Skip to content

Latest commit

 

History

History

eng-mul

opus-2020-07-14.zip

  • dataset: opus
  • model: transformer
  • source language(s): eng
  • target language(s): abk acm ady afb afh_Latn afr akl_Latn aln amh ang_Latn apc ara arg arq ary arz asm ast avk_Latn awa aze_Latn bak bam_Latn bel bel_Latn ben bho bod bos_Latn bre brx brx_Latn bul bul_Latn cat ceb ces cha che chr chv cjy_Hans cjy_Hant cmn cmn_Hans cmn_Hant cor cos crh crh_Latn csb_Latn cym dan deu dsb dtp dws_Latn egl ell enm_Latn epo est eus ewe ext fao fij fin fkv_Latn fra frm_Latn frr fry fuc fuv gan gcf_Latn gil gla gle glg glv gom gos got_Goth grc_Grek grn gsw guj hat hau_Latn haw heb hif_Latn hil hin hnj_Latn hoc hoc_Latn hrv hsb hun hye iba ibo ido ido_Latn ike_Latn ile_Latn ilo ina_Latn ind isl ita izh jav jav_Java jbo jbo_Cyrl jbo_Latn jdt_Cyrl jpn kab kal kan kat kaz_Cyrl kaz_Latn kek_Latn kha khm khm_Latn kin kir_Cyrl kjh kpv krl ksh kum kur_Arab kur_Latn lad lad_Latn lao lat_Latn lav ldn_Latn lfn_Cyrl lfn_Latn lij lin lit liv_Latn lkt lld_Latn lmo ltg ltz lug lzh lzh_Hans mad mah mai mal mar max_Latn mdf mfe mhr mic min mkd mlg mlt mnw moh mon mri mwl mww mya myv nan nau nav nds niu nld nno nob nob_Hebr nog non_Latn nov_Latn npi nya oci ori orv_Cyrl oss ota_Arab ota_Latn pag pan_Guru pap pau pdc pes pes_Latn pes_Thaa pms pnb pol por ppl_Latn prg_Latn pus quc qya qya_Latn rap rif_Latn roh rom ron rue run rus sag sah san_Deva scn sco sgs shs_Latn shy_Latn sin sjn_Latn slv sma sme smo sna snd_Arab som spa sqi srp_Cyrl srp_Latn stq sun swe swg swh tah tam tat tat_Arab tat_Latn tel tet tgk_Cyrl tha tir tlh_Latn tly_Latn tmw_Latn toi_Latn ton tpw_Latn tso tuk tuk_Latn tur tvl tyv tzl tzl_Latn udm uig_Arab uig_Cyrl ukr umb urd uzb_Cyrl uzb_Latn vec vie vie_Hani vol_Latn vro war wln wol wuu xal xho yid yor yue yue_Hans yue_Hant zho zho_Hans zho_Hant zlm_Latn zsm_Latn zul zza
  • model: transformer
  • pre-processing: normalization + SentencePiece (spm32k,spm32k)
  • a sentence initial language token is required in the form of >>id<< (id = valid target language ID)
  • download: opus-2020-07-14.zip
  • test set translations: opus-2020-07-14.test.txt
  • test set scores: opus-2020-07-14.eval.txt

Benchmarks

testset BLEU chr-F
Tatoeba-test.eng-abk.eng.abk 1.3 0.230
Tatoeba-test.eng-ady.eng.ady 1.8 0.008
Tatoeba-test.eng-afh.eng.afh 1.1 0.066
Tatoeba-test.eng-afr.eng.afr 45.2 0.642
Tatoeba-test.eng-akl.eng.akl 1.5 0.076
Tatoeba-test.eng-amh.eng.amh 7.7 0.422
Tatoeba-test.eng-ang.eng.ang 5.8 0.139
Tatoeba-test.eng-ara.eng.ara 4.9 0.293
Tatoeba-test.eng-arg.eng.arg 1.4 0.116
Tatoeba-test.eng-asm.eng.asm 2.1 0.216
Tatoeba-test.eng-ast.eng.ast 12.0 0.367
Tatoeba-test.eng-avk.eng.avk 0.2 0.099
Tatoeba-test.eng-awa.eng.awa 0.2 0.002
Tatoeba-test.eng-aze.eng.aze 15.3 0.450
Tatoeba-test.eng-bak.eng.bak 7.2 0.323
Tatoeba-test.eng-bam.eng.bam 0.0 0.037
Tatoeba-test.eng-bel.eng.bel 11.4 0.361
Tatoeba-test.eng-ben.eng.ben 8.1 0.349
Tatoeba-test.eng-bho.eng.bho 0.2 0.003
Tatoeba-test.eng-bod.eng.bod 0.0 0.152
Tatoeba-test.eng-bre.eng.bre 5.7 0.240
Tatoeba-test.eng-brx.eng.brx 0.8 0.050
Tatoeba-test.eng-bul.eng.bul 29.0 0.512
Tatoeba-test.eng-cat.eng.cat 33.0 0.560
Tatoeba-test.eng-ceb.eng.ceb 9.1 0.386
Tatoeba-test.eng-ces.eng.ces 19.5 0.429
Tatoeba-test.eng-cha.eng.cha 4.0 0.215
Tatoeba-test.eng-che.eng.che 0.7 0.013
Tatoeba-test.eng-chm.eng.chm 2.5 0.218
Tatoeba-test.eng-chr.eng.chr 1.7 0.145
Tatoeba-test.eng-chv.eng.chv 6.0 0.330
Tatoeba-test.eng-cor.eng.cor 0.6 0.083
Tatoeba-test.eng-cos.eng.cos 4.3 0.165
Tatoeba-test.eng-crh.eng.crh 9.3 0.326
Tatoeba-test.eng-csb.eng.csb 6.3 0.212
Tatoeba-test.eng-cym.eng.cym 14.3 0.378
Tatoeba-test.eng-dan.eng.dan 38.0 0.591
Tatoeba-test.eng-deu.eng.deu 21.5 0.472
Tatoeba-test.eng-dsb.eng.dsb 0.7 0.108
Tatoeba-test.eng-dtp.eng.dtp 0.3 0.086
Tatoeba-test.eng-dws.eng.dws 0.7 0.081
Tatoeba-test.eng-egl.eng.egl 0.5 0.082
Tatoeba-test.eng-ell.eng.ell 24.1 0.484
Tatoeba-test.eng-enm.eng.enm 4.8 0.307
Tatoeba-test.eng-epo.eng.epo 29.9 0.540
Tatoeba-test.eng-est.eng.est 26.0 0.500
Tatoeba-test.eng-eus.eng.eus 14.7 0.430
Tatoeba-test.eng-ewe.eng.ewe 6.4 0.244
Tatoeba-test.eng-ext.eng.ext 4.4 0.225
Tatoeba-test.eng-fao.eng.fao 5.7 0.271
Tatoeba-test.eng-fas.eng.fas 6.2 0.252
Tatoeba-test.eng-fij.eng.fij 32.3 0.356
Tatoeba-test.eng-fin.eng.fin 15.6 0.420
Tatoeba-test.eng-fkv.eng.fkv 0.9 0.144
Tatoeba-test.eng-fra.eng.fra 28.0 0.520
Tatoeba-test.eng-frm.eng.frm 0.4 0.122
Tatoeba-test.eng-frr.eng.frr 3.6 0.032
Tatoeba-test.eng-fry.eng.fry 14.5 0.375
Tatoeba-test.eng-ful.eng.ful 0.4 0.073
Tatoeba-test.eng-gcf.eng.gcf 0.5 0.095
Tatoeba-test.eng-gil.eng.gil 37.0 0.621
Tatoeba-test.eng-gla.eng.gla 9.7 0.319
Tatoeba-test.eng-gle.eng.gle 21.7 0.440
Tatoeba-test.eng-glg.eng.glg 30.1 0.544
Tatoeba-test.eng-glv.eng.glv 22.0 0.411
Tatoeba-test.eng-gos.eng.gos 0.5 0.135
Tatoeba-test.eng-got.eng.got 0.2 0.010
Tatoeba-test.eng-grc.eng.grc 0.0 0.005
Tatoeba-test.eng-grn.eng.grn 1.5 0.131
Tatoeba-test.eng-gsw.eng.gsw 0.6 0.108
Tatoeba-test.eng-guj.eng.guj 13.9 0.330
Tatoeba-test.eng-hat.eng.hat 29.3 0.509
Tatoeba-test.eng-hau.eng.hau 3.0 0.350
Tatoeba-test.eng-haw.eng.haw 0.4 0.079
Tatoeba-test.eng-heb.eng.heb 17.7 0.398
Tatoeba-test.eng-hif.eng.hif 2.8 0.283
Tatoeba-test.eng-hil.eng.hil 12.5 0.448
Tatoeba-test.eng-hin.eng.hin 10.6 0.363
Tatoeba-test.eng-hmn.eng.hmn 0.2 0.071
Tatoeba-test.eng-hoc.eng.hoc 0.2 0.078
Tatoeba-test.eng-hsb.eng.hsb 4.5 0.277
Tatoeba-test.eng-hun.eng.hun 13.6 0.383
Tatoeba-test.eng-hye.eng.hye 12.9 0.337
Tatoeba-test.eng-iba.eng.iba 13.0 0.411
Tatoeba-test.eng-ibo.eng.ibo 2.2 0.217
Tatoeba-test.eng-ido.eng.ido 9.8 0.377
Tatoeba-test.eng-iku.eng.iku 2.5 0.087
Tatoeba-test.eng-ile.eng.ile 4.3 0.280
Tatoeba-test.eng-ilo.eng.ilo 24.6 0.497
Tatoeba-test.eng-ina.eng.ina 19.5 0.479
Tatoeba-test.eng-isl.eng.isl 10.6 0.345
Tatoeba-test.eng-ita.eng.ita 30.4 0.559
Tatoeba-test.eng-izh.eng.izh 1.5 0.053
Tatoeba-test.eng-jav.eng.jav 7.2 0.316
Tatoeba-test.eng-jbo.eng.jbo 0.2 0.121
Tatoeba-test.eng-jdt.eng.jdt 0.6 0.000
Tatoeba-test.eng-jpn.eng.jpn 0.0 0.001
Tatoeba-test.eng-kab.eng.kab 4.6 0.247
Tatoeba-test.eng-kal.eng.kal 10.0 0.339
Tatoeba-test.eng-kan.eng.kan 3.9 0.332
Tatoeba-test.eng-kat.eng.kat 1.2 0.120
Tatoeba-test.eng-kaz.eng.kaz 8.7 0.308
Tatoeba-test.eng-kek.eng.kek 3.3 0.191
Tatoeba-test.eng-kha.eng.kha 0.4 0.097
Tatoeba-test.eng-khm.eng.khm 0.3 0.227
Tatoeba-test.eng-kin.eng.kin 7.0 0.372
Tatoeba-test.eng-kir.eng.kir 15.4 0.364
Tatoeba-test.eng-kjh.eng.kjh 0.4 0.007
Tatoeba-test.eng-kok.eng.kok 5.5 0.005
Tatoeba-test.eng-kom.eng.kom 0.6 0.006
Tatoeba-test.eng-krl.eng.krl 2.2 0.099
Tatoeba-test.eng-ksh.eng.ksh 0.9 0.120
Tatoeba-test.eng-kum.eng.kum 1.6 0.008
Tatoeba-test.eng-kur.eng.kur 4.3 0.164
Tatoeba-test.eng-lad.eng.lad 0.5 0.150
Tatoeba-test.eng-lah.eng.lah 0.2 0.001
Tatoeba-test.eng-lao.eng.lao 0.7 0.409
Tatoeba-test.eng-lat.eng.lat 7.3 0.352
Tatoeba-test.eng-lav.eng.lav 15.0 0.428
Tatoeba-test.eng-ldn.eng.ldn 0.5 0.073
Tatoeba-test.eng-lfn.eng.lfn 0.4 0.114
Tatoeba-test.eng-lij.eng.lij 0.7 0.121
Tatoeba-test.eng-lin.eng.lin 1.0 0.251
Tatoeba-test.eng-lit.eng.lit 12.4 0.430
Tatoeba-test.eng-liv.eng.liv 1.0 0.042
Tatoeba-test.eng-lkt.eng.lkt 0.5 0.059
Tatoeba-test.eng-lld.eng.lld 0.2 0.104
Tatoeba-test.eng-lmo.eng.lmo 2.7 0.150
Tatoeba-test.eng-ltz.eng.ltz 19.8 0.413
Tatoeba-test.eng-lug.eng.lug 24.0 0.600
Tatoeba-test.eng-mad.eng.mad 0.7 0.058
Tatoeba-test.eng-mah.eng.mah 8.9 0.323
Tatoeba-test.eng-mai.eng.mai 10.7 0.396
Tatoeba-test.eng-mal.eng.mal 6.9 0.400
Tatoeba-test.eng-mar.eng.mar 10.7 0.358
Tatoeba-test.eng-mdf.eng.mdf 1.7 0.007
Tatoeba-test.eng-mfe.eng.mfe 83.6 0.911
Tatoeba-test.eng-mic.eng.mic 1.5 0.096
Tatoeba-test.eng-mkd.eng.mkd 28.3 0.512
Tatoeba-test.eng-mlg.eng.mlg 26.7 0.416
Tatoeba-test.eng-mlt.eng.mlt 7.9 0.371
Tatoeba-test.eng-mnw.eng.mnw 0.7 0.003
Tatoeba-test.eng-moh.eng.moh 0.9 0.069
Tatoeba-test.eng-mon.eng.mon 6.8 0.288
Tatoeba-test.eng-mri.eng.mri 8.0 0.307
Tatoeba-test.eng-msa.eng.msa 22.7 0.476
Tatoeba-test.eng.multi 18.5 0.415
Tatoeba-test.eng-mwl.eng.mwl 2.5 0.132
Tatoeba-test.eng-mya.eng.mya 0.2 0.107
Tatoeba-test.eng-myv.eng.myv 0.5 0.013
Tatoeba-test.eng-nau.eng.nau 0.5 0.110
Tatoeba-test.eng-nav.eng.nav 2.7 0.145
Tatoeba-test.eng-nds.eng.nds 9.8 0.345
Tatoeba-test.eng-nep.eng.nep 0.3 0.003
Tatoeba-test.eng-niu.eng.niu 33.5 0.464
Tatoeba-test.eng-nld.eng.nld 33.0 0.560
Tatoeba-test.eng-nog.eng.nog 0.2 0.009
Tatoeba-test.eng-non.eng.non 0.5 0.096
Tatoeba-test.eng-nov.eng.nov 1.5 0.209
Tatoeba-test.eng-nya.eng.nya 9.7 0.446
Tatoeba-test.eng-oci.eng.oci 4.7 0.266
Tatoeba-test.eng-ori.eng.ori 3.4 0.235
Tatoeba-test.eng-orv.eng.orv 0.2 0.007
Tatoeba-test.eng-oss.eng.oss 2.9 0.220
Tatoeba-test.eng-ota.eng.ota 0.2 0.055
Tatoeba-test.eng-pag.eng.pag 21.5 0.499
Tatoeba-test.eng-pan.eng.pan 4.6 0.281
Tatoeba-test.eng-pap.eng.pap 46.5 0.623
Tatoeba-test.eng-pau.eng.pau 1.0 0.099
Tatoeba-test.eng-pdc.eng.pdc 2.7 0.156
Tatoeba-test.eng-pms.eng.pms 0.6 0.136
Tatoeba-test.eng-pol.eng.pol 18.5 0.457
Tatoeba-test.eng-por.eng.por 28.7 0.546
Tatoeba-test.eng-ppl.eng.ppl 0.4 0.100
Tatoeba-test.eng-prg.eng.prg 0.3 0.090
Tatoeba-test.eng-pus.eng.pus 0.7 0.127
Tatoeba-test.eng-quc.eng.quc 2.7 0.214
Tatoeba-test.eng-qya.eng.qya 0.4 0.092
Tatoeba-test.eng-rap.eng.rap 2.2 0.054
Tatoeba-test.eng-rif.eng.rif 1.3 0.110
Tatoeba-test.eng-roh.eng.roh 1.9 0.219
Tatoeba-test.eng-rom.eng.rom 9.2 0.338
Tatoeba-test.eng-ron.eng.ron 23.4 0.495
Tatoeba-test.eng-rue.eng.rue 0.1 0.010
Tatoeba-test.eng-run.eng.run 8.3 0.399
Tatoeba-test.eng-rus.eng.rus 19.2 0.431
Tatoeba-test.eng-sag.eng.sag 4.7 0.160
Tatoeba-test.eng-sah.eng.sah 0.3 0.006
Tatoeba-test.eng-san.eng.san 2.1 0.130
Tatoeba-test.eng-scn.eng.scn 0.6 0.064
Tatoeba-test.eng-sco.eng.sco 30.3 0.526
Tatoeba-test.eng-sgs.eng.sgs 2.7 0.107
Tatoeba-test.eng-shs.eng.shs 1.0 0.058
Tatoeba-test.eng-shy.eng.shy 0.3 0.045
Tatoeba-test.eng-sin.eng.sin 3.8 0.296
Tatoeba-test.eng-sjn.eng.sjn 0.4 0.083
Tatoeba-test.eng-slv.eng.slv 10.6 0.297
Tatoeba-test.eng-sma.eng.sma 0.5 0.063
Tatoeba-test.eng-sme.eng.sme 12.2 0.259
Tatoeba-test.eng-smo.eng.smo 24.6 0.453
Tatoeba-test.eng-sna.eng.sna 19.3 0.492
Tatoeba-test.eng-snd.eng.snd 2.4 0.166
Tatoeba-test.eng-som.eng.som 16.0 0.192
Tatoeba-test.eng-spa.eng.spa 33.3 0.576
Tatoeba-test.eng-sqi.eng.sqi 27.9 0.538
Tatoeba-test.eng-stq.eng.stq 5.9 0.316
Tatoeba-test.eng-sun.eng.sun 38.0 0.441
Tatoeba-test.eng-swe.eng.swe 38.6 0.580
Tatoeba-test.eng-swg.eng.swg 0.6 0.127
Tatoeba-test.eng-tah.eng.tah 5.5 0.276
Tatoeba-test.eng-tam.eng.tam 5.1 0.371
Tatoeba-test.eng-tat.eng.tat 8.2 0.298
Tatoeba-test.eng-tel.eng.tel 4.2 0.306
Tatoeba-test.eng-tet.eng.tet 0.6 0.096
Tatoeba-test.eng-tgk.eng.tgk 9.1 0.311
Tatoeba-test.eng-tha.eng.tha 2.2 0.428
Tatoeba-test.eng-tir.eng.tir 1.5 0.186
Tatoeba-test.eng-tlh.eng.tlh 0.0 0.035
Tatoeba-test.eng-tly.eng.tly 0.4 0.067
Tatoeba-test.eng-toi.eng.toi 7.0 0.229
Tatoeba-test.eng-ton.eng.ton 32.7 0.611
Tatoeba-test.eng-tpw.eng.tpw 1.0 0.127
Tatoeba-test.eng-tso.eng.tso 10.4 0.483
Tatoeba-test.eng-tuk.eng.tuk 8.0 0.360
Tatoeba-test.eng-tur.eng.tur 19.9 0.494
Tatoeba-test.eng-tvl.eng.tvl 31.5 0.510
Tatoeba-test.eng-tyv.eng.tyv 19.8 0.327
Tatoeba-test.eng-tzl.eng.tzl 0.4 0.109
Tatoeba-test.eng-udm.eng.udm 8.2 0.192
Tatoeba-test.eng-uig.eng.uig 1.8 0.246
Tatoeba-test.eng-ukr.eng.ukr 20.7 0.438
Tatoeba-test.eng-umb.eng.umb 6.1 0.298
Tatoeba-test.eng-urd.eng.urd 6.9 0.312
Tatoeba-test.eng-uzb.eng.uzb 3.8 0.298
Tatoeba-test.eng-vec.eng.vec 0.6 0.085
Tatoeba-test.eng-vie.eng.vie 20.8 0.392
Tatoeba-test.eng-vol.eng.vol 0.2 0.078
Tatoeba-test.eng-war.eng.war 11.0 0.403
Tatoeba-test.eng-wln.eng.wln 4.6 0.185
Tatoeba-test.eng-wol.eng.wol 7.1 0.214
Tatoeba-test.eng-xal.eng.xal 0.1 0.007
Tatoeba-test.eng-xho.eng.xho 18.6 0.555
Tatoeba-test.eng-yid.eng.yid 2.7 0.218
Tatoeba-test.eng-yor.eng.yor 10.2 0.230
Tatoeba-test.eng-zho.eng.zho 20.9 0.186
Tatoeba-test.eng-zul.eng.zul 32.1 0.666
Tatoeba-test.eng-zza.eng.zza 0.5 0.077

opus-2020-07-20.zip

  • dataset: opus
  • model: transformer
  • source language(s): eng
  • target language(s): abk acm ady afb afh_Latn afr akl_Latn aln amh ang_Latn apc ara arg arq ary arz asm ast avk_Latn awa aze_Latn bak bam_Latn bel bel_Latn ben bho bod bos_Latn bre brx brx_Latn bul bul_Latn cat ceb ces cha che chr chv cjy_Hans cjy_Hant cmn cmn_Hans cmn_Hant cor cos crh crh_Latn csb_Latn cym dan deu dsb dtp dws_Latn egl ell enm_Latn epo est eus ewe ext fao fij fin fkv_Latn fra frm_Latn frr fry fuc fuv gan gcf_Latn gil gla gle glg glv gom gos got_Goth grc_Grek grn gsw guj hat hau_Latn haw heb hif_Latn hil hin hnj_Latn hoc hoc_Latn hrv hsb hun hye iba ibo ido ido_Latn ike_Latn ile_Latn ilo ina_Latn ind isl ita izh jav jav_Java jbo jbo_Cyrl jbo_Latn jdt_Cyrl jpn kab kal kan kat kaz_Cyrl kaz_Latn kek_Latn kha khm khm_Latn kin kir_Cyrl kjh kpv krl ksh kum kur_Arab kur_Latn lad lad_Latn lao lat_Latn lav ldn_Latn lfn_Cyrl lfn_Latn lij lin lit liv_Latn lkt lld_Latn lmo ltg ltz lug lzh lzh_Hans mad mah mai mal mar max_Latn mdf mfe mhr mic min mkd mlg mlt mnw moh mon mri mwl mww mya myv nan nau nav nds niu nld nno nob nob_Hebr nog non_Latn nov_Latn npi nya oci ori orv_Cyrl oss ota_Arab ota_Latn pag pan_Guru pap pau pdc pes pes_Latn pes_Thaa pms pnb pol por ppl_Latn prg_Latn pus quc qya qya_Latn rap rif_Latn roh rom ron rue run rus sag sah san_Deva scn sco sgs shs_Latn shy_Latn sin sjn_Latn slv sma sme smo sna snd_Arab som spa sqi srp_Cyrl srp_Latn stq sun swe swg swh tah tam tat tat_Arab tat_Latn tel tet tgk_Cyrl tha tir tlh_Latn tly_Latn tmw_Latn toi_Latn ton tpw_Latn tso tuk tuk_Latn tur tvl tyv tzl tzl_Latn udm uig_Arab uig_Cyrl ukr umb urd uzb_Cyrl uzb_Latn vec vie vie_Hani vol_Latn vro war wln wol wuu xal xho yid yor yue yue_Hans yue_Hant zho zho_Hans zho_Hant zlm_Latn zsm_Latn zul zza
  • model: transformer
  • pre-processing: normalization + SentencePiece (spm32k,spm32k)
  • a sentence initial language token is required in the form of >>id<< (id = valid target language ID)
  • download: opus-2020-07-20.zip
  • test set translations: opus-2020-07-20.test.txt
  • test set scores: opus-2020-07-20.eval.txt

Benchmarks

testset BLEU chr-F
Tatoeba-test.eng-abk.eng.abk 1.2 0.244
Tatoeba-test.eng-ady.eng.ady 1.1 0.007
Tatoeba-test.eng-afh.eng.afh 0.2 0.028
Tatoeba-test.eng-afr.eng.afr 45.9 0.648
Tatoeba-test.eng-akl.eng.akl 1.1 0.069
Tatoeba-test.eng-amh.eng.amh 9.0 0.458
Tatoeba-test.eng-ang.eng.ang 5.5 0.141
Tatoeba-test.eng-ara.eng.ara 5.5 0.306
Tatoeba-test.eng-arg.eng.arg 1.5 0.112
Tatoeba-test.eng-asm.eng.asm 2.1 0.233
Tatoeba-test.eng-ast.eng.ast 12.7 0.378
Tatoeba-test.eng-avk.eng.avk 0.1 0.088
Tatoeba-test.eng-awa.eng.awa 0.1 0.002
Tatoeba-test.eng-aze.eng.aze 17.7 0.476
Tatoeba-test.eng-bak.eng.bak 6.7 0.381
Tatoeba-test.eng-bam.eng.bam 8.3 0.037
Tatoeba-test.eng-bel.eng.bel 12.4 0.382
Tatoeba-test.eng-ben.eng.ben 8.8 0.365
Tatoeba-test.eng-bho.eng.bho 0.1 0.003
Tatoeba-test.eng-bod.eng.bod 0.0 0.145
Tatoeba-test.eng-bre.eng.bre 6.2 0.260
Tatoeba-test.eng-brx.eng.brx 0.6 0.056
Tatoeba-test.eng-bul.eng.bul 30.9 0.528
Tatoeba-test.eng-cat.eng.cat 34.6 0.578
Tatoeba-test.eng-ceb.eng.ceb 9.5 0.395
Tatoeba-test.eng-ces.eng.ces 21.6 0.452
Tatoeba-test.eng-cha.eng.cha 4.0 0.224
Tatoeba-test.eng-che.eng.che 0.6 0.012
Tatoeba-test.eng-chm.eng.chm 2.3 0.214
Tatoeba-test.eng-chr.eng.chr 1.9 0.151
Tatoeba-test.eng-chv.eng.chv 6.2 0.352
Tatoeba-test.eng-cor.eng.cor 0.6 0.084
Tatoeba-test.eng-cos.eng.cos 3.2 0.146
Tatoeba-test.eng-crh.eng.crh 9.6 0.327
Tatoeba-test.eng-csb.eng.csb 3.0 0.217
Tatoeba-test.eng-cym.eng.cym 16.0 0.396
Tatoeba-test.eng-dan.eng.dan 40.2 0.605
Tatoeba-test.eng-deu.eng.deu 23.2 0.487
Tatoeba-test.eng-dsb.eng.dsb 0.6 0.115
Tatoeba-test.eng-dtp.eng.dtp 0.3 0.079
Tatoeba-test.eng-dws.eng.dws 0.8 0.078
Tatoeba-test.eng-egl.eng.egl 0.3 0.082
Tatoeba-test.eng-ell.eng.ell 26.3 0.514
Tatoeba-test.eng-enm.eng.enm 1.9 0.220
Tatoeba-test.eng-epo.eng.epo 31.5 0.552
Tatoeba-test.eng-est.eng.est 27.6 0.517
Tatoeba-test.eng-eus.eng.eus 16.8 0.455
Tatoeba-test.eng-ewe.eng.ewe 6.1 0.260
Tatoeba-test.eng-ext.eng.ext 2.3 0.172
Tatoeba-test.eng-fao.eng.fao 6.0 0.280
Tatoeba-test.eng-fas.eng.fas 6.7 0.258
Tatoeba-test.eng-fij.eng.fij 29.4 0.363
Tatoeba-test.eng-fin.eng.fin 17.0 0.440
Tatoeba-test.eng-fkv.eng.fkv 0.6 0.129
Tatoeba-test.eng-fra.eng.fra 29.7 0.531
Tatoeba-test.eng-frm.eng.frm 0.3 0.105
Tatoeba-test.eng-frr.eng.frr 7.6 0.062
Tatoeba-test.eng-fry.eng.fry 15.3 0.392
Tatoeba-test.eng-ful.eng.ful 0.3 0.057
Tatoeba-test.eng-gcf.eng.gcf 0.2 0.078
Tatoeba-test.eng-gil.eng.gil 46.3 0.672
Tatoeba-test.eng-gla.eng.gla 10.0 0.336
Tatoeba-test.eng-gle.eng.gle 24.3 0.467
Tatoeba-test.eng-glg.eng.glg 30.7 0.552
Tatoeba-test.eng-glv.eng.glv 21.5 0.372
Tatoeba-test.eng-gos.eng.gos 0.3 0.115
Tatoeba-test.eng-got.eng.got 0.2 0.009
Tatoeba-test.eng-grc.eng.grc 0.0 0.004
Tatoeba-test.eng-grn.eng.grn 1.6 0.139
Tatoeba-test.eng-gsw.eng.gsw 0.4 0.099
Tatoeba-test.eng-guj.eng.guj 15.1 0.344
Tatoeba-test.eng-hat.eng.hat 31.8 0.520
Tatoeba-test.eng-hau.eng.hau 5.7 0.360
Tatoeba-test.eng-haw.eng.haw 0.3 0.071
Tatoeba-test.eng-hbs.eng.hbs 26.8 0.513
Tatoeba-test.eng-heb.eng.heb 19.5 0.422
Tatoeba-test.eng-hif.eng.hif 0.7 0.205
Tatoeba-test.eng-hil.eng.hil 11.4 0.453
Tatoeba-test.eng-hin.eng.hin 11.8 0.381
Tatoeba-test.eng-hmn.eng.hmn 0.1 0.068
Tatoeba-test.eng-hoc.eng.hoc 0.1 0.073
Tatoeba-test.eng-hsb.eng.hsb 4.6 0.303
Tatoeba-test.eng-hun.eng.hun 15.1 0.406
Tatoeba-test.eng-hye.eng.hye 12.9 0.356
Tatoeba-test.eng-iba.eng.iba 15.3 0.418
Tatoeba-test.eng-ibo.eng.ibo 2.2 0.231
Tatoeba-test.eng-ido.eng.ido 10.0 0.385
Tatoeba-test.eng-iku.eng.iku 2.2 0.085
Tatoeba-test.eng-ile.eng.ile 4.1 0.273
Tatoeba-test.eng-ilo.eng.ilo 25.8 0.510
Tatoeba-test.eng-ina.eng.ina 20.6 0.486
Tatoeba-test.eng-isl.eng.isl 11.3 0.360
Tatoeba-test.eng-ita.eng.ita 32.4 0.577
Tatoeba-test.eng-izh.eng.izh 1.7 0.039
Tatoeba-test.eng-jav.eng.jav 8.0 0.320
Tatoeba-test.eng-jbo.eng.jbo 0.2 0.126
Tatoeba-test.eng-jdt.eng.jdt 5.3 0.008
Tatoeba-test.eng-jpn.eng.jpn 0.0 0.001
Tatoeba-test.eng-kab.eng.kab 4.6 0.249
Tatoeba-test.eng-kal.eng.kal 11.1 0.351
Tatoeba-test.eng-kan.eng.kan 4.2 0.337
Tatoeba-test.eng-kat.eng.kat 1.4 0.116
Tatoeba-test.eng-kaz.eng.kaz 9.5 0.326
Tatoeba-test.eng-kek.eng.kek 3.8 0.204
Tatoeba-test.eng-kha.eng.kha 0.4 0.089
Tatoeba-test.eng-khm.eng.khm 0.4 0.237
Tatoeba-test.eng-kin.eng.kin 10.2 0.398
Tatoeba-test.eng-kir.eng.kir 13.5 0.378
Tatoeba-test.eng-kjh.eng.kjh 0.8 0.009
Tatoeba-test.eng-kok.eng.kok 0.6 0.000
Tatoeba-test.eng-kom.eng.kom 0.6 0.008
Tatoeba-test.eng-krl.eng.krl 1.3 0.080
Tatoeba-test.eng-ksh.eng.ksh 0.4 0.109
Tatoeba-test.eng-kum.eng.kum 1.1 0.008
Tatoeba-test.eng-kur.eng.kur 7.2 0.165
Tatoeba-test.eng-lad.eng.lad 0.4 0.138
Tatoeba-test.eng-lah.eng.lah 0.1 0.000
Tatoeba-test.eng-lao.eng.lao 0.7 0.432
Tatoeba-test.eng-lat.eng.lat 7.6 0.357
Tatoeba-test.eng-lav.eng.lav 17.8 0.452
Tatoeba-test.eng-ldn.eng.ldn 0.4 0.065
Tatoeba-test.eng-lfn.eng.lfn 0.4 0.109
Tatoeba-test.eng-lij.eng.lij 0.6 0.120
Tatoeba-test.eng-lin.eng.lin 1.2 0.259
Tatoeba-test.eng-lit.eng.lit 14.4 0.457
Tatoeba-test.eng-liv.eng.liv 0.6 0.036
Tatoeba-test.eng-lkt.eng.lkt 0.5 0.051
Tatoeba-test.eng-lld.eng.lld 0.3 0.138
Tatoeba-test.eng-lmo.eng.lmo 2.4 0.135
Tatoeba-test.eng-ltz.eng.ltz 17.8 0.398
Tatoeba-test.eng-lug.eng.lug 21.7 0.570
Tatoeba-test.eng-mad.eng.mad 0.7 0.050
Tatoeba-test.eng-mah.eng.mah 9.4 0.345
Tatoeba-test.eng-mai.eng.mai 10.7 0.396
Tatoeba-test.eng-mal.eng.mal 7.2 0.413
Tatoeba-test.eng-mar.eng.mar 11.7 0.380
Tatoeba-test.eng-mdf.eng.mdf 0.3 0.004
Tatoeba-test.eng-mfe.eng.mfe 83.6 0.911
Tatoeba-test.eng-mic.eng.mic 3.2 0.181
Tatoeba-test.eng-mkd.eng.mkd 29.6 0.525
Tatoeba-test.eng-mlg.eng.mlg 27.3 0.433
Tatoeba-test.eng-mlt.eng.mlt 9.3 0.395
Tatoeba-test.eng-mnw.eng.mnw 1.5 0.003
Tatoeba-test.eng-moh.eng.moh 0.6 0.060
Tatoeba-test.eng-mon.eng.mon 8.4 0.307
Tatoeba-test.eng-mri.eng.mri 8.7 0.316
Tatoeba-test.eng-msa.eng.msa 22.5 0.484
Tatoeba-test.eng.multi 19.5 0.430
Tatoeba-test.eng-mwl.eng.mwl 0.5 0.087
Tatoeba-test.eng-mya.eng.mya 0.1 0.111
Tatoeba-test.eng-myv.eng.myv 0.4 0.010
Tatoeba-test.eng-nau.eng.nau 0.6 0.130
Tatoeba-test.eng-nav.eng.nav 2.3 0.157
Tatoeba-test.eng-nds.eng.nds 10.2 0.348
Tatoeba-test.eng-nep.eng.nep 0.2 0.003
Tatoeba-test.eng-niu.eng.niu 31.9 0.470
Tatoeba-test.eng-nld.eng.nld 35.1 0.578
Tatoeba-test.eng-nog.eng.nog 0.2 0.009
Tatoeba-test.eng-non.eng.non 0.2 0.085
Tatoeba-test.eng-nor.eng.nor 33.7 0.559
Tatoeba-test.eng-nov.eng.nov 1.4 0.205
Tatoeba-test.eng-nya.eng.nya 8.8 0.490
Tatoeba-test.eng-oci.eng.oci 5.1 0.269
Tatoeba-test.eng-ori.eng.ori 4.0 0.256
Tatoeba-test.eng-orv.eng.orv 0.2 0.007
Tatoeba-test.eng-oss.eng.oss 3.3 0.220
Tatoeba-test.eng-ota.eng.ota 0.1 0.052
Tatoeba-test.eng-pag.eng.pag 20.3 0.492
Tatoeba-test.eng-pan.eng.pan 5.9 0.289
Tatoeba-test.eng-pap.eng.pap 46.4 0.628
Tatoeba-test.eng-pau.eng.pau 0.8 0.096
Tatoeba-test.eng-pdc.eng.pdc 2.4 0.149
Tatoeba-test.eng-pms.eng.pms 0.5 0.110
Tatoeba-test.eng-pol.eng.pol 20.5 0.478
Tatoeba-test.eng-por.eng.por 30.0 0.558
Tatoeba-test.eng-ppl.eng.ppl 0.6 0.091
Tatoeba-test.eng-prg.eng.prg 0.2 0.082
Tatoeba-test.eng-pus.eng.pus 1.3 0.144
Tatoeba-test.eng-quc.eng.quc 2.4 0.218
Tatoeba-test.eng-qya.eng.qya 0.4 0.087
Tatoeba-test.eng-rap.eng.rap 1.8 0.046
Tatoeba-test.eng-rif.eng.rif 1.4 0.111
Tatoeba-test.eng-roh.eng.roh 1.5 0.226
Tatoeba-test.eng-rom.eng.rom 8.7 0.340
Tatoeba-test.eng-ron.eng.ron 24.9 0.512
Tatoeba-test.eng-rue.eng.rue 0.1 0.008
Tatoeba-test.eng-run.eng.run 8.8 0.415
Tatoeba-test.eng-rus.eng.rus 21.1 0.449
Tatoeba-test.eng-sag.eng.sag 5.0 0.158
Tatoeba-test.eng-sah.eng.sah 0.2 0.006
Tatoeba-test.eng-san.eng.san 2.2 0.136
Tatoeba-test.eng-scn.eng.scn 0.4 0.047
Tatoeba-test.eng-sco.eng.sco 29.6 0.545
Tatoeba-test.eng-sgs.eng.sgs 2.6 0.103
Tatoeba-test.eng-shs.eng.shs 0.9 0.054
Tatoeba-test.eng-shy.eng.shy 0.3 0.063
Tatoeba-test.eng-sin.eng.sin 10.0 0.310
Tatoeba-test.eng-sjn.eng.sjn 0.3 0.072
Tatoeba-test.eng-slv.eng.slv 11.5 0.303
Tatoeba-test.eng-sma.eng.sma 0.6 0.065
Tatoeba-test.eng-sme.eng.sme 13.1 0.281
Tatoeba-test.eng-smo.eng.smo 21.9 0.443
Tatoeba-test.eng-sna.eng.sna 27.2 0.533
Tatoeba-test.eng-snd.eng.snd 5.7 0.337
Tatoeba-test.eng-som.eng.som 0.0 0.296
Tatoeba-test.eng-spa.eng.spa 35.5 0.591
Tatoeba-test.eng-sqi.eng.sqi 30.3 0.557
Tatoeba-test.eng-stq.eng.stq 6.5 0.331
Tatoeba-test.eng-sun.eng.sun 44.5 0.458
Tatoeba-test.eng-swa.eng.swa 1.0 0.105
Tatoeba-test.eng-swe.eng.swe 41.0 0.598
Tatoeba-test.eng-swg.eng.swg 0.5 0.117
Tatoeba-test.eng-tah.eng.tah 11.8 0.297
Tatoeba-test.eng-tam.eng.tam 4.7 0.391
Tatoeba-test.eng-tat.eng.tat 9.1 0.313
Tatoeba-test.eng-tel.eng.tel 5.4 0.323
Tatoeba-test.eng-tet.eng.tet 0.6 0.093
Tatoeba-test.eng-tgk.eng.tgk 8.2 0.319
Tatoeba-test.eng-tha.eng.tha 2.0 0.445
Tatoeba-test.eng-tir.eng.tir 1.5 0.190
Tatoeba-test.eng-tlh.eng.tlh 0.0 0.077
Tatoeba-test.eng-tly.eng.tly 0.2 0.049
Tatoeba-test.eng-toi.eng.toi 7.0 0.223
Tatoeba-test.eng-ton.eng.ton 31.2 0.590
Tatoeba-test.eng-tpw.eng.tpw 0.4 0.102
Tatoeba-test.eng-tso.eng.tso 10.4 0.468
Tatoeba-test.eng-tuk.eng.tuk 8.8 0.369
Tatoeba-test.eng-tur.eng.tur 21.9 0.519
Tatoeba-test.eng-tvl.eng.tvl 34.0 0.523
Tatoeba-test.eng-tyv.eng.tyv 18.7 0.381
Tatoeba-test.eng-tzl.eng.tzl 0.3 0.099
Tatoeba-test.eng-udm.eng.udm 8.7 0.249
Tatoeba-test.eng-uig.eng.uig 2.0 0.256
Tatoeba-test.eng-ukr.eng.ukr 22.6 0.459
Tatoeba-test.eng-umb.eng.umb 4.2 0.291
Tatoeba-test.eng-urd.eng.urd 8.0 0.327
Tatoeba-test.eng-uzb.eng.uzb 6.3 0.326
Tatoeba-test.eng-vec.eng.vec 1.0 0.086
Tatoeba-test.eng-vie.eng.vie 22.3 0.408
Tatoeba-test.eng-vol.eng.vol 0.2 0.069
Tatoeba-test.eng-war.eng.war 11.3 0.409
Tatoeba-test.eng-wln.eng.wln 7.1 0.199
Tatoeba-test.eng-wol.eng.wol 8.3 0.253
Tatoeba-test.eng-xal.eng.xal 0.0 0.006
Tatoeba-test.eng-xho.eng.xho 20.2 0.558
Tatoeba-test.eng-yid.eng.yid 3.0 0.237
Tatoeba-test.eng-yor.eng.yor 11.9 0.270
Tatoeba-test.eng-zho.eng.zho 21.8 0.192
Tatoeba-test.eng-zul.eng.zul 34.8 0.663
Tatoeba-test.eng-zza.eng.zza 0.4 0.067

opus-2020-07-27.zip

  • dataset: opus
  • model: transformer
  • source language(s): eng
  • target language(s): abk acm ady afb afh_Latn afr akl_Latn aln amh ang_Latn apc ara arg arq ary arz asm ast avk_Latn awa aze_Latn bak bam_Latn bel bel_Latn ben bho bod bos_Latn bre brx brx_Latn bul bul_Latn cat ceb ces cha che chr chv cjy_Hans cjy_Hant cmn cmn_Hans cmn_Hant cor cos crh crh_Latn csb_Latn cym dan deu dsb dtp dws_Latn egl ell enm_Latn epo est eus ewe ext fao fij fin fkv_Latn fra frm_Latn frr fry fuc fuv gan gcf_Latn gil gla gle glg glv gom gos got_Goth grc_Grek grn gsw guj hat hau_Latn haw heb hif_Latn hil hin hnj_Latn hoc hoc_Latn hrv hsb hun hye iba ibo ido ido_Latn ike_Latn ile_Latn ilo ina_Latn ind isl ita izh jav jav_Java jbo jbo_Cyrl jbo_Latn jdt_Cyrl jpn kab kal kan kat kaz_Cyrl kaz_Latn kek_Latn kha khm khm_Latn kin kir_Cyrl kjh kpv krl ksh kum kur_Arab kur_Latn lad lad_Latn lao lat_Latn lav ldn_Latn lfn_Cyrl lfn_Latn lij lin lit liv_Latn lkt lld_Latn lmo ltg ltz lug lzh lzh_Hans mad mah mai mal mar max_Latn mdf mfe mhr mic min mkd mlg mlt mnw moh mon mri mwl mww mya myv nan nau nav nds niu nld nno nob nob_Hebr nog non_Latn nov_Latn npi nya oci ori orv_Cyrl oss ota_Arab ota_Latn pag pan_Guru pap pau pdc pes pes_Latn pes_Thaa pms pnb pol por ppl_Latn prg_Latn pus quc qya qya_Latn rap rif_Latn roh rom ron rue run rus sag sah san_Deva scn sco sgs shs_Latn shy_Latn sin sjn_Latn slv sma sme smo sna snd_Arab som spa sqi srp_Cyrl srp_Latn stq sun swe swg swh tah tam tat tat_Arab tat_Latn tel tet tgk_Cyrl tha tir tlh_Latn tly_Latn tmw_Latn toi_Latn ton tpw_Latn tso tuk tuk_Latn tur tvl tyv tzl tzl_Latn udm uig_Arab uig_Cyrl ukr umb urd uzb_Cyrl uzb_Latn vec vie vie_Hani vol_Latn vro war wln wol wuu xal xho yid yor yue yue_Hans yue_Hant zho zho_Hans zho_Hant zlm_Latn zsm_Latn zul zza
  • model: transformer
  • pre-processing: normalization + SentencePiece (spm32k,spm32k)
  • a sentence initial language token is required in the form of >>id<< (id = valid target language ID)
  • download: opus-2020-07-27.zip
  • test set translations: opus-2020-07-27.test.txt
  • test set scores: opus-2020-07-27.eval.txt

Benchmarks

testset BLEU chr-F
newsdev2014-enghin.eng.hin 4.5 0.281
newsdev2015-enfi-engfin.eng.fin 8.5 0.406
newsdev2016-enro-engron.eng.ron 16.7 0.482
newsdev2016-entr-engtur.eng.tur 7.9 0.392
newsdev2017-enlv-englav.eng.lav 12.3 0.434
newsdev2017-enzh-engzho.eng.zho 7.3 0.062
newsdev2018-enet-engest.eng.est 10.3 0.413
newsdev2019-engu-engguj.eng.guj 4.5 0.266
newsdev2019-enlt-englit.eng.lit 10.6 0.424
newsdiscussdev2015-enfr-engfra.eng.fra 22.0 0.514
newsdiscusstest2015-enfr-engfra.eng.fra 24.6 0.537
newssyscomb2009-engces.eng.ces 9.8 0.385
newssyscomb2009-engdeu.eng.deu 12.8 0.454
newssyscomb2009-engfra.eng.fra 20.7 0.518
newssyscomb2009-enghun.eng.hun 7.7 0.363
newssyscomb2009-engita.eng.ita 21.5 0.535
newssyscomb2009-engspa.eng.spa 23.7 0.529
news-test2008-engces.eng.ces 8.8 0.370
news-test2008-engdeu.eng.deu 13.5 0.445
news-test2008-engfra.eng.fra 19.1 0.494
news-test2008-engspa.eng.spa 22.4 0.515
newstest2009-engces.eng.ces 9.3 0.383
newstest2009-engdeu.eng.deu 13.0 0.449
newstest2009-engfra.eng.fra 20.0 0.509
newstest2009-enghun.eng.hun 8.0 0.360
newstest2009-engita.eng.ita 21.5 0.531
newstest2009-engspa.eng.spa 23.4 0.531
newstest2010-engces.eng.ces 9.7 0.384
newstest2010-engdeu.eng.deu 14.5 0.457
newstest2010-engfra.eng.fra 21.1 0.518
newstest2010-engspa.eng.spa 26.5 0.551
newstest2011-engces.eng.ces 10.3 0.384
newstest2011-engdeu.eng.deu 13.5 0.443
newstest2011-engfra.eng.fra 23.5 0.538
newstest2011-engspa.eng.spa 27.9 0.555
newstest2012-engces.eng.ces 9.4 0.368
newstest2012-engdeu.eng.deu 13.6 0.443
newstest2012-engfra.eng.fra 22.4 0.525
newstest2012-engrus.eng.rus 15.1 0.455
newstest2012-engspa.eng.spa 27.3 0.551
newstest2013-engces.eng.ces 10.2 0.384
newstest2013-engdeu.eng.deu 15.5 0.462
newstest2013-engfra.eng.fra 22.1 0.511
newstest2013-engrus.eng.rus 11.4 0.408
newstest2013-engspa.eng.spa 24.6 0.530
newstest2014-hien-enghin.eng.hin 6.7 0.301
newstest2015-encs-engces.eng.ces 10.0 0.384
newstest2015-ende-engdeu.eng.deu 17.9 0.484
newstest2015-enfi-engfin.eng.fin 9.3 0.410
newstest2015-enru-engrus.eng.rus 13.7 0.438
newstest2016-encs-engces.eng.ces 11.5 0.399
newstest2016-ende-engdeu.eng.deu 20.4 0.508
newstest2016-enfi-engfin.eng.fin 10.3 0.425
newstest2016-enro-engron.eng.ron 15.8 0.464
newstest2016-enru-engrus.eng.rus 12.6 0.426
newstest2016-entr-engtur.eng.tur 7.6 0.374
newstest2017-encs-engces.eng.ces 9.1 0.369
newstest2017-ende-engdeu.eng.deu 17.0 0.475
newstest2017-enfi-engfin.eng.fin 11.3 0.429
newstest2017-enlv-englav.eng.lav 9.3 0.405
newstest2017-enru-engrus.eng.rus 13.2 0.438
newstest2017-entr-engtur.eng.tur 7.5 0.368
newstest2017-enzh-engzho.eng.zho 7.2 0.075
newstest2018-encs-engces.eng.ces 9.2 0.373
newstest2018-ende-engdeu.eng.deu 22.9 0.528
newstest2018-enet-engest.eng.est 11.4 0.425
newstest2018-enfi-engfin.eng.fin 7.5 0.389
newstest2018-enru-engrus.eng.rus 11.5 0.427
newstest2018-entr-engtur.eng.tur 7.0 0.372
newstest2018-enzh-engzho.eng.zho 7.9 0.073
newstest2019-encs-engces.eng.ces 10.0 0.381
newstest2019-ende-engdeu.eng.deu 20.2 0.505
newstest2019-enfi-engfin.eng.fin 9.8 0.409
newstest2019-engu-engguj.eng.guj 4.8 0.277
newstest2019-enlt-englit.eng.lit 7.3 0.378
newstest2019-enru-engrus.eng.rus 12.2 0.408
newstest2019-enzh-engzho.eng.zho 5.9 0.100
newstestB2016-enfi-engfin.eng.fin 8.3 0.404
newstestB2017-enfi-engfin.eng.fin 9.1 0.405
Tatoeba-test.eng-abk.eng.abk 1.3 0.247
Tatoeba-test.eng-ady.eng.ady 0.2 0.003
Tatoeba-test.eng-afh.eng.afh 0.2 0.028
Tatoeba-test.eng-afr.eng.afr 46.5 0.655
Tatoeba-test.eng-akl.eng.akl 0.7 0.057
Tatoeba-test.eng-amh.eng.amh 9.1 0.450
Tatoeba-test.eng-ang.eng.ang 5.8 0.141
Tatoeba-test.eng-ara.eng.ara 5.9 0.314
Tatoeba-test.eng-arg.eng.arg 1.4 0.124
Tatoeba-test.eng-asm.eng.asm 2.0 0.234
Tatoeba-test.eng-ast.eng.ast 13.6 0.382
Tatoeba-test.eng-avk.eng.avk 0.1 0.059
Tatoeba-test.eng-awa.eng.awa 0.1 0.001
Tatoeba-test.eng-aze.eng.aze 18.7 0.489
Tatoeba-test.eng-bak.eng.bak 8.3 0.376
Tatoeba-test.eng-bam.eng.bam 8.3 0.037
Tatoeba-test.eng-bel.eng.bel 13.0 0.387
Tatoeba-test.eng-ben.eng.ben 9.6 0.372
Tatoeba-test.eng-bho.eng.bho 0.1 0.002
Tatoeba-test.eng-bod.eng.bod 0.0 0.150
Tatoeba-test.eng-bre.eng.bre 7.2 0.264
Tatoeba-test.eng-brx.eng.brx 0.3 0.050
Tatoeba-test.eng-bul.eng.bul 31.4 0.534
Tatoeba-test.eng-cat.eng.cat 35.2 0.582
Tatoeba-test.eng-ceb.eng.ceb 9.2 0.399
Tatoeba-test.eng-ces.eng.ces 22.8 0.463
Tatoeba-test.eng-cha.eng.cha 5.1 0.227
Tatoeba-test.eng-che.eng.che 0.2 0.011
Tatoeba-test.eng-chm.eng.chm 2.5 0.234
Tatoeba-test.eng-chr.eng.chr 3.7 0.156
Tatoeba-test.eng-chv.eng.chv 6.4 0.362
Tatoeba-test.eng-cor.eng.cor 0.7 0.088
Tatoeba-test.eng-cos.eng.cos 1.4 0.108
Tatoeba-test.eng-crh.eng.crh 5.9 0.318
Tatoeba-test.eng-csb.eng.csb 2.8 0.224
Tatoeba-test.eng-cym.eng.cym 16.5 0.401
Tatoeba-test.eng-dan.eng.dan 41.2 0.614
Tatoeba-test.eng-deu.eng.deu 23.8 0.494
Tatoeba-test.eng-dsb.eng.dsb 0.1 0.051
Tatoeba-test.eng-dtp.eng.dtp 0.1 0.056
Tatoeba-test.eng-dws.eng.dws 0.1 0.037
Tatoeba-test.eng-egl.eng.egl 0.1 0.054
Tatoeba-test.eng-ell.eng.ell 27.0 0.523
Tatoeba-test.eng-enm.eng.enm 0.4 0.119
Tatoeba-test.eng-epo.eng.epo 32.6 0.562
Tatoeba-test.eng-est.eng.est 29.1 0.530
Tatoeba-test.eng-eus.eng.eus 17.4 0.469
Tatoeba-test.eng-ewe.eng.ewe 6.5 0.259
Tatoeba-test.eng-ext.eng.ext 2.2 0.152
Tatoeba-test.eng-fao.eng.fao 5.7 0.280
Tatoeba-test.eng-fas.eng.fas 6.8 0.260
Tatoeba-test.eng-fij.eng.fij 29.6 0.357
Tatoeba-test.eng-fin.eng.fin 17.7 0.447
Tatoeba-test.eng-fkv.eng.fkv 0.3 0.103
Tatoeba-test.eng-fra.eng.fra 30.4 0.537
Tatoeba-test.eng-frm.eng.frm 0.1 0.089
Tatoeba-test.eng-frr.eng.frr 1.3 0.006
Tatoeba-test.eng-fry.eng.fry 16.1 0.383
Tatoeba-test.eng-ful.eng.ful 0.1 0.030
Tatoeba-test.eng-gcf.eng.gcf 0.0 0.034
Tatoeba-test.eng-gil.eng.gil 44.4 0.680
Tatoeba-test.eng-gla.eng.gla 10.2 0.339
Tatoeba-test.eng-gle.eng.gle 25.4 0.479
Tatoeba-test.eng-glg.eng.glg 31.6 0.559
Tatoeba-test.eng-glv.eng.glv 20.2 0.404
Tatoeba-test.eng-gos.eng.gos 0.1 0.079
Tatoeba-test.eng-got.eng.got 0.1 0.005
Tatoeba-test.eng-grc.eng.grc 0.0 0.003
Tatoeba-test.eng-grn.eng.grn 1.6 0.137
Tatoeba-test.eng-gsw.eng.gsw 0.2 0.064
Tatoeba-test.eng-guj.eng.guj 16.2 0.355
Tatoeba-test.eng-hat.eng.hat 33.1 0.545
Tatoeba-test.eng-hau.eng.hau 3.4 0.375
Tatoeba-test.eng-haw.eng.haw 0.1 0.043
Tatoeba-test.eng-hbs.eng.hbs 27.6 0.519
Tatoeba-test.eng-heb.eng.heb 20.4 0.433
Tatoeba-test.eng-hif.eng.hif 0.3 0.142
Tatoeba-test.eng-hil.eng.hil 12.7 0.452
Tatoeba-test.eng-hin.eng.hin 12.3 0.389
Tatoeba-test.eng-hmn.eng.hmn 0.1 0.039
Tatoeba-test.eng-hoc.eng.hoc 0.0 0.053
Tatoeba-test.eng-hsb.eng.hsb 5.0 0.292
Tatoeba-test.eng-hun.eng.hun 15.7 0.416
Tatoeba-test.eng-hye.eng.hye 13.1 0.359
Tatoeba-test.eng-iba.eng.iba 15.6 0.423
Tatoeba-test.eng-ibo.eng.ibo 2.2 0.225
Tatoeba-test.eng-ido.eng.ido 10.7 0.391
Tatoeba-test.eng-iku.eng.iku 1.1 0.055
Tatoeba-test.eng-ile.eng.ile 4.1 0.275
Tatoeba-test.eng-ilo.eng.ilo 25.9 0.511
Tatoeba-test.eng-ina.eng.ina 20.7 0.488
Tatoeba-test.eng-isl.eng.isl 11.4 0.367
Tatoeba-test.eng-ita.eng.ita 33.4 0.585
Tatoeba-test.eng-izh.eng.izh 0.2 0.041
Tatoeba-test.eng-jav.eng.jav 7.9 0.319
Tatoeba-test.eng-jbo.eng.jbo 0.2 0.127
Tatoeba-test.eng-jdt.eng.jdt 1.1 0.005
Tatoeba-test.eng-jpn.eng.jpn 0.0 0.000
Tatoeba-test.eng-kab.eng.kab 5.4 0.261
Tatoeba-test.eng-kal.eng.kal 10.1 0.334
Tatoeba-test.eng-kan.eng.kan 4.3 0.350
Tatoeba-test.eng-kat.eng.kat 1.5 0.120
Tatoeba-test.eng-kaz.eng.kaz 9.3 0.343
Tatoeba-test.eng-kek.eng.kek 4.0 0.206
Tatoeba-test.eng-kha.eng.kha 0.1 0.060
Tatoeba-test.eng-khm.eng.khm 0.4 0.242
Tatoeba-test.eng-kin.eng.kin 7.6 0.399
Tatoeba-test.eng-kir.eng.kir 12.6 0.375
Tatoeba-test.eng-kjh.eng.kjh 0.2 0.004
Tatoeba-test.eng-kok.eng.kok 0.5 0.000
Tatoeba-test.eng-kom.eng.kom 0.5 0.008
Tatoeba-test.eng-krl.eng.krl 0.8 0.062
Tatoeba-test.eng-ksh.eng.ksh 0.3 0.093
Tatoeba-test.eng-kum.eng.kum 0.3 0.004
Tatoeba-test.eng-kur.eng.kur 6.9 0.165
Tatoeba-test.eng-lad.eng.lad 0.2 0.096
Tatoeba-test.eng-lah.eng.lah 0.1 0.000
Tatoeba-test.eng-lao.eng.lao 0.6 0.423
Tatoeba-test.eng-lat.eng.lat 7.7 0.359
Tatoeba-test.eng-lav.eng.lav 18.5 0.462
Tatoeba-test.eng-ldn.eng.ldn 0.1 0.047
Tatoeba-test.eng-lfn.eng.lfn 0.5 0.112
Tatoeba-test.eng-lij.eng.lij 0.3 0.078
Tatoeba-test.eng-lin.eng.lin 1.1 0.259
Tatoeba-test.eng-lit.eng.lit 15.1 0.466
Tatoeba-test.eng-liv.eng.liv 0.1 0.021
Tatoeba-test.eng-lkt.eng.lkt 0.2 0.029
Tatoeba-test.eng-lld.eng.lld 0.1 0.051
Tatoeba-test.eng-lmo.eng.lmo 0.3 0.078
Tatoeba-test.eng-ltz.eng.ltz 19.8 0.415
Tatoeba-test.eng-lug.eng.lug 4.2 0.474
Tatoeba-test.eng-mad.eng.mad 0.4 0.035
Tatoeba-test.eng-mah.eng.mah 9.7 0.330
Tatoeba-test.eng-mai.eng.mai 10.7 0.396
Tatoeba-test.eng-mal.eng.mal 6.9 0.414
Tatoeba-test.eng-mar.eng.mar 12.3 0.389
Tatoeba-test.eng-mdf.eng.mdf 0.1 0.002
Tatoeba-test.eng-mfe.eng.mfe 83.6 0.911
Tatoeba-test.eng-mic.eng.mic 0.8 0.055
Tatoeba-test.eng-mkd.eng.mkd 30.3 0.531
Tatoeba-test.eng-mlg.eng.mlg 26.9 0.450
Tatoeba-test.eng-mlt.eng.mlt 10.8 0.412
Tatoeba-test.eng-mnw.eng.mnw 0.0 0.000
Tatoeba-test.eng-moh.eng.moh 0.3 0.047
Tatoeba-test.eng-mon.eng.mon 8.7 0.310
Tatoeba-test.eng-mri.eng.mri 8.4 0.316
Tatoeba-test.eng-msa.eng.msa 19.5 0.484
Tatoeba-test.eng.multi 19.1 0.434
Tatoeba-test.eng-mwl.eng.mwl 0.8 0.096
Tatoeba-test.eng-mya.eng.mya 0.1 0.108
Tatoeba-test.eng-myv.eng.myv 0.2 0.007
Tatoeba-test.eng-nau.eng.nau 0.4 0.087
Tatoeba-test.eng-nav.eng.nav 2.8 0.167
Tatoeba-test.eng-nds.eng.nds 10.1 0.353
Tatoeba-test.eng-nep.eng.nep 0.2 0.002
Tatoeba-test.eng-niu.eng.niu 36.3 0.501
Tatoeba-test.eng-nld.eng.nld 36.3 0.587
Tatoeba-test.eng-nog.eng.nog 0.1 0.002
Tatoeba-test.eng-non.eng.non 0.1 0.072
Tatoeba-test.eng-nor.eng.nor 35.1 0.567
Tatoeba-test.eng-nov.eng.nov 0.4 0.131
Tatoeba-test.eng-nya.eng.nya 9.2 0.488
Tatoeba-test.eng-oci.eng.oci 5.5 0.280
Tatoeba-test.eng-ori.eng.ori 6.6 0.270
Tatoeba-test.eng-orv.eng.orv 0.0 0.003
Tatoeba-test.eng-oss.eng.oss 3.4 0.228
Tatoeba-test.eng-ota.eng.ota 0.0 0.034
Tatoeba-test.eng-pag.eng.pag 19.8 0.467
Tatoeba-test.eng-pan.eng.pan 7.7 0.297
Tatoeba-test.eng-pap.eng.pap 46.7 0.635
Tatoeba-test.eng-pau.eng.pau 0.4 0.073
Tatoeba-test.eng-pdc.eng.pdc 0.9 0.072
Tatoeba-test.eng-pms.eng.pms 0.2 0.088
Tatoeba-test.eng-pol.eng.pol 21.4 0.487
Tatoeba-test.eng-por.eng.por 30.6 0.563
Tatoeba-test.eng-ppl.eng.ppl 0.3 0.073
Tatoeba-test.eng-prg.eng.prg 0.1 0.053
Tatoeba-test.eng-pus.eng.pus 0.7 0.138
Tatoeba-test.eng-quc.eng.quc 2.7 0.216
Tatoeba-test.eng-qya.eng.qya 0.2 0.065
Tatoeba-test.eng-rap.eng.rap 0.2 0.038
Tatoeba-test.eng-rif.eng.rif 0.3 0.051
Tatoeba-test.eng-roh.eng.roh 1.2 0.221
Tatoeba-test.eng-rom.eng.rom 8.0 0.333
Tatoeba-test.eng-ron.eng.ron 25.8 0.520
Tatoeba-test.eng-rue.eng.rue 0.1 0.005
Tatoeba-test.eng-run.eng.run 9.3 0.421
Tatoeba-test.eng-rus.eng.rus 21.5 0.455
Tatoeba-test.eng-sag.eng.sag 4.8 0.145
Tatoeba-test.eng-sah.eng.sah 0.1 0.003
Tatoeba-test.eng-san.eng.san 1.8 0.131
Tatoeba-test.eng-scn.eng.scn 0.4 0.047
Tatoeba-test.eng-sco.eng.sco 32.6 0.561
Tatoeba-test.eng-sgs.eng.sgs 0.7 0.070
Tatoeba-test.eng-shs.eng.shs 0.3 0.033
Tatoeba-test.eng-shy.eng.shy 0.3 0.060
Tatoeba-test.eng-sin.eng.sin 10.0 0.322
Tatoeba-test.eng-sjn.eng.sjn 0.1 0.040
Tatoeba-test.eng-slv.eng.slv 11.9 0.309
Tatoeba-test.eng-sma.eng.sma 0.1 0.035
Tatoeba-test.eng-sme.eng.sme 12.2 0.272
Tatoeba-test.eng-smo.eng.smo 15.3 0.413
Tatoeba-test.eng-sna.eng.sna 27.3 0.523
Tatoeba-test.eng-snd.eng.snd 6.0 0.349
Tatoeba-test.eng-som.eng.som 0.0 0.296
Tatoeba-test.eng-spa.eng.spa 36.0 0.596
Tatoeba-test.eng-sqi.eng.sqi 30.4 0.561
Tatoeba-test.eng-stq.eng.stq 2.1 0.297
Tatoeba-test.eng-sun.eng.sun 43.0 0.462
Tatoeba-test.eng-swa.eng.swa 0.4 0.076
Tatoeba-test.eng-swe.eng.swe 41.5 0.604
Tatoeba-test.eng-swg.eng.swg 0.3 0.096
Tatoeba-test.eng-tah.eng.tah 9.9 0.286
Tatoeba-test.eng-tam.eng.tam 5.6 0.399
Tatoeba-test.eng-tat.eng.tat 9.6 0.316
Tatoeba-test.eng-tel.eng.tel 5.7 0.322
Tatoeba-test.eng-tet.eng.tet 0.2 0.052
Tatoeba-test.eng-tgk.eng.tgk 9.5 0.324
Tatoeba-test.eng-tha.eng.tha 2.3 0.448
Tatoeba-test.eng-tir.eng.tir 1.4 0.184
Tatoeba-test.eng-tlh.eng.tlh 0.0 0.057
Tatoeba-test.eng-tly.eng.tly 0.1 0.045
Tatoeba-test.eng-toi.eng.toi 7.0 0.253
Tatoeba-test.eng-ton.eng.ton 31.0 0.578
Tatoeba-test.eng-tpw.eng.tpw 0.6 0.101
Tatoeba-test.eng-tso.eng.tso 10.4 0.468
Tatoeba-test.eng-tuk.eng.tuk 9.0 0.375
Tatoeba-test.eng-tur.eng.tur 22.9 0.530
Tatoeba-test.eng-tvl.eng.tvl 20.9 0.444
Tatoeba-test.eng-tyv.eng.tyv 20.9 0.369
Tatoeba-test.eng-tzl.eng.tzl 0.1 0.058
Tatoeba-test.eng-udm.eng.udm 8.3 0.276
Tatoeba-test.eng-uig.eng.uig 2.1 0.258
Tatoeba-test.eng-ukr.eng.ukr 23.5 0.470
Tatoeba-test.eng-umb.eng.umb 4.2 0.289
Tatoeba-test.eng-urd.eng.urd 8.1 0.336
Tatoeba-test.eng-uzb.eng.uzb 5.9 0.328
Tatoeba-test.eng-vec.eng.vec 0.4 0.072
Tatoeba-test.eng-vie.eng.vie 22.7 0.414
Tatoeba-test.eng-vol.eng.vol 0.3 0.083
Tatoeba-test.eng-war.eng.war 11.2 0.411
Tatoeba-test.eng-wln.eng.wln 7.2 0.198
Tatoeba-test.eng-wol.eng.wol 8.1 0.206
Tatoeba-test.eng-xal.eng.xal 0.0 0.003
Tatoeba-test.eng-xho.eng.xho 22.2 0.560
Tatoeba-test.eng-yid.eng.yid 3.9 0.244
Tatoeba-test.eng-yor.eng.yor 10.3 0.243
Tatoeba-test.eng-zho.eng.zho 22.1 0.194
Tatoeba-test.eng-zul.eng.zul 33.3 0.653
Tatoeba-test.eng-zza.eng.zza 0.1 0.045

opus2m-2020-08-01.zip

  • dataset: opus2m
  • model: transformer
  • source language(s): eng
  • target language(s): abk acm ady afb afh_Latn afr akl_Latn aln amh ang_Latn apc ara arg arq ary arz asm ast avk_Latn awa aze_Latn bak bam_Latn bel bel_Latn ben bho bod bos_Latn bre brx brx_Latn bul bul_Latn cat ceb ces cha che chr chv cjy_Hans cjy_Hant cmn cmn_Hans cmn_Hant cor cos crh crh_Latn csb_Latn cym dan deu dsb dtp dws_Latn egl ell enm_Latn epo est eus ewe ext fao fij fin fkv_Latn fra frm_Latn frr fry fuc fuv gan gcf_Latn gil gla gle glg glv gom gos got_Goth grc_Grek grn gsw guj hat hau_Latn haw heb hif_Latn hil hin hnj_Latn hoc hoc_Latn hrv hsb hun hye iba ibo ido ido_Latn ike_Latn ile_Latn ilo ina_Latn ind isl ita izh jav jav_Java jbo jbo_Cyrl jbo_Latn jdt_Cyrl jpn kab kal kan kat kaz_Cyrl kaz_Latn kek_Latn kha khm khm_Latn kin kir_Cyrl kjh kpv krl ksh kum kur_Arab kur_Latn lad lad_Latn lao lat_Latn lav ldn_Latn lfn_Cyrl lfn_Latn lij lin lit liv_Latn lkt lld_Latn lmo ltg ltz lug lzh lzh_Hans mad mah mai mal mar max_Latn mdf mfe mhr mic min mkd mlg mlt mnw moh mon mri mwl mww mya myv nan nau nav nds niu nld nno nob nob_Hebr nog non_Latn nov_Latn npi nya oci ori orv_Cyrl oss ota_Arab ota_Latn pag pan_Guru pap pau pdc pes pes_Latn pes_Thaa pms pnb pol por ppl_Latn prg_Latn pus quc qya qya_Latn rap rif_Latn roh rom ron rue run rus sag sah san_Deva scn sco sgs shs_Latn shy_Latn sin sjn_Latn slv sma sme smo sna snd_Arab som spa sqi srp_Cyrl srp_Latn stq sun swe swg swh tah tam tat tat_Arab tat_Latn tel tet tgk_Cyrl tha tir tlh_Latn tly_Latn tmw_Latn toi_Latn ton tpw_Latn tso tuk tuk_Latn tur tvl tyv tzl tzl_Latn udm uig_Arab uig_Cyrl ukr umb urd uzb_Cyrl uzb_Latn vec vie vie_Hani vol_Latn vro war wln wol wuu xal xho yid yor yue yue_Hans yue_Hant zho zho_Hans zho_Hant zlm_Latn zsm_Latn zul zza
  • model: transformer
  • pre-processing: normalization + SentencePiece (spm32k,spm32k)
  • a sentence initial language token is required in the form of >>id<< (id = valid target language ID)
  • download: opus2m-2020-08-01.zip
  • test set translations: opus2m-2020-08-01.test.txt
  • test set scores: opus2m-2020-08-01.eval.txt

Benchmarks

testset BLEU chr-F
newsdev2014-enghin.eng.hin 5.0 0.288
newsdev2015-enfi-engfin.eng.fin 9.3 0.418
newsdev2016-enro-engron.eng.ron 17.2 0.488
newsdev2016-entr-engtur.eng.tur 8.2 0.402
newsdev2017-enlv-englav.eng.lav 12.9 0.444
newsdev2017-enzh-engzho.eng.zho 17.6 0.170
newsdev2018-enet-engest.eng.est 10.9 0.423
newsdev2019-engu-engguj.eng.guj 5.2 0.284
newsdev2019-enlt-englit.eng.lit 11.0 0.431
newsdiscussdev2015-enfr-engfra.eng.fra 22.6 0.521
newsdiscusstest2015-enfr-engfra.eng.fra 25.9 0.546
newssyscomb2009-engces.eng.ces 10.3 0.394
newssyscomb2009-engdeu.eng.deu 13.3 0.459
newssyscomb2009-engfra.eng.fra 21.5 0.522
newssyscomb2009-enghun.eng.hun 8.1 0.371
newssyscomb2009-engita.eng.ita 22.1 0.540
newssyscomb2009-engspa.eng.spa 23.8 0.531
news-test2008-engces.eng.ces 9.0 0.376
news-test2008-engdeu.eng.deu 14.2 0.451
news-test2008-engfra.eng.fra 19.8 0.500
news-test2008-engspa.eng.spa 22.8 0.518
newstest2009-engces.eng.ces 9.8 0.392
newstest2009-engdeu.eng.deu 13.7 0.454
newstest2009-engfra.eng.fra 20.7 0.514
newstest2009-enghun.eng.hun 8.4 0.370
newstest2009-engita.eng.ita 22.4 0.538
newstest2009-engspa.eng.spa 23.5 0.532
newstest2010-engces.eng.ces 10.0 0.393
newstest2010-engdeu.eng.deu 15.2 0.463
newstest2010-engfra.eng.fra 22.0 0.524
newstest2010-engspa.eng.spa 27.2 0.556
newstest2011-engces.eng.ces 10.8 0.392
newstest2011-engdeu.eng.deu 14.2 0.449
newstest2011-engfra.eng.fra 24.3 0.544
newstest2011-engspa.eng.spa 28.3 0.559
newstest2012-engces.eng.ces 9.9 0.377
newstest2012-engdeu.eng.deu 14.3 0.449
newstest2012-engfra.eng.fra 23.2 0.530
newstest2012-engrus.eng.rus 16.0 0.463
newstest2012-engspa.eng.spa 27.8 0.555
newstest2013-engces.eng.ces 11.0 0.392
newstest2013-engdeu.eng.deu 16.4 0.469
newstest2013-engfra.eng.fra 22.6 0.515
newstest2013-engrus.eng.rus 12.1 0.414
newstest2013-engspa.eng.spa 24.9 0.532
newstest2014-hien-enghin.eng.hin 7.2 0.311
newstest2015-encs-engces.eng.ces 10.9 0.396
newstest2015-ende-engdeu.eng.deu 18.3 0.490
newstest2015-enfi-engfin.eng.fin 10.1 0.421
newstest2015-enru-engrus.eng.rus 14.5 0.445
newstest2016-encs-engces.eng.ces 12.2 0.408
newstest2016-ende-engdeu.eng.deu 21.4 0.517
newstest2016-enfi-engfin.eng.fin 11.2 0.435
newstest2016-enro-engron.eng.ron 16.6 0.472
newstest2016-enru-engrus.eng.rus 13.4 0.435
newstest2016-entr-engtur.eng.tur 8.1 0.385
newstest2017-encs-engces.eng.ces 9.6 0.377
newstest2017-ende-engdeu.eng.deu 17.9 0.482
newstest2017-enfi-engfin.eng.fin 11.8 0.440
newstest2017-enlv-englav.eng.lav 9.6 0.412
newstest2017-enru-engrus.eng.rus 14.1 0.446
newstest2017-entr-engtur.eng.tur 8.0 0.378
newstest2017-enzh-engzho.eng.zho 16.8 0.175
newstest2018-encs-engces.eng.ces 9.8 0.380
newstest2018-ende-engdeu.eng.deu 23.8 0.536
newstest2018-enet-engest.eng.est 11.8 0.433
newstest2018-enfi-engfin.eng.fin 7.8 0.398
newstest2018-enru-engrus.eng.rus 12.2 0.434
newstest2018-entr-engtur.eng.tur 7.5 0.383
newstest2018-enzh-engzho.eng.zho 18.3 0.179
newstest2019-encs-engces.eng.ces 10.7 0.389
newstest2019-ende-engdeu.eng.deu 21.0 0.512
newstest2019-enfi-engfin.eng.fin 10.4 0.420
newstest2019-engu-engguj.eng.guj 5.8 0.297
newstest2019-enlt-englit.eng.lit 8.0 0.388
newstest2019-enru-engrus.eng.rus 13.0 0.415
newstest2019-enzh-engzho.eng.zho 15.0 0.192
newstestB2016-enfi-engfin.eng.fin 9.0 0.414
newstestB2017-enfi-engfin.eng.fin 9.5 0.415
Tatoeba-test.eng-abk.eng.abk 4.2 0.275
Tatoeba-test.eng-ady.eng.ady 0.4 0.006
Tatoeba-test.eng-afh.eng.afh 1.0 0.058
Tatoeba-test.eng-afr.eng.afr 47.0 0.663
Tatoeba-test.eng-akl.eng.akl 2.7 0.080
Tatoeba-test.eng-amh.eng.amh 8.5 0.455
Tatoeba-test.eng-ang.eng.ang 6.2 0.138
Tatoeba-test.eng-ara.eng.ara 6.3 0.325
Tatoeba-test.eng-arg.eng.arg 1.5 0.107
Tatoeba-test.eng-asm.eng.asm 2.1 0.265
Tatoeba-test.eng-ast.eng.ast 15.7 0.393
Tatoeba-test.eng-avk.eng.avk 0.2 0.095
Tatoeba-test.eng-awa.eng.awa 0.1 0.002
Tatoeba-test.eng-aze.eng.aze 19.0 0.500
Tatoeba-test.eng-bak.eng.bak 12.7 0.379
Tatoeba-test.eng-bam.eng.bam 8.3 0.037
Tatoeba-test.eng-bel.eng.bel 13.5 0.396
Tatoeba-test.eng-ben.eng.ben 10.0 0.383
Tatoeba-test.eng-bho.eng.bho 0.1 0.003
Tatoeba-test.eng-bod.eng.bod 0.0 0.147
Tatoeba-test.eng-bre.eng.bre 7.6 0.275
Tatoeba-test.eng-brx.eng.brx 0.8 0.060
Tatoeba-test.eng-bul.eng.bul 32.1 0.542
Tatoeba-test.eng-cat.eng.cat 37.0 0.595
Tatoeba-test.eng-ceb.eng.ceb 9.6 0.409
Tatoeba-test.eng-ces.eng.ces 24.0 0.475
Tatoeba-test.eng-cha.eng.cha 3.9 0.228
Tatoeba-test.eng-che.eng.che 0.7 0.013
Tatoeba-test.eng-chm.eng.chm 2.6 0.212
Tatoeba-test.eng-chr.eng.chr 6.0 0.190
Tatoeba-test.eng-chv.eng.chv 6.5 0.369
Tatoeba-test.eng-cor.eng.cor 0.9 0.086
Tatoeba-test.eng-cos.eng.cos 4.2 0.174
Tatoeba-test.eng-crh.eng.crh 9.9 0.361
Tatoeba-test.eng-csb.eng.csb 3.4 0.230
Tatoeba-test.eng-cym.eng.cym 18.0 0.418
Tatoeba-test.eng-dan.eng.dan 42.5 0.624
Tatoeba-test.eng-deu.eng.deu 25.2 0.505
Tatoeba-test.eng-dsb.eng.dsb 0.9 0.121
Tatoeba-test.eng-dtp.eng.dtp 0.3 0.084
Tatoeba-test.eng-dws.eng.dws 0.2 0.040
Tatoeba-test.eng-egl.eng.egl 0.4 0.085
Tatoeba-test.eng-ell.eng.ell 28.7 0.543
Tatoeba-test.eng-enm.eng.enm 3.3 0.295
Tatoeba-test.eng-epo.eng.epo 33.4 0.570
Tatoeba-test.eng-est.eng.est 30.3 0.545
Tatoeba-test.eng-eus.eng.eus 18.5 0.486
Tatoeba-test.eng-ewe.eng.ewe 6.8 0.272
Tatoeba-test.eng-ext.eng.ext 5.0 0.228
Tatoeba-test.eng-fao.eng.fao 5.2 0.277
Tatoeba-test.eng-fas.eng.fas 6.9 0.265
Tatoeba-test.eng-fij.eng.fij 31.5 0.365
Tatoeba-test.eng-fin.eng.fin 18.5 0.459
Tatoeba-test.eng-fkv.eng.fkv 0.9 0.132
Tatoeba-test.eng-fra.eng.fra 31.5 0.546
Tatoeba-test.eng-frm.eng.frm 0.9 0.128
Tatoeba-test.eng-frr.eng.frr 3.0 0.025
Tatoeba-test.eng-fry.eng.fry 14.4 0.387
Tatoeba-test.eng-ful.eng.ful 0.4 0.061
Tatoeba-test.eng-gcf.eng.gcf 0.3 0.075
Tatoeba-test.eng-gil.eng.gil 47.4 0.706
Tatoeba-test.eng-gla.eng.gla 10.9 0.341
Tatoeba-test.eng-gle.eng.gle 26.8 0.493
Tatoeba-test.eng-glg.eng.glg 32.5 0.565
Tatoeba-test.eng-glv.eng.glv 21.5 0.395
Tatoeba-test.eng-gos.eng.gos 0.3 0.124
Tatoeba-test.eng-got.eng.got 0.2 0.010
Tatoeba-test.eng-grc.eng.grc 0.0 0.005
Tatoeba-test.eng-grn.eng.grn 1.5 0.129
Tatoeba-test.eng-gsw.eng.gsw 0.6 0.106
Tatoeba-test.eng-guj.eng.guj 15.4 0.347
Tatoeba-test.eng-hat.eng.hat 31.1 0.527
Tatoeba-test.eng-hau.eng.hau 6.5 0.385
Tatoeba-test.eng-haw.eng.haw 0.2 0.066
Tatoeba-test.eng-hbs.eng.hbs 28.7 0.531
Tatoeba-test.eng-heb.eng.heb 21.3 0.443
Tatoeba-test.eng-hif.eng.hif 2.8 0.268
Tatoeba-test.eng-hil.eng.hil 12.0 0.463
Tatoeba-test.eng-hin.eng.hin 13.0 0.401
Tatoeba-test.eng-hmn.eng.hmn 0.2 0.073
Tatoeba-test.eng-hoc.eng.hoc 0.2 0.077
Tatoeba-test.eng-hsb.eng.hsb 5.7 0.308
Tatoeba-test.eng-hun.eng.hun 17.1 0.431
Tatoeba-test.eng-hye.eng.hye 15.0 0.378
Tatoeba-test.eng-iba.eng.iba 16.0 0.437
Tatoeba-test.eng-ibo.eng.ibo 2.9 0.221
Tatoeba-test.eng-ido.eng.ido 11.5 0.403
Tatoeba-test.eng-iku.eng.iku 2.3 0.089
Tatoeba-test.eng-ile.eng.ile 4.3 0.282
Tatoeba-test.eng-ilo.eng.ilo 26.4 0.522
Tatoeba-test.eng-ina.eng.ina 20.9 0.493
Tatoeba-test.eng-isl.eng.isl 12.5 0.375
Tatoeba-test.eng-ita.eng.ita 33.9 0.592
Tatoeba-test.eng-izh.eng.izh 4.6 0.050
Tatoeba-test.eng-jav.eng.jav 7.8 0.328
Tatoeba-test.eng-jbo.eng.jbo 0.1 0.123
Tatoeba-test.eng-jdt.eng.jdt 6.4 0.008
Tatoeba-test.eng-jpn.eng.jpn 0.0 0.000
Tatoeba-test.eng-kab.eng.kab 5.9 0.261
Tatoeba-test.eng-kal.eng.kal 13.4 0.382
Tatoeba-test.eng-kan.eng.kan 4.8 0.358
Tatoeba-test.eng-kat.eng.kat 1.8 0.115
Tatoeba-test.eng-kaz.eng.kaz 8.8 0.354
Tatoeba-test.eng-kek.eng.kek 3.7 0.188
Tatoeba-test.eng-kha.eng.kha 0.5 0.094
Tatoeba-test.eng-khm.eng.khm 0.4 0.243
Tatoeba-test.eng-kin.eng.kin 5.2 0.362
Tatoeba-test.eng-kir.eng.kir 17.2 0.416
Tatoeba-test.eng-kjh.eng.kjh 0.6 0.009
Tatoeba-test.eng-kok.eng.kok 5.5 0.005
Tatoeba-test.eng-kom.eng.kom 2.4 0.012
Tatoeba-test.eng-krl.eng.krl 2.0 0.099
Tatoeba-test.eng-ksh.eng.ksh 0.4 0.074
Tatoeba-test.eng-kum.eng.kum 0.9 0.007
Tatoeba-test.eng-kur.eng.kur 9.1 0.174
Tatoeba-test.eng-lad.eng.lad 1.2 0.154
Tatoeba-test.eng-lah.eng.lah 0.1 0.001
Tatoeba-test.eng-lao.eng.lao 0.6 0.426
Tatoeba-test.eng-lat.eng.lat 8.2 0.366
Tatoeba-test.eng-lav.eng.lav 20.4 0.475
Tatoeba-test.eng-ldn.eng.ldn 0.3 0.059
Tatoeba-test.eng-lfn.eng.lfn 0.5 0.104
Tatoeba-test.eng-lij.eng.lij 0.2 0.094
Tatoeba-test.eng-lin.eng.lin 1.2 0.276
Tatoeba-test.eng-lit.eng.lit 17.4 0.488
Tatoeba-test.eng-liv.eng.liv 0.3 0.039
Tatoeba-test.eng-lkt.eng.lkt 0.3 0.041
Tatoeba-test.eng-lld.eng.lld 0.1 0.083
Tatoeba-test.eng-lmo.eng.lmo 1.4 0.154
Tatoeba-test.eng-ltz.eng.ltz 19.1 0.395
Tatoeba-test.eng-lug.eng.lug 4.2 0.382
Tatoeba-test.eng-mad.eng.mad 2.1 0.075
Tatoeba-test.eng-mah.eng.mah 9.5 0.331
Tatoeba-test.eng-mai.eng.mai 9.3 0.372
Tatoeba-test.eng-mal.eng.mal 8.3 0.437
Tatoeba-test.eng-mar.eng.mar 13.5 0.410
Tatoeba-test.eng-mdf.eng.mdf 2.3 0.008
Tatoeba-test.eng-mfe.eng.mfe 83.6 0.905
Tatoeba-test.eng-mic.eng.mic 7.6 0.214
Tatoeba-test.eng-mkd.eng.mkd 31.8 0.540
Tatoeba-test.eng-mlg.eng.mlg 31.3 0.464
Tatoeba-test.eng-mlt.eng.mlt 11.7 0.427
Tatoeba-test.eng-mnw.eng.mnw 0.1 0.000
Tatoeba-test.eng-moh.eng.moh 0.6 0.067
Tatoeba-test.eng-mon.eng.mon 8.5 0.323
Tatoeba-test.eng-mri.eng.mri 8.5 0.320
Tatoeba-test.eng-msa.eng.msa 24.5 0.498
Tatoeba-test.eng.multi 22.4 0.451
Tatoeba-test.eng-mwl.eng.mwl 3.8 0.169
Tatoeba-test.eng-mya.eng.mya 0.2 0.123
Tatoeba-test.eng-myv.eng.myv 1.1 0.014
Tatoeba-test.eng-nau.eng.nau 0.6 0.109
Tatoeba-test.eng-nav.eng.nav 1.8 0.149
Tatoeba-test.eng-nds.eng.nds 11.3 0.365
Tatoeba-test.eng-nep.eng.nep 0.5 0.004
Tatoeba-test.eng-niu.eng.niu 34.4 0.501
Tatoeba-test.eng-nld.eng.nld 37.6 0.598
Tatoeba-test.eng-nog.eng.nog 0.2 0.010
Tatoeba-test.eng-non.eng.non 0.2 0.096
Tatoeba-test.eng-nor.eng.nor 36.3 0.577
Tatoeba-test.eng-nov.eng.nov 0.9 0.180
Tatoeba-test.eng-nya.eng.nya 9.8 0.524
Tatoeba-test.eng-oci.eng.oci 6.3 0.288
Tatoeba-test.eng-ori.eng.ori 5.3 0.273
Tatoeba-test.eng-orv.eng.orv 0.2 0.007
Tatoeba-test.eng-oss.eng.oss 3.0 0.230
Tatoeba-test.eng-ota.eng.ota 0.2 0.053
Tatoeba-test.eng-pag.eng.pag 20.2 0.513
Tatoeba-test.eng-pan.eng.pan 6.4 0.301
Tatoeba-test.eng-pap.eng.pap 44.7 0.624
Tatoeba-test.eng-pau.eng.pau 0.8 0.098
Tatoeba-test.eng-pdc.eng.pdc 2.9 0.143
Tatoeba-test.eng-pms.eng.pms 0.6 0.124
Tatoeba-test.eng-pol.eng.pol 22.7 0.500
Tatoeba-test.eng-por.eng.por 31.6 0.570
Tatoeba-test.eng-ppl.eng.ppl 0.5 0.085
Tatoeba-test.eng-prg.eng.prg 0.1 0.078
Tatoeba-test.eng-pus.eng.pus 0.9 0.137
Tatoeba-test.eng-quc.eng.quc 2.7 0.255
Tatoeba-test.eng-qya.eng.qya 0.4 0.084
Tatoeba-test.eng-rap.eng.rap 1.9 0.050
Tatoeba-test.eng-rif.eng.rif 1.3 0.102
Tatoeba-test.eng-roh.eng.roh 1.4 0.169
Tatoeba-test.eng-rom.eng.rom 7.8 0.329
Tatoeba-test.eng-ron.eng.ron 27.0 0.530
Tatoeba-test.eng-rue.eng.rue 0.1 0.009
Tatoeba-test.eng-run.eng.run 9.8 0.434
Tatoeba-test.eng-rus.eng.rus 22.2 0.465
Tatoeba-test.eng-sag.eng.sag 4.8 0.155
Tatoeba-test.eng-sah.eng.sah 0.2 0.007
Tatoeba-test.eng-san.eng.san 1.7 0.143
Tatoeba-test.eng-scn.eng.scn 1.5 0.083
Tatoeba-test.eng-sco.eng.sco 30.3 0.514
Tatoeba-test.eng-sgs.eng.sgs 1.6 0.104
Tatoeba-test.eng-shs.eng.shs 0.7 0.049
Tatoeba-test.eng-shy.eng.shy 0.6 0.064
Tatoeba-test.eng-sin.eng.sin 5.4 0.317
Tatoeba-test.eng-sjn.eng.sjn 0.3 0.074
Tatoeba-test.eng-slv.eng.slv 12.8 0.313
Tatoeba-test.eng-sma.eng.sma 0.8 0.063
Tatoeba-test.eng-sme.eng.sme 13.2 0.290
Tatoeba-test.eng-smo.eng.smo 12.1 0.416
Tatoeba-test.eng-sna.eng.sna 27.1 0.533
Tatoeba-test.eng-snd.eng.snd 6.0 0.359
Tatoeba-test.eng-som.eng.som 16.0 0.274
Tatoeba-test.eng-spa.eng.spa 36.7 0.603
Tatoeba-test.eng-sqi.eng.sqi 32.3 0.573
Tatoeba-test.eng-stq.eng.stq 0.6 0.198
Tatoeba-test.eng-sun.eng.sun 39.0 0.447
Tatoeba-test.eng-swa.eng.swa 1.1 0.109
Tatoeba-test.eng-swe.eng.swe 42.7 0.614
Tatoeba-test.eng-swg.eng.swg 0.6 0.118
Tatoeba-test.eng-tah.eng.tah 12.4 0.294
Tatoeba-test.eng-tam.eng.tam 5.0 0.404
Tatoeba-test.eng-tat.eng.tat 9.9 0.326
Tatoeba-test.eng-tel.eng.tel 4.7 0.326
Tatoeba-test.eng-tet.eng.tet 0.7 0.100
Tatoeba-test.eng-tgk.eng.tgk 5.5 0.304
Tatoeba-test.eng-tha.eng.tha 2.2 0.456
Tatoeba-test.eng-tir.eng.tir 1.5 0.197
Tatoeba-test.eng-tlh.eng.tlh 0.0 0.032
Tatoeba-test.eng-tly.eng.tly 0.3 0.061
Tatoeba-test.eng-toi.eng.toi 8.3 0.219
Tatoeba-test.eng-ton.eng.ton 32.7 0.619
Tatoeba-test.eng-tpw.eng.tpw 1.4 0.136
Tatoeba-test.eng-tso.eng.tso 9.6 0.465
Tatoeba-test.eng-tuk.eng.tuk 9.4 0.383
Tatoeba-test.eng-tur.eng.tur 24.1 0.542
Tatoeba-test.eng-tvl.eng.tvl 8.9 0.398
Tatoeba-test.eng-tyv.eng.tyv 10.4 0.249
Tatoeba-test.eng-tzl.eng.tzl 0.2 0.098
Tatoeba-test.eng-udm.eng.udm 6.5 0.212
Tatoeba-test.eng-uig.eng.uig 2.1 0.266
Tatoeba-test.eng-ukr.eng.ukr 24.3 0.479
Tatoeba-test.eng-umb.eng.umb 4.4 0.274
Tatoeba-test.eng-urd.eng.urd 8.6 0.344
Tatoeba-test.eng-uzb.eng.uzb 6.9 0.343
Tatoeba-test.eng-vec.eng.vec 1.0 0.094
Tatoeba-test.eng-vie.eng.vie 23.2 0.420
Tatoeba-test.eng-vol.eng.vol 0.3 0.086
Tatoeba-test.eng-war.eng.war 11.4 0.415
Tatoeba-test.eng-wln.eng.wln 8.4 0.218
Tatoeba-test.eng-wol.eng.wol 11.5 0.252
Tatoeba-test.eng-xal.eng.xal 0.1 0.007
Tatoeba-test.eng-xho.eng.xho 19.5 0.552
Tatoeba-test.eng-yid.eng.yid 4.0 0.256
Tatoeba-test.eng-yor.eng.yor 8.8 0.247
Tatoeba-test.eng-zho.eng.zho 21.8 0.192
Tatoeba-test.eng-zul.eng.zul 34.3 0.655
Tatoeba-test.eng-zza.eng.zza 0.5 0.080