Hugging Face's logo

Helsinki-NLP
/
opus-mt-en-ine

Translation
Transformers
PyTorch
TensorFlow
marian
text2text-generation
Inference Endpoints
Model card Files Community
2
system HF staff commited on
Commit
0ccad4b
·
1 Parent(s): c00437f

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +253 -0
README.md ADDED
@@ -0,0 +1,253 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: en
3
+ tags:
4
+ - translation
5
+
6
+ license: apache-2.0
7
+ ---
8
+
9
+ ### eng-ine
10
+
11
+ * source group: English
12
+ * target group: Indo-European languages
13
+ * OPUS readme: [eng-ine](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/eng-ine/README.md)
14
+
15
+ * model: transformer
16
+ * source language(s): eng
17
+ * target language(s): afr aln ang_Latn arg asm ast awa bel bel_Latn ben bho bos_Latn bre bul bul_Latn cat ces cor cos csb_Latn cym dan deu dsb egl ell enm_Latn ext fao fra frm_Latn frr fry gcf_Latn gla gle glg glv gom gos got_Goth grc_Grek gsw guj hat hif_Latn hin hrv hsb hye ind isl ita jdt_Cyrl ksh kur_Arab kur_Latn lad lad_Latn lat_Latn lav lij lit lld_Latn lmo ltg ltz mai mar max_Latn mfe min mkd mwl nds nld nno nob nob_Hebr non_Latn npi oci ori orv_Cyrl oss pan_Guru pap pdc pes pes_Latn pes_Thaa pms pnb pol por prg_Latn pus roh rom ron rue rus san_Deva scn sco sgs sin slv snd_Arab spa sqi srp_Cyrl srp_Latn stq swe swg tgk_Cyrl tly_Latn tmw_Latn ukr urd vec wln yid zlm_Latn zsm_Latn zza
18
+ * model: transformer
19
+ * pre-processing: normalization + SentencePiece (spm32k,spm32k)
20
+ * a sentence initial language token is required in the form of `>>id<<` (id = valid target language ID)
21
+ * download original weights: [opus2m-2020-08-01.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/eng-ine/opus2m-2020-08-01.zip)
22
+ * test set translations: [opus2m-2020-08-01.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/eng-ine/opus2m-2020-08-01.test.txt)
23
+ * test set scores: [opus2m-2020-08-01.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/eng-ine/opus2m-2020-08-01.eval.txt)
24
+
25
+ ## Benchmarks
26
+
27
+ | testset | BLEU | chr-F |
28
+ |-----------------------|-------|-------|
29
+ | newsdev2014-enghin.eng.hin | 6.2 | 0.317 |
30
+ | newsdev2016-enro-engron.eng.ron | 22.1 | 0.525 |
31
+ | newsdev2017-enlv-englav.eng.lav | 17.4 | 0.486 |
32
+ | newsdev2019-engu-engguj.eng.guj | 6.5 | 0.303 |
33
+ | newsdev2019-enlt-englit.eng.lit | 14.9 | 0.476 |
34
+ | newsdiscussdev2015-enfr-engfra.eng.fra | 26.4 | 0.547 |
35
+ | newsdiscusstest2015-enfr-engfra.eng.fra | 30.0 | 0.575 |
36
+ | newssyscomb2009-engces.eng.ces | 14.7 | 0.442 |
37
+ | newssyscomb2009-engdeu.eng.deu | 16.7 | 0.487 |
38
+ | newssyscomb2009-engfra.eng.fra | 24.8 | 0.547 |
39
+ | newssyscomb2009-engita.eng.ita | 25.2 | 0.562 |
40
+ | newssyscomb2009-engspa.eng.spa | 27.0 | 0.554 |
41
+ | news-test2008-engces.eng.ces | 13.0 | 0.417 |
42
+ | news-test2008-engdeu.eng.deu | 17.4 | 0.480 |
43
+ | news-test2008-engfra.eng.fra | 22.3 | 0.519 |
44
+ | news-test2008-engspa.eng.spa | 24.9 | 0.532 |
45
+ | newstest2009-engces.eng.ces | 13.6 | 0.432 |
46
+ | newstest2009-engdeu.eng.deu | 16.6 | 0.482 |
47
+ | newstest2009-engfra.eng.fra | 23.5 | 0.535 |
48
+ | newstest2009-engita.eng.ita | 25.5 | 0.561 |
49
+ | newstest2009-engspa.eng.spa | 26.3 | 0.551 |
50
+ | newstest2010-engces.eng.ces | 14.2 | 0.436 |
51
+ | newstest2010-engdeu.eng.deu | 18.3 | 0.492 |
52
+ | newstest2010-engfra.eng.fra | 25.7 | 0.550 |
53
+ | newstest2010-engspa.eng.spa | 30.5 | 0.578 |
54
+ | newstest2011-engces.eng.ces | 15.1 | 0.439 |
55
+ | newstest2011-engdeu.eng.deu | 17.1 | 0.478 |
56
+ | newstest2011-engfra.eng.fra | 28.0 | 0.569 |
57
+ | newstest2011-engspa.eng.spa | 31.9 | 0.580 |
58
+ | newstest2012-engces.eng.ces | 13.6 | 0.418 |
59
+ | newstest2012-engdeu.eng.deu | 17.0 | 0.475 |
60
+ | newstest2012-engfra.eng.fra | 26.1 | 0.553 |
61
+ | newstest2012-engrus.eng.rus | 21.4 | 0.506 |
62
+ | newstest2012-engspa.eng.spa | 31.4 | 0.577 |
63
+ | newstest2013-engces.eng.ces | 15.3 | 0.438 |
64
+ | newstest2013-engdeu.eng.deu | 20.3 | 0.501 |
65
+ | newstest2013-engfra.eng.fra | 26.0 | 0.540 |
66
+ | newstest2013-engrus.eng.rus | 16.1 | 0.449 |
67
+ | newstest2013-engspa.eng.spa | 28.6 | 0.555 |
68
+ | newstest2014-hien-enghin.eng.hin | 9.5 | 0.344 |
69
+ | newstest2015-encs-engces.eng.ces | 14.8 | 0.440 |
70
+ | newstest2015-ende-engdeu.eng.deu | 22.6 | 0.523 |
71
+ | newstest2015-enru-engrus.eng.rus | 18.8 | 0.483 |
72
+ | newstest2016-encs-engces.eng.ces | 16.8 | 0.457 |
73
+ | newstest2016-ende-engdeu.eng.deu | 26.2 | 0.555 |
74
+ | newstest2016-enro-engron.eng.ron | 21.2 | 0.510 |
75
+ | newstest2016-enru-engrus.eng.rus | 17.6 | 0.471 |
76
+ | newstest2017-encs-engces.eng.ces | 13.6 | 0.421 |
77
+ | newstest2017-ende-engdeu.eng.deu | 21.5 | 0.516 |
78
+ | newstest2017-enlv-englav.eng.lav | 13.0 | 0.452 |
79
+ | newstest2017-enru-engrus.eng.rus | 18.7 | 0.486 |
80
+ | newstest2018-encs-engces.eng.ces | 13.5 | 0.425 |
81
+ | newstest2018-ende-engdeu.eng.deu | 29.8 | 0.581 |
82
+ | newstest2018-enru-engrus.eng.rus | 16.1 | 0.472 |
83
+ | newstest2019-encs-engces.eng.ces | 14.8 | 0.435 |
84
+ | newstest2019-ende-engdeu.eng.deu | 26.6 | 0.554 |
85
+ | newstest2019-engu-engguj.eng.guj | 6.9 | 0.313 |
86
+ | newstest2019-enlt-englit.eng.lit | 10.6 | 0.429 |
87
+ | newstest2019-enru-engrus.eng.rus | 17.5 | 0.452 |
88
+ | Tatoeba-test.eng-afr.eng.afr | 52.1 | 0.708 |
89
+ | Tatoeba-test.eng-ang.eng.ang | 5.1 | 0.131 |
90
+ | Tatoeba-test.eng-arg.eng.arg | 1.2 | 0.099 |
91
+ | Tatoeba-test.eng-asm.eng.asm | 2.9 | 0.259 |
92
+ | Tatoeba-test.eng-ast.eng.ast | 14.1 | 0.408 |
93
+ | Tatoeba-test.eng-awa.eng.awa | 0.3 | 0.002 |
94
+ | Tatoeba-test.eng-bel.eng.bel | 18.1 | 0.450 |
95
+ | Tatoeba-test.eng-ben.eng.ben | 13.5 | 0.432 |
96
+ | Tatoeba-test.eng-bho.eng.bho | 0.3 | 0.003 |
97
+ | Tatoeba-test.eng-bre.eng.bre | 10.4 | 0.318 |
98
+ | Tatoeba-test.eng-bul.eng.bul | 38.7 | 0.592 |
99
+ | Tatoeba-test.eng-cat.eng.cat | 42.0 | 0.633 |
100
+ | Tatoeba-test.eng-ces.eng.ces | 32.3 | 0.546 |
101
+ | Tatoeba-test.eng-cor.eng.cor | 0.5 | 0.079 |
102
+ | Tatoeba-test.eng-cos.eng.cos | 3.1 | 0.148 |
103
+ | Tatoeba-test.eng-csb.eng.csb | 1.4 | 0.216 |
104
+ | Tatoeba-test.eng-cym.eng.cym | 22.4 | 0.470 |
105
+ | Tatoeba-test.eng-dan.eng.dan | 49.7 | 0.671 |
106
+ | Tatoeba-test.eng-deu.eng.deu | 31.7 | 0.554 |
107
+ | Tatoeba-test.eng-dsb.eng.dsb | 1.1 | 0.139 |
108
+ | Tatoeba-test.eng-egl.eng.egl | 0.9 | 0.089 |
109
+ | Tatoeba-test.eng-ell.eng.ell | 42.7 | 0.640 |
110
+ | Tatoeba-test.eng-enm.eng.enm | 3.5 | 0.259 |
111
+ | Tatoeba-test.eng-ext.eng.ext | 6.4 | 0.235 |
112
+ | Tatoeba-test.eng-fao.eng.fao | 6.6 | 0.285 |
113
+ | Tatoeba-test.eng-fas.eng.fas | 5.7 | 0.257 |
114
+ | Tatoeba-test.eng-fra.eng.fra | 38.4 | 0.595 |
115
+ | Tatoeba-test.eng-frm.eng.frm | 0.9 | 0.149 |
116
+ | Tatoeba-test.eng-frr.eng.frr | 8.4 | 0.145 |
117
+ | Tatoeba-test.eng-fry.eng.fry | 16.5 | 0.411 |
118
+ | Tatoeba-test.eng-gcf.eng.gcf | 0.6 | 0.098 |
119
+ | Tatoeba-test.eng-gla.eng.gla | 11.6 | 0.361 |
120
+ | Tatoeba-test.eng-gle.eng.gle | 32.5 | 0.546 |
121
+ | Tatoeba-test.eng-glg.eng.glg | 38.4 | 0.602 |
122
+ | Tatoeba-test.eng-glv.eng.glv | 23.1 | 0.418 |
123
+ | Tatoeba-test.eng-gos.eng.gos | 0.7 | 0.137 |
124
+ | Tatoeba-test.eng-got.eng.got | 0.2 | 0.010 |
125
+ | Tatoeba-test.eng-grc.eng.grc | 0.0 | 0.005 |
126
+ | Tatoeba-test.eng-gsw.eng.gsw | 0.9 | 0.108 |
127
+ | Tatoeba-test.eng-guj.eng.guj | 20.8 | 0.391 |
128
+ | Tatoeba-test.eng-hat.eng.hat | 34.0 | 0.537 |
129
+ | Tatoeba-test.eng-hbs.eng.hbs | 33.7 | 0.567 |
130
+ | Tatoeba-test.eng-hif.eng.hif | 2.8 | 0.269 |
131
+ | Tatoeba-test.eng-hin.eng.hin | 15.6 | 0.437 |
132
+ | Tatoeba-test.eng-hsb.eng.hsb | 5.4 | 0.320 |
133
+ | Tatoeba-test.eng-hye.eng.hye | 17.4 | 0.426 |
134
+ | Tatoeba-test.eng-isl.eng.isl | 17.4 | 0.436 |
135
+ | Tatoeba-test.eng-ita.eng.ita | 40.4 | 0.636 |
136
+ | Tatoeba-test.eng-jdt.eng.jdt | 6.4 | 0.008 |
137
+ | Tatoeba-test.eng-kok.eng.kok | 6.6 | 0.005 |
138
+ | Tatoeba-test.eng-ksh.eng.ksh | 0.8 | 0.123 |
139
+ | Tatoeba-test.eng-kur.eng.kur | 10.2 | 0.209 |
140
+ | Tatoeba-test.eng-lad.eng.lad | 0.8 | 0.163 |
141
+ | Tatoeba-test.eng-lah.eng.lah | 0.2 | 0.001 |
142
+ | Tatoeba-test.eng-lat.eng.lat | 9.4 | 0.372 |
143
+ | Tatoeba-test.eng-lav.eng.lav | 30.3 | 0.559 |
144
+ | Tatoeba-test.eng-lij.eng.lij | 1.0 | 0.130 |
145
+ | Tatoeba-test.eng-lit.eng.lit | 25.3 | 0.560 |
146
+ | Tatoeba-test.eng-lld.eng.lld | 0.4 | 0.139 |
147
+ | Tatoeba-test.eng-lmo.eng.lmo | 0.6 | 0.108 |
148
+ | Tatoeba-test.eng-ltz.eng.ltz | 18.1 | 0.388 |
149
+ | Tatoeba-test.eng-mai.eng.mai | 17.2 | 0.464 |
150
+ | Tatoeba-test.eng-mar.eng.mar | 18.0 | 0.451 |
151
+ | Tatoeba-test.eng-mfe.eng.mfe | 81.0 | 0.899 |
152
+ | Tatoeba-test.eng-mkd.eng.mkd | 37.6 | 0.587 |
153
+ | Tatoeba-test.eng-msa.eng.msa | 27.7 | 0.519 |
154
+ | Tatoeba-test.eng.multi | 32.6 | 0.539 |
155
+ | Tatoeba-test.eng-mwl.eng.mwl | 3.8 | 0.134 |
156
+ | Tatoeba-test.eng-nds.eng.nds | 14.3 | 0.401 |
157
+ | Tatoeba-test.eng-nep.eng.nep | 0.5 | 0.002 |
158
+ | Tatoeba-test.eng-nld.eng.nld | 44.0 | 0.642 |
159
+ | Tatoeba-test.eng-non.eng.non | 0.7 | 0.118 |
160
+ | Tatoeba-test.eng-nor.eng.nor | 42.7 | 0.623 |
161
+ | Tatoeba-test.eng-oci.eng.oci | 7.2 | 0.295 |
162
+ | Tatoeba-test.eng-ori.eng.ori | 2.7 | 0.257 |
163
+ | Tatoeba-test.eng-orv.eng.orv | 0.2 | 0.008 |
164
+ | Tatoeba-test.eng-oss.eng.oss | 2.9 | 0.264 |
165
+ | Tatoeba-test.eng-pan.eng.pan | 7.4 | 0.337 |
166
+ | Tatoeba-test.eng-pap.eng.pap | 48.5 | 0.656 |
167
+ | Tatoeba-test.eng-pdc.eng.pdc | 1.8 | 0.145 |
168
+ | Tatoeba-test.eng-pms.eng.pms | 0.7 | 0.136 |
169
+ | Tatoeba-test.eng-pol.eng.pol | 31.1 | 0.563 |
170
+ | Tatoeba-test.eng-por.eng.por | 37.0 | 0.605 |
171
+ | Tatoeba-test.eng-prg.eng.prg | 0.2 | 0.100 |
172
+ | Tatoeba-test.eng-pus.eng.pus | 1.0 | 0.134 |
173
+ | Tatoeba-test.eng-roh.eng.roh | 2.3 | 0.236 |
174
+ | Tatoeba-test.eng-rom.eng.rom | 7.8 | 0.340 |
175
+ | Tatoeba-test.eng-ron.eng.ron | 34.3 | 0.585 |
176
+ | Tatoeba-test.eng-rue.eng.rue | 0.2 | 0.010 |
177
+ | Tatoeba-test.eng-rus.eng.rus | 29.6 | 0.526 |
178
+ | Tatoeba-test.eng-san.eng.san | 2.4 | 0.125 |
179
+ | Tatoeba-test.eng-scn.eng.scn | 1.6 | 0.079 |
180
+ | Tatoeba-test.eng-sco.eng.sco | 33.6 | 0.562 |
181
+ | Tatoeba-test.eng-sgs.eng.sgs | 3.4 | 0.114 |
182
+ | Tatoeba-test.eng-sin.eng.sin | 9.2 | 0.349 |
183
+ | Tatoeba-test.eng-slv.eng.slv | 15.6 | 0.334 |
184
+ | Tatoeba-test.eng-snd.eng.snd | 9.1 | 0.324 |
185
+ | Tatoeba-test.eng-spa.eng.spa | 43.4 | 0.645 |
186
+ | Tatoeba-test.eng-sqi.eng.sqi | 39.0 | 0.621 |
187
+ | Tatoeba-test.eng-stq.eng.stq | 10.8 | 0.373 |
188
+ | Tatoeba-test.eng-swe.eng.swe | 49.9 | 0.663 |
189
+ | Tatoeba-test.eng-swg.eng.swg | 0.7 | 0.137 |
190
+ | Tatoeba-test.eng-tgk.eng.tgk | 6.4 | 0.346 |
191
+ | Tatoeba-test.eng-tly.eng.tly | 0.5 | 0.055 |
192
+ | Tatoeba-test.eng-ukr.eng.ukr | 31.4 | 0.536 |
193
+ | Tatoeba-test.eng-urd.eng.urd | 11.1 | 0.389 |
194
+ | Tatoeba-test.eng-vec.eng.vec | 1.3 | 0.110 |
195
+ | Tatoeba-test.eng-wln.eng.wln | 6.8 | 0.233 |
196
+ | Tatoeba-test.eng-yid.eng.yid | 5.8 | 0.295 |
197
+ | Tatoeba-test.eng-zza.eng.zza | 0.8 | 0.086 |
198
+
199
+
200
+ ### System Info:
201
+ - hf_name: eng-ine
202
+
203
+ - source_languages: eng
204
+
205
+ - target_languages: ine
206
+
207
+ - opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/eng-ine/README.md
208
+
209
+ - original_repo: Tatoeba-Challenge
210
+
211
+ - tags: ['translation']
212
+
213
+ - prepro: normalization + SentencePiece (spm32k,spm32k)
214
+
215
+ - url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/eng-ine/opus2m-2020-08-01.zip
216
+
217
+ - url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/eng-ine/opus2m-2020-08-01.test.txt
218
+
219
+ - src_alpha3: eng
220
+
221
+ - tgt_alpha3: ine
222
+
223
+ - short_pair: en-ine
224
+
225
+ - chrF2_score: 0.539
226
+
227
+ - bleu: 32.6
228
+
229
+ - brevity_penalty: 0.973
230
+
231
+ - ref_len: 68664.0
232
+
233
+ - src_name: English
234
+
235
+ - tgt_name: Indo-European languages
236
+
237
+ - train_date: 2020-08-01
238
+
239
+ - src_alpha2: en
240
+
241
+ - tgt_alpha2: ine
242
+
243
+ - prefer_old: False
244
+
245
+ - long_pair: eng-ine
246
+
247
+ - helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535
248
+
249
+ - transformers_git_sha: 46e9f53347bbe9e989f0335f98465f30886d8173
250
+
251
+ - port_machine: brutasse
252
+
253
+ - port_time: 2020-08-18-01:48