 24 additions & 5 deletions models/Makefile
Original file line number Diff line number Diff line change
Expand Up @@ -20,10 +20,29 @@ all: ${SCOREFILES} ${BLEUSCOREFILES} ${CHRFSCOREFILES} ${COMETSCOREFILES}
rm -f $@

%-scores.fix: %-scores.txt
sed 's/\(news.*[0-9][0-9][0-9][0-9]\)\-[a-z][a-z][a-z][a-z] /\1 /' $< |\
if [ -e $< ]; then \
sed 's/\(news.*[0-9][0-9][0-9][0-9]\)\-[a-z][a-z][a-z][a-z] /\1 /' $< |\
sed -e 's/\(news.*2021\)\.[a-z][a-z]\-[a-z][a-z] /\1 /' |\
rev | uniq -f1 | rev > $@; \
if [ `diff $< $@ | wc -l` -gt 0 ]; then \
mv $@ $<; \
fi; \
rm -f $@; \
fi

%.bleu-scores.txt: %.scores.txt
cut -f1,2,4 $< | \
sed 's/\(news.*[0-9][0-9][0-9][0-9]\)\-[a-z][a-z][a-z][a-z] /\1 /' |\
sed -e 's/\(news.*2021\)\.[a-z][a-z]\-[a-z][a-z] /\1 /' |\
rev | uniq -f1 | rev > $@
if [ `diff $< $@ | wc -l` -gt 0 ]; then \
mv $@ $<; \
fi
rm -f $@


%.chrf-scores.txt: %.scores.txt
cut -f1,2,3 $< |\
sed 's/\(news.*[0-9][0-9][0-9][0-9]\)\-[a-z][a-z][a-z][a-z] /\1 /' |\
sed -e 's/\(news.*2021\)\.[a-z][a-z]\-[a-z][a-z] /\1 /' |\
rev | uniq -f1 | rev > $@

## cannot produce comet scores here
%.comet-scores.txt:
@echo "don't know how to produce comet scores"
86 changes: 50 additions & 36 deletions models/aav-aav/opus-2020-10-04.bleu-scores.txt
Original file line number Diff line number Diff line change
@@ -1,50 +1,64 @@
eng-hoc_Latn tatoeba-test-v2020-07-28 0.0
eng-hoc_Latn tatoeba-test-v2021-03-30 0.0
eng-hoc_Latn tatoeba-test-v2021-08-07 0.0
eng-hoc tatoeba-test-v2020-07-28 0.0
eng-hoc tatoeba-test-v2021-03-30 0.0
eng-hoc tatoeba-test-v2021-08-07 0.0
eng-kha tatoeba-test-v2020-07-28 0.1
eng-kha tatoeba-test-v2021-03-30 0.1
eng-kha tatoeba-test-v2021-08-07 0.1
eng-khm flores101-dev 0.5
eng-vie flores101-dev 22.2
khm-eng flores101-dev 9.9
khm-vie flores101-dev 9.6
eng-khm flores101-devtest 0.6
eng-vie flores101-devtest 22.2
khm-eng flores101-devtest 8.6
khm-vie flores101-devtest 8.8
vie-eng flores101-devtest 16.2
vie-khm flores101-devtest 0.2
vie-eng flores101-dev 17.3
vie-khm flores101-dev 0.2
eng-khm flores200-dev 0.5
eng-khm flores200-devtest 0.6
eng-khm newstest2020 0.4
khm-eng newstest2020 5.4
eng-hoc tatoeba-test-v2020-07-28 0.0
eng-hoc_Latn tatoeba-test-v2020-07-28 0.0
eng-kha tatoeba-test-v2020-07-28 0.1
eng-khm tatoeba-test-v2020-07-28 0.2
eng-vie tatoeba-test-v2020-07-28 30.2
hoc-eng tatoeba-test-v2020-07-28 0.2
hoc_Latn-eng tatoeba-test-v2020-07-28 0.2
kha-eng tatoeba-test-v2020-07-28 0.8
khm-eng tatoeba-test-v2020-07-28 11.7
vie-eng tatoeba-test-v2020-07-28 31.6
eng-hoc tatoeba-test-v2021-03-30 0.0
eng-hoc_Latn tatoeba-test-v2021-03-30 0.0
eng-kha tatoeba-test-v2021-03-30 0.1
eng-khm tatoeba-test-v2021-03-30 0.2
eng-vie tatoeba-test-v2021-03-30 30.2
hoc-eng tatoeba-test-v2021-03-30 0.2
hoc_Latn-eng tatoeba-test-v2021-03-30 0.2
kha-eng tatoeba-test-v2021-03-30 0.8
khm-eng tatoeba-test-v2021-03-30 11.7
vie-eng tatoeba-test-v2021-03-30 31.6
eng-hoc tatoeba-test-v2021-08-07 0.0
eng-hoc_Latn tatoeba-test-v2021-08-07 0.0
eng-kha tatoeba-test-v2021-08-07 0.1
eng-khm tatoeba-test-v2021-08-07 0.2
eng-khm tico19-test 1.0
eng-khm wikipedia.dev.km-en 1.0
eng-khm wikipedia.devtest.km-en 0.0
eng-vie flores101-dev 22.2
eng-vie flores101-devtest 22.2
eng-vie flores200-dev 22.1
eng-vie flores200-devtest 21.8
eng-vie tatoeba-test-v2020-07-28 30.2
eng-vie tatoeba-test-v2021-03-30 30.2
eng-vie tatoeba-test-v2021-08-07 30.2
hoc-eng tatoeba-test-v2020-07-28 0.2
hoc-eng tatoeba-test-v2021-03-30 0.2
hoc-eng tatoeba-test-v2021-08-07 0.2
hoc_Latn-eng tatoeba-test-v2020-07-28 0.2
hoc_Latn-eng tatoeba-test-v2021-03-30 0.2
hoc_Latn-eng tatoeba-test-v2021-08-07 0.2
kha-eng tatoeba-test-v2020-07-28 0.8
kha-eng tatoeba-test-v2021-03-30 0.8
kha-eng tatoeba-test-v2021-08-07 0.8
khm-eng flores101-dev 9.9
khm-eng flores101-devtest 8.6
khm-eng flores200-dev 10.0
khm-eng flores200-devtest 8.6
khm-eng newstest2020 5.3
khm-eng tatoeba-test-v2020-07-28 11.7
khm-eng tatoeba-test-v2021-03-30 11.7
khm-eng tatoeba-test-v2021-08-07 11.6
vie-eng tatoeba-test-v2021-08-07 31.6
eng-khm tico19-test 1.0
khm-eng tico19-test 9.4
eng-khm wikipedia.dev.km-en 1.0
khm-eng wikipedia.dev 6.1
khm-eng wikipedia.dev.km-en 6.2
eng-khm wikipedia.devtest.km-en 0.0
khm-eng wikipedia.devtest 5.0
khm-eng wikipedia.devtest.km-en 5.2
khm-vie flores101-dev 9.6
khm-vie flores101-devtest 8.8
khm-vie flores200-dev 9.5
khm-vie flores200-devtest 8.6
vie-eng flores101-dev 17.3
vie-eng flores101-devtest 16.2
vie-eng flores200-dev 17.1
vie-eng flores200-devtest 15.9
vie-eng tatoeba-test-v2020-07-28 31.6
vie-eng tatoeba-test-v2021-03-30 31.6
vie-eng tatoeba-test-v2021-08-07 31.6
vie-khm flores101-dev 0.2
vie-khm flores101-devtest 0.2
vie-khm flores200-dev 0.2
vie-khm flores200-devtest 0.3
88 changes: 51 additions & 37 deletions models/aav-aav/opus-2020-10-04.chrf-scores.txt
Original file line number Diff line number Diff line change
@@ -1,50 +1,64 @@
eng-khm flores101-dev 0.194
eng-vie flores101-dev 0.424
khm-eng flores101-dev 0.368
khm-vie flores101-dev 0.310
eng-khm flores101-devtest 0.191
eng-vie flores101-devtest 0.419
khm-eng flores101-devtest 0.347
khm-vie flores101-devtest 0.296
vie-eng flores101-devtest 0.426
vie-khm flores101-devtest 0.115
vie-eng flores101-dev 0.436
vie-khm flores101-dev 0.113
eng-khm newstest2020 0.183
khm-eng newstest2020 0.272
eng-hoc tatoeba-test-v2020-07-28 0.009
eng-hoc_Latn tatoeba-test-v2020-07-28 0.009
eng-kha tatoeba-test-v2020-07-28 0.011
eng-khm tatoeba-test-v2020-07-28 0.245
eng-vie tatoeba-test-v2020-07-28 0.480
hoc-eng tatoeba-test-v2020-07-28 0.099
hoc_Latn-eng tatoeba-test-v2020-07-28 0.099
kha-eng tatoeba-test-v2020-07-28 0.119
khm-eng tatoeba-test-v2020-07-28 0.298
vie-eng tatoeba-test-v2020-07-28 0.511
eng-hoc tatoeba-test-v2021-03-30 0.009
eng-hoc_Latn tatoeba-test-v2021-03-30 0.009
eng-kha tatoeba-test-v2021-03-30 0.011
eng-khm tatoeba-test-v2021-03-30 0.245
eng-vie tatoeba-test-v2021-03-30 0.480
hoc-eng tatoeba-test-v2021-03-30 0.099
hoc_Latn-eng tatoeba-test-v2021-03-30 0.099
kha-eng tatoeba-test-v2021-03-30 0.119
khm-eng tatoeba-test-v2021-03-30 0.298
vie-eng tatoeba-test-v2021-03-30 0.511
eng-hoc tatoeba-test-v2021-08-07 0.009
eng-hoc_Latn tatoeba-test-v2021-08-07 0.009
eng-hoc tatoeba-test-v2020-07-28 0.009
eng-hoc tatoeba-test-v2021-03-30 0.009
eng-hoc tatoeba-test-v2021-08-07 0.009
eng-kha tatoeba-test-v2020-07-28 0.011
eng-kha tatoeba-test-v2021-03-30 0.011
eng-kha tatoeba-test-v2021-08-07 0.011
eng-khm flores101-dev 0.194
eng-khm flores101-devtest 0.191
eng-khm flores200-dev 0.19390
eng-khm flores200-devtest 0.19321
eng-khm newstest2020 0.18618
eng-khm tatoeba-test-v2020-07-28 0.245
eng-khm tatoeba-test-v2021-03-30 0.245
eng-khm tatoeba-test-v2021-08-07 0.242
eng-khm tico19-test 0.287
eng-khm wikipedia.dev.km-en 0.163
eng-khm wikipedia.devtest.km-en 0.145
eng-vie flores101-dev 0.424
eng-vie flores101-devtest 0.419
eng-vie flores200-dev 0.42304
eng-vie flores200-devtest 0.41636
eng-vie tatoeba-test-v2020-07-28 0.480
eng-vie tatoeba-test-v2021-03-30 0.480
eng-vie tatoeba-test-v2021-08-07 0.480
hoc-eng tatoeba-test-v2020-07-28 0.099
hoc-eng tatoeba-test-v2021-03-30 0.099
hoc-eng tatoeba-test-v2021-08-07 0.099
hoc_Latn-eng tatoeba-test-v2020-07-28 0.099
hoc_Latn-eng tatoeba-test-v2021-03-30 0.099
hoc_Latn-eng tatoeba-test-v2021-08-07 0.099
kha-eng tatoeba-test-v2020-07-28 0.119
kha-eng tatoeba-test-v2021-03-30 0.119
kha-eng tatoeba-test-v2021-08-07 0.119
khm-eng flores101-dev 0.368
khm-eng flores101-devtest 0.347
khm-eng flores200-dev 0.36863
khm-eng flores200-devtest 0.34642
khm-eng newstest2020 0.27289
khm-eng tatoeba-test-v2020-07-28 0.298
khm-eng tatoeba-test-v2021-03-30 0.298
khm-eng tatoeba-test-v2021-08-07 0.296
vie-eng tatoeba-test-v2021-08-07 0.511
eng-khm tico19-test 0.287
khm-eng tico19-test 0.340
eng-khm wikipedia.dev.km-en 0.163
khm-eng wikipedia.dev 0.27117
khm-eng wikipedia.dev.km-en 0.270
eng-khm wikipedia.devtest.km-en 0.145
khm-eng wikipedia.devtest 0.24758
khm-eng wikipedia.devtest.km-en 0.249
khm-vie flores101-dev 0.310
khm-vie flores101-devtest 0.296
khm-vie flores200-dev 0.30634
khm-vie flores200-devtest 0.29532
vie-eng flores101-dev 0.436
vie-eng flores101-devtest 0.426
vie-eng flores200-dev 0.43626
vie-eng flores200-devtest 0.42508
vie-eng tatoeba-test-v2020-07-28 0.511
vie-eng tatoeba-test-v2021-03-30 0.511
vie-eng tatoeba-test-v2021-08-07 0.511
vie-khm flores101-dev 0.113
vie-khm flores101-devtest 0.115
vie-khm flores200-dev 0.11254
vie-khm flores200-devtest 0.11154
84 changes: 52 additions & 32 deletions models/aav-aav/opus4m+btTCv20210807-2021-09-30.bleu-scores.txt
Original file line number Diff line number Diff line change
@@ -1,44 +1,64 @@
eng-khm flores101-dev 5.1
eng-vie flores101-dev 35.9
khm-eng flores101-dev 21.5
khm-vie flores101-dev 22.2
eng-khm flores101-devtest 4.9
eng-vie flores101-devtest 36.0
khm-eng flores101-devtest 19.9
khm-vie flores101-devtest 22.0
vie-eng flores101-devtest 27.5
vie-khm flores101-devtest 4.2
vie-eng flores101-dev 29.3
vie-khm flores101-dev 4.0
eng-khm newstest2020 0.4
khm-eng newstest2020 13.0
eng-hoc tatoeba-test-v2020-07-28 0.3
eng-kha tatoeba-test-v2020-07-28 0.7
eng-khm tatoeba-test-v2020-07-28 0.6
eng-vie tatoeba-test-v2020-07-28 34.0
hoc-eng tatoeba-test-v2020-07-28 0.3
kha-eng tatoeba-test-v2020-07-28 0.8
khm-eng tatoeba-test-v2020-07-28 17.4
vie-eng tatoeba-test-v2020-07-28 36.9
eng-hoc tatoeba-test-v2021-03-30 0.3
eng-kha tatoeba-test-v2021-03-30 0.7
eng-khm tatoeba-test-v2021-03-30 0.6
eng-vie tatoeba-test-v2021-03-30 34.0
hoc-eng tatoeba-test-v2021-03-30 0.3
kha-eng tatoeba-test-v2021-03-30 0.8
khm-eng tatoeba-test-v2021-03-30 17.4
vie-eng tatoeba-test-v2021-03-30 36.9
eng-hoc tatoeba-test-v2021-08-07 0.3
eng-kha tatoeba-test-v2020-07-28 0.7
eng-kha tatoeba-test-v2021-03-30 0.7
eng-kha tatoeba-test-v2021-08-07 0.7
eng-khm flores101-dev 5.1
eng-khm flores101-devtest 4.9
eng-khm flores200-dev 5.0
eng-khm flores200-devtest 4.9
eng-khm newstest2020 0.5
eng-khm tatoeba-test-v2020-07-28 0.6
eng-khm tatoeba-test-v2021-03-30 0.6
eng-khm tatoeba-test-v2021-08-07 0.6
eng-khm tico19-test 14.6
eng-khm wikipedia.dev.km-en 1.7
eng-khm wikipedia.devtest.km-en 0.3
eng-vie flores101-dev 35.9
eng-vie flores101-devtest 36.0
eng-vie flores200-dev 36.0
eng-vie flores200-devtest 36.0
eng-vie tatoeba-test-v2020-07-28 34.0
eng-vie tatoeba-test-v2021-03-30 34.0
eng-vie tatoeba-test-v2021-08-07 34.1
hoc-eng tatoeba-test-v2020-07-28 0.3
hoc-eng tatoeba-test-v2021-03-30 0.3
hoc-eng tatoeba-test-v2021-08-07 0.3
kha-eng tatoeba-test-v2020-07-28 0.8
kha-eng tatoeba-test-v2021-03-30 0.8
kha-eng tatoeba-test-v2021-08-07 0.8
khm-eng flores101-dev 21.5
khm-eng flores101-devtest 19.9
khm-eng flores200-dev 21.6
khm-eng flores200-devtest 20.1
khm-eng newstest2020 13.4
khm-eng tatoeba-test-v2020-07-28 17.4
khm-eng tatoeba-test-v2021-03-30 17.4
khm-eng tatoeba-test-v2021-08-07 17.1
vie-eng tatoeba-test-v2021-08-07 36.9
eng-khm tico19-test 14.6
khm-eng tico19-test 30.7
eng-khm wikipedia.dev.km-en 1.7
khm-eng wikipedia.dev 14.0
khm-eng wikipedia.dev.km-en 13.1
eng-khm wikipedia.devtest.km-en 0.3
khm-eng wikipedia.devtest 11.5
khm-eng wikipedia.devtest.km-en 10.5
khm-vie flores101-dev 22.2
khm-vie flores101-devtest 22.0
khm-vie flores200-dev 22.0
khm-vie flores200-devtest 21.8
sat-eng flores200-dev 0.2
sat-eng flores200-devtest 0.2
sat-khm flores200-dev 0.0
sat-khm flores200-devtest 0.0
sat-vie flores200-dev 0.1
sat-vie flores200-devtest 0.2
vie-eng flores101-dev 29.3
vie-eng flores101-devtest 27.5
vie-eng flores200-dev 29.0
vie-eng flores200-devtest 27.3
vie-eng tatoeba-test-v2020-07-28 36.9
vie-eng tatoeba-test-v2021-03-30 36.9
vie-eng tatoeba-test-v2021-08-07 36.9
vie-khm flores101-dev 4.0
vie-khm flores101-devtest 4.2
vie-khm flores200-dev 3.8
vie-khm flores200-devtest 4.0
Loading

0 comments on commit ea1ca55

Please sign in to comment.