tiedeman commited on
Commit
53f9130
1 Parent(s): 23f276a

Initial commit

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ *.spm filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,1696 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ language:
4
+ - abi
5
+ - acd
6
+ - ade
7
+ - adj
8
+ - ak
9
+ - akp
10
+ - ann
11
+ - anv
12
+ - atg
13
+ - avn
14
+ - bas
15
+ - bav
16
+ - bba
17
+ - beh
18
+ - bem
19
+ - bfd
20
+ - bfo
21
+ - bim
22
+ - biv
23
+ - bkv
24
+ - blh
25
+ - bmq
26
+ - bmv
27
+ - bom
28
+ - bov
29
+ - box
30
+ - bqj
31
+ - bss
32
+ - btt
33
+ - bud
34
+ - bwu
35
+ - cce
36
+ - cjk
37
+ - cko
38
+ - cme
39
+ - csk
40
+ - cwe
41
+ - cwt
42
+ - dag
43
+ - de
44
+ - dga
45
+ - dgi
46
+ - dig
47
+ - dop
48
+ - dug
49
+ - dyi
50
+ - dyo
51
+ - ee
52
+ - efi
53
+ - en
54
+ - es
55
+ - fal
56
+ - ff
57
+ - fon
58
+ - fr
59
+ - gej
60
+ - gkn
61
+ - gng
62
+ - gog
63
+ - gud
64
+ - gur
65
+ - guw
66
+ - gux
67
+ - gwr
68
+ - hag
69
+ - hay
70
+ - heh
71
+ - hz
72
+ - ife
73
+ - ig
74
+ - iri
75
+ - izr
76
+ - jbu
77
+ - jmc
78
+ - kam
79
+ - kbp
80
+ - kdc
81
+ - kdl
82
+ - kdn
83
+ - ken
84
+ - keu
85
+ - kez
86
+ - kg
87
+ - ki
88
+ - kia
89
+ - kj
90
+ - kki
91
+ - kkj
92
+ - kma
93
+ - kmb
94
+ - ksb
95
+ - ktj
96
+ - kub
97
+ - kus
98
+ - kyf
99
+ - las
100
+ - lee
101
+ - lef
102
+ - lem
103
+ - lg
104
+ - lia
105
+ - lip
106
+ - ln
107
+ - lob
108
+ - lon
109
+ - lua
110
+ - luy
111
+ - maw
112
+ - mcp
113
+ - mcu
114
+ - mda
115
+ - mfq
116
+ - mgo
117
+ - mnf
118
+ - mnh
119
+ - mor
120
+ - mos
121
+ - muh
122
+ - myk
123
+ - myx
124
+ - mzk
125
+ - mzm
126
+ - mzw
127
+ - ncu
128
+ - nd
129
+ - ndz
130
+ - nfr
131
+ - ng
132
+ - nhu
133
+ - nim
134
+ - nin
135
+ - nmz
136
+ - nnb
137
+ - nnh
138
+ - nnw
139
+ - nr
140
+ - nso
141
+ - ntm
142
+ - ntr
143
+ - nuj
144
+ - nwb
145
+ - ny
146
+ - nyf
147
+ - nyn
148
+ - nyo
149
+ - nyy
150
+ - nzi
151
+ - oku
152
+ - old
153
+ - ozm
154
+ - pai
155
+ - pbl
156
+ - pkb
157
+ - pt
158
+ - rim
159
+ - rn
160
+ - rw
161
+ - seh
162
+ - sg
163
+ - sig
164
+ - sil
165
+ - sld
166
+ - sn
167
+ - snw
168
+ - soy
169
+ - spp
170
+ - ss
171
+ - st
172
+ - suk
173
+ - sw
174
+ - sxb
175
+ - tbz
176
+ - tem
177
+ - thk
178
+ - tik
179
+ - tlj
180
+ - tn
181
+ - toh
182
+ - toi
183
+ - tpm
184
+ - ts
185
+ - tsw
186
+ - tum
187
+ - tw
188
+ - umb
189
+ - vag
190
+ - ve
191
+ - vmw
192
+ - vun
193
+ - wmw
194
+ - wo
195
+ - wob
196
+ - xh
197
+ - xog
198
+ - xon
199
+ - xrb
200
+ - xsm
201
+ - xuo
202
+ - yam
203
+ - yaz
204
+ - yo
205
+ - zu
206
+
207
+ tags:
208
+ - translation
209
+ - opus-mt-tc-bible
210
+
211
+ license: apache-2.0
212
+ model-index:
213
+ - name: opus-mt-tc-bible-big-alv-deu_eng_fra_por_spa
214
+ results:
215
+ - task:
216
+ name: Translation bem-eng
217
+ type: translation
218
+ args: bem-eng
219
+ dataset:
220
+ name: flores200-devtest
221
+ type: flores200-devtest
222
+ args: bem-eng
223
+ metrics:
224
+ - name: BLEU
225
+ type: bleu
226
+ value: 13.1
227
+ - name: chr-F
228
+ type: chrf
229
+ value: 0.37071
230
+ - task:
231
+ name: Translation ibo-eng
232
+ type: translation
233
+ args: ibo-eng
234
+ dataset:
235
+ name: flores200-devtest
236
+ type: flores200-devtest
237
+ args: ibo-eng
238
+ metrics:
239
+ - name: BLEU
240
+ type: bleu
241
+ value: 14.6
242
+ - name: chr-F
243
+ type: chrf
244
+ value: 0.38994
245
+ - task:
246
+ name: Translation kin-eng
247
+ type: translation
248
+ args: kin-eng
249
+ dataset:
250
+ name: flores200-devtest
251
+ type: flores200-devtest
252
+ args: kin-eng
253
+ metrics:
254
+ - name: BLEU
255
+ type: bleu
256
+ value: 18.1
257
+ - name: chr-F
258
+ type: chrf
259
+ value: 0.41964
260
+ - task:
261
+ name: Translation kin-fra
262
+ type: translation
263
+ args: kin-fra
264
+ dataset:
265
+ name: flores200-devtest
266
+ type: flores200-devtest
267
+ args: kin-fra
268
+ metrics:
269
+ - name: BLEU
270
+ type: bleu
271
+ value: 10.7
272
+ - name: chr-F
273
+ type: chrf
274
+ value: 0.34887
275
+ - task:
276
+ name: Translation kon-eng
277
+ type: translation
278
+ args: kon-eng
279
+ dataset:
280
+ name: flores200-devtest
281
+ type: flores200-devtest
282
+ args: kon-eng
283
+ metrics:
284
+ - name: BLEU
285
+ type: bleu
286
+ value: 11.3
287
+ - name: chr-F
288
+ type: chrf
289
+ value: 0.34262
290
+ - task:
291
+ name: Translation lin-eng
292
+ type: translation
293
+ args: lin-eng
294
+ dataset:
295
+ name: flores200-devtest
296
+ type: flores200-devtest
297
+ args: lin-eng
298
+ metrics:
299
+ - name: BLEU
300
+ type: bleu
301
+ value: 14.2
302
+ - name: chr-F
303
+ type: chrf
304
+ value: 0.37728
305
+ - task:
306
+ name: Translation lin-fra
307
+ type: translation
308
+ args: lin-fra
309
+ dataset:
310
+ name: flores200-devtest
311
+ type: flores200-devtest
312
+ args: lin-fra
313
+ metrics:
314
+ - name: BLEU
315
+ type: bleu
316
+ value: 11.0
317
+ - name: chr-F
318
+ type: chrf
319
+ value: 0.35052
320
+ - task:
321
+ name: Translation lug-eng
322
+ type: translation
323
+ args: lug-eng
324
+ dataset:
325
+ name: flores200-devtest
326
+ type: flores200-devtest
327
+ args: lug-eng
328
+ metrics:
329
+ - name: BLEU
330
+ type: bleu
331
+ value: 10.6
332
+ - name: chr-F
333
+ type: chrf
334
+ value: 0.31805
335
+ - task:
336
+ name: Translation nso-eng
337
+ type: translation
338
+ args: nso-eng
339
+ dataset:
340
+ name: flores200-devtest
341
+ type: flores200-devtest
342
+ args: nso-eng
343
+ metrics:
344
+ - name: BLEU
345
+ type: bleu
346
+ value: 22.3
347
+ - name: chr-F
348
+ type: chrf
349
+ value: 0.45662
350
+ - task:
351
+ name: Translation nso-fra
352
+ type: translation
353
+ args: nso-fra
354
+ dataset:
355
+ name: flores200-devtest
356
+ type: flores200-devtest
357
+ args: nso-fra
358
+ metrics:
359
+ - name: BLEU
360
+ type: bleu
361
+ value: 10.5
362
+ - name: chr-F
363
+ type: chrf
364
+ value: 0.33732
365
+ - task:
366
+ name: Translation nya-eng
367
+ type: translation
368
+ args: nya-eng
369
+ dataset:
370
+ name: flores200-devtest
371
+ type: flores200-devtest
372
+ args: nya-eng
373
+ metrics:
374
+ - name: BLEU
375
+ type: bleu
376
+ value: 15.8
377
+ - name: chr-F
378
+ type: chrf
379
+ value: 0.39887
380
+ - task:
381
+ name: Translation run-eng
382
+ type: translation
383
+ args: run-eng
384
+ dataset:
385
+ name: flores200-devtest
386
+ type: flores200-devtest
387
+ args: run-eng
388
+ metrics:
389
+ - name: BLEU
390
+ type: bleu
391
+ value: 15.5
392
+ - name: chr-F
393
+ type: chrf
394
+ value: 0.39846
395
+ - task:
396
+ name: Translation run-fra
397
+ type: translation
398
+ args: run-fra
399
+ dataset:
400
+ name: flores200-devtest
401
+ type: flores200-devtest
402
+ args: run-fra
403
+ metrics:
404
+ - name: BLEU
405
+ type: bleu
406
+ value: 10.8
407
+ - name: chr-F
408
+ type: chrf
409
+ value: 0.34845
410
+ - task:
411
+ name: Translation run-por
412
+ type: translation
413
+ args: run-por
414
+ dataset:
415
+ name: flores200-devtest
416
+ type: flores200-devtest
417
+ args: run-por
418
+ metrics:
419
+ - name: BLEU
420
+ type: bleu
421
+ value: 10.2
422
+ - name: chr-F
423
+ type: chrf
424
+ value: 0.33836
425
+ - task:
426
+ name: Translation sna-eng
427
+ type: translation
428
+ args: sna-eng
429
+ dataset:
430
+ name: flores200-devtest
431
+ type: flores200-devtest
432
+ args: sna-eng
433
+ metrics:
434
+ - name: BLEU
435
+ type: bleu
436
+ value: 17.2
437
+ - name: chr-F
438
+ type: chrf
439
+ value: 0.41974
440
+ - task:
441
+ name: Translation sna-fra
442
+ type: translation
443
+ args: sna-fra
444
+ dataset:
445
+ name: flores200-devtest
446
+ type: flores200-devtest
447
+ args: sna-fra
448
+ metrics:
449
+ - name: BLEU
450
+ type: bleu
451
+ value: 11.3
452
+ - name: chr-F
453
+ type: chrf
454
+ value: 0.36443
455
+ - task:
456
+ name: Translation sna-por
457
+ type: translation
458
+ args: sna-por
459
+ dataset:
460
+ name: flores200-devtest
461
+ type: flores200-devtest
462
+ args: sna-por
463
+ metrics:
464
+ - name: BLEU
465
+ type: bleu
466
+ value: 10.2
467
+ - name: chr-F
468
+ type: chrf
469
+ value: 0.34260
470
+ - task:
471
+ name: Translation sot-eng
472
+ type: translation
473
+ args: sot-eng
474
+ dataset:
475
+ name: flores200-devtest
476
+ type: flores200-devtest
477
+ args: sot-eng
478
+ metrics:
479
+ - name: BLEU
480
+ type: bleu
481
+ value: 20.7
482
+ - name: chr-F
483
+ type: chrf
484
+ value: 0.45415
485
+ - task:
486
+ name: Translation sot-fra
487
+ type: translation
488
+ args: sot-fra
489
+ dataset:
490
+ name: flores200-devtest
491
+ type: flores200-devtest
492
+ args: sot-fra
493
+ metrics:
494
+ - name: BLEU
495
+ type: bleu
496
+ value: 10.7
497
+ - name: chr-F
498
+ type: chrf
499
+ value: 0.34608
500
+ - task:
501
+ name: Translation ssw-eng
502
+ type: translation
503
+ args: ssw-eng
504
+ dataset:
505
+ name: flores200-devtest
506
+ type: flores200-devtest
507
+ args: ssw-eng
508
+ metrics:
509
+ - name: BLEU
510
+ type: bleu
511
+ value: 16.1
512
+ - name: chr-F
513
+ type: chrf
514
+ value: 0.39768
515
+ - task:
516
+ name: Translation swh-deu
517
+ type: translation
518
+ args: swh-deu
519
+ dataset:
520
+ name: flores200-devtest
521
+ type: flores200-devtest
522
+ args: swh-deu
523
+ metrics:
524
+ - name: BLEU
525
+ type: bleu
526
+ value: 10.9
527
+ - name: chr-F
528
+ type: chrf
529
+ value: 0.38892
530
+ - task:
531
+ name: Translation swh-eng
532
+ type: translation
533
+ args: swh-eng
534
+ dataset:
535
+ name: flores200-devtest
536
+ type: flores200-devtest
537
+ args: swh-eng
538
+ metrics:
539
+ - name: BLEU
540
+ type: bleu
541
+ value: 29.1
542
+ - name: chr-F
543
+ type: chrf
544
+ value: 0.54048
545
+ - task:
546
+ name: Translation swh-fra
547
+ type: translation
548
+ args: swh-fra
549
+ dataset:
550
+ name: flores200-devtest
551
+ type: flores200-devtest
552
+ args: swh-fra
553
+ metrics:
554
+ - name: BLEU
555
+ type: bleu
556
+ value: 18.2
557
+ - name: chr-F
558
+ type: chrf
559
+ value: 0.44837
560
+ - task:
561
+ name: Translation swh-por
562
+ type: translation
563
+ args: swh-por
564
+ dataset:
565
+ name: flores200-devtest
566
+ type: flores200-devtest
567
+ args: swh-por
568
+ metrics:
569
+ - name: BLEU
570
+ type: bleu
571
+ value: 17.6
572
+ - name: chr-F
573
+ type: chrf
574
+ value: 0.44062
575
+ - task:
576
+ name: Translation swh-spa
577
+ type: translation
578
+ args: swh-spa
579
+ dataset:
580
+ name: flores200-devtest
581
+ type: flores200-devtest
582
+ args: swh-spa
583
+ metrics:
584
+ - name: BLEU
585
+ type: bleu
586
+ value: 11.6
587
+ - name: chr-F
588
+ type: chrf
589
+ value: 0.38855
590
+ - task:
591
+ name: Translation tsn-eng
592
+ type: translation
593
+ args: tsn-eng
594
+ dataset:
595
+ name: flores200-devtest
596
+ type: flores200-devtest
597
+ args: tsn-eng
598
+ metrics:
599
+ - name: BLEU
600
+ type: bleu
601
+ value: 15.3
602
+ - name: chr-F
603
+ type: chrf
604
+ value: 0.40410
605
+ - task:
606
+ name: Translation tsn-fra
607
+ type: translation
608
+ args: tsn-fra
609
+ dataset:
610
+ name: flores200-devtest
611
+ type: flores200-devtest
612
+ args: tsn-fra
613
+ metrics:
614
+ - name: BLEU
615
+ type: bleu
616
+ value: 10.3
617
+ - name: chr-F
618
+ type: chrf
619
+ value: 0.34284
620
+ - task:
621
+ name: Translation tso-eng
622
+ type: translation
623
+ args: tso-eng
624
+ dataset:
625
+ name: flores200-devtest
626
+ type: flores200-devtest
627
+ args: tso-eng
628
+ metrics:
629
+ - name: BLEU
630
+ type: bleu
631
+ value: 17.6
632
+ - name: chr-F
633
+ type: chrf
634
+ value: 0.41504
635
+ - task:
636
+ name: Translation tso-fra
637
+ type: translation
638
+ args: tso-fra
639
+ dataset:
640
+ name: flores200-devtest
641
+ type: flores200-devtest
642
+ args: tso-fra
643
+ metrics:
644
+ - name: BLEU
645
+ type: bleu
646
+ value: 10.1
647
+ - name: chr-F
648
+ type: chrf
649
+ value: 0.33502
650
+ - task:
651
+ name: Translation xho-eng
652
+ type: translation
653
+ args: xho-eng
654
+ dataset:
655
+ name: flores200-devtest
656
+ type: flores200-devtest
657
+ args: xho-eng
658
+ metrics:
659
+ - name: BLEU
660
+ type: bleu
661
+ value: 23.7
662
+ - name: chr-F
663
+ type: chrf
664
+ value: 0.47667
665
+ - task:
666
+ name: Translation xho-fra
667
+ type: translation
668
+ args: xho-fra
669
+ dataset:
670
+ name: flores200-devtest
671
+ type: flores200-devtest
672
+ args: xho-fra
673
+ metrics:
674
+ - name: BLEU
675
+ type: bleu
676
+ value: 14.1
677
+ - name: chr-F
678
+ type: chrf
679
+ value: 0.39392
680
+ - task:
681
+ name: Translation xho-por
682
+ type: translation
683
+ args: xho-por
684
+ dataset:
685
+ name: flores200-devtest
686
+ type: flores200-devtest
687
+ args: xho-por
688
+ metrics:
689
+ - name: BLEU
690
+ type: bleu
691
+ value: 12.3
692
+ - name: chr-F
693
+ type: chrf
694
+ value: 0.37032
695
+ - task:
696
+ name: Translation zul-eng
697
+ type: translation
698
+ args: zul-eng
699
+ dataset:
700
+ name: flores200-devtest
701
+ type: flores200-devtest
702
+ args: zul-eng
703
+ metrics:
704
+ - name: BLEU
705
+ type: bleu
706
+ value: 23.4
707
+ - name: chr-F
708
+ type: chrf
709
+ value: 0.47798
710
+ - task:
711
+ name: Translation zul-fra
712
+ type: translation
713
+ args: zul-fra
714
+ dataset:
715
+ name: flores200-devtest
716
+ type: flores200-devtest
717
+ args: zul-fra
718
+ metrics:
719
+ - name: BLEU
720
+ type: bleu
721
+ value: 13.9
722
+ - name: chr-F
723
+ type: chrf
724
+ value: 0.39504
725
+ - task:
726
+ name: Translation zul-por
727
+ type: translation
728
+ args: zul-por
729
+ dataset:
730
+ name: flores200-devtest
731
+ type: flores200-devtest
732
+ args: zul-por
733
+ metrics:
734
+ - name: BLEU
735
+ type: bleu
736
+ value: 12.3
737
+ - name: chr-F
738
+ type: chrf
739
+ value: 0.36947
740
+ - task:
741
+ name: Translation ibo-eng
742
+ type: translation
743
+ args: ibo-eng
744
+ dataset:
745
+ name: flores101-devtest
746
+ type: flores_101
747
+ args: ibo eng devtest
748
+ metrics:
749
+ - name: BLEU
750
+ type: bleu
751
+ value: 12.5
752
+ - name: chr-F
753
+ type: chrf
754
+ value: 0.36320
755
+ - task:
756
+ name: Translation nya-eng
757
+ type: translation
758
+ args: nya-eng
759
+ dataset:
760
+ name: flores101-devtest
761
+ type: flores_101
762
+ args: nya eng devtest
763
+ metrics:
764
+ - name: BLEU
765
+ type: bleu
766
+ value: 13.2
767
+ - name: chr-F
768
+ type: chrf
769
+ value: 0.36765
770
+ - task:
771
+ name: Translation swh-por
772
+ type: translation
773
+ args: swh-por
774
+ dataset:
775
+ name: flores101-devtest
776
+ type: flores_101
777
+ args: swh por devtest
778
+ metrics:
779
+ - name: BLEU
780
+ type: bleu
781
+ value: 14.7
782
+ - name: chr-F
783
+ type: chrf
784
+ value: 0.40847
785
+ - task:
786
+ name: Translation xho-por
787
+ type: translation
788
+ args: xho-por
789
+ dataset:
790
+ name: flores101-devtest
791
+ type: flores_101
792
+ args: xho por devtest
793
+ metrics:
794
+ - name: BLEU
795
+ type: bleu
796
+ value: 10.1
797
+ - name: chr-F
798
+ type: chrf
799
+ value: 0.33906
800
+ - task:
801
+ name: Translation zul-fra
802
+ type: translation
803
+ args: zul-fra
804
+ dataset:
805
+ name: flores101-devtest
806
+ type: flores_101
807
+ args: zul fra devtest
808
+ metrics:
809
+ - name: BLEU
810
+ type: bleu
811
+ value: 11.2
812
+ - name: chr-F
813
+ type: chrf
814
+ value: 0.35968
815
+ - task:
816
+ name: Translation bem-eng
817
+ type: translation
818
+ args: bem-eng
819
+ dataset:
820
+ name: ntrex128
821
+ type: ntrex128
822
+ args: bem-eng
823
+ metrics:
824
+ - name: BLEU
825
+ type: bleu
826
+ value: 14.9
827
+ - name: chr-F
828
+ type: chrf
829
+ value: 0.38068
830
+ - task:
831
+ name: Translation bem-spa
832
+ type: translation
833
+ args: bem-spa
834
+ dataset:
835
+ name: ntrex128
836
+ type: ntrex128
837
+ args: bem-spa
838
+ metrics:
839
+ - name: BLEU
840
+ type: bleu
841
+ value: 10.2
842
+ - name: chr-F
843
+ type: chrf
844
+ value: 0.33394
845
+ - task:
846
+ name: Translation ibo-eng
847
+ type: translation
848
+ args: ibo-eng
849
+ dataset:
850
+ name: ntrex128
851
+ type: ntrex128
852
+ args: ibo-eng
853
+ metrics:
854
+ - name: BLEU
855
+ type: bleu
856
+ value: 17.4
857
+ - name: chr-F
858
+ type: chrf
859
+ value: 0.42002
860
+ - task:
861
+ name: Translation ibo-fra
862
+ type: translation
863
+ args: ibo-fra
864
+ dataset:
865
+ name: ntrex128
866
+ type: ntrex128
867
+ args: ibo-fra
868
+ metrics:
869
+ - name: BLEU
870
+ type: bleu
871
+ value: 10.3
872
+ - name: chr-F
873
+ type: chrf
874
+ value: 0.34993
875
+ - task:
876
+ name: Translation ibo-spa
877
+ type: translation
878
+ args: ibo-spa
879
+ dataset:
880
+ name: ntrex128
881
+ type: ntrex128
882
+ args: ibo-spa
883
+ metrics:
884
+ - name: BLEU
885
+ type: bleu
886
+ value: 11.7
887
+ - name: chr-F
888
+ type: chrf
889
+ value: 0.36108
890
+ - task:
891
+ name: Translation kin-eng
892
+ type: translation
893
+ args: kin-eng
894
+ dataset:
895
+ name: ntrex128
896
+ type: ntrex128
897
+ args: kin-eng
898
+ metrics:
899
+ - name: BLEU
900
+ type: bleu
901
+ value: 16.9
902
+ - name: chr-F
903
+ type: chrf
904
+ value: 0.42892
905
+ - task:
906
+ name: Translation kin-fra
907
+ type: translation
908
+ args: kin-fra
909
+ dataset:
910
+ name: ntrex128
911
+ type: ntrex128
912
+ args: kin-fra
913
+ metrics:
914
+ - name: BLEU
915
+ type: bleu
916
+ value: 10.7
917
+ - name: chr-F
918
+ type: chrf
919
+ value: 0.35842
920
+ - task:
921
+ name: Translation kin-por
922
+ type: translation
923
+ args: kin-por
924
+ dataset:
925
+ name: ntrex128
926
+ type: ntrex128
927
+ args: kin-por
928
+ metrics:
929
+ - name: BLEU
930
+ type: bleu
931
+ value: 10.0
932
+ - name: chr-F
933
+ type: chrf
934
+ value: 0.34399
935
+ - task:
936
+ name: Translation kin-spa
937
+ type: translation
938
+ args: kin-spa
939
+ dataset:
940
+ name: ntrex128
941
+ type: ntrex128
942
+ args: kin-spa
943
+ metrics:
944
+ - name: BLEU
945
+ type: bleu
946
+ value: 12.5
947
+ - name: chr-F
948
+ type: chrf
949
+ value: 0.37224
950
+ - task:
951
+ name: Translation nde-eng
952
+ type: translation
953
+ args: nde-eng
954
+ dataset:
955
+ name: ntrex128
956
+ type: ntrex128
957
+ args: nde-eng
958
+ metrics:
959
+ - name: BLEU
960
+ type: bleu
961
+ value: 13.8
962
+ - name: chr-F
963
+ type: chrf
964
+ value: 0.39640
965
+ - task:
966
+ name: Translation nde-spa
967
+ type: translation
968
+ args: nde-spa
969
+ dataset:
970
+ name: ntrex128
971
+ type: ntrex128
972
+ args: nde-spa
973
+ metrics:
974
+ - name: BLEU
975
+ type: bleu
976
+ value: 10.0
977
+ - name: chr-F
978
+ type: chrf
979
+ value: 0.34638
980
+ - task:
981
+ name: Translation nso-eng
982
+ type: translation
983
+ args: nso-eng
984
+ dataset:
985
+ name: ntrex128
986
+ type: ntrex128
987
+ args: nso-eng
988
+ metrics:
989
+ - name: BLEU
990
+ type: bleu
991
+ value: 17.0
992
+ - name: chr-F
993
+ type: chrf
994
+ value: 0.42278
995
+ - task:
996
+ name: Translation nso-spa
997
+ type: translation
998
+ args: nso-spa
999
+ dataset:
1000
+ name: ntrex128
1001
+ type: ntrex128
1002
+ args: nso-spa
1003
+ metrics:
1004
+ - name: BLEU
1005
+ type: bleu
1006
+ value: 11.3
1007
+ - name: chr-F
1008
+ type: chrf
1009
+ value: 0.35027
1010
+ - task:
1011
+ name: Translation nya-eng
1012
+ type: translation
1013
+ args: nya-eng
1014
+ dataset:
1015
+ name: ntrex128
1016
+ type: ntrex128
1017
+ args: nya-eng
1018
+ metrics:
1019
+ - name: BLEU
1020
+ type: bleu
1021
+ value: 19.2
1022
+ - name: chr-F
1023
+ type: chrf
1024
+ value: 0.42702
1025
+ - task:
1026
+ name: Translation nya-fra
1027
+ type: translation
1028
+ args: nya-fra
1029
+ dataset:
1030
+ name: ntrex128
1031
+ type: ntrex128
1032
+ args: nya-fra
1033
+ metrics:
1034
+ - name: BLEU
1035
+ type: bleu
1036
+ value: 10.7
1037
+ - name: chr-F
1038
+ type: chrf
1039
+ value: 0.35503
1040
+ - task:
1041
+ name: Translation nya-por
1042
+ type: translation
1043
+ args: nya-por
1044
+ dataset:
1045
+ name: ntrex128
1046
+ type: ntrex128
1047
+ args: nya-por
1048
+ metrics:
1049
+ - name: BLEU
1050
+ type: bleu
1051
+ value: 10.4
1052
+ - name: chr-F
1053
+ type: chrf
1054
+ value: 0.33912
1055
+ - task:
1056
+ name: Translation nya-spa
1057
+ type: translation
1058
+ args: nya-spa
1059
+ dataset:
1060
+ name: ntrex128
1061
+ type: ntrex128
1062
+ args: nya-spa
1063
+ metrics:
1064
+ - name: BLEU
1065
+ type: bleu
1066
+ value: 12.5
1067
+ - name: chr-F
1068
+ type: chrf
1069
+ value: 0.36355
1070
+ - task:
1071
+ name: Translation ssw-eng
1072
+ type: translation
1073
+ args: ssw-eng
1074
+ dataset:
1075
+ name: ntrex128
1076
+ type: ntrex128
1077
+ args: ssw-eng
1078
+ metrics:
1079
+ - name: BLEU
1080
+ type: bleu
1081
+ value: 18.0
1082
+ - name: chr-F
1083
+ type: chrf
1084
+ value: 0.43041
1085
+ - task:
1086
+ name: Translation ssw-spa
1087
+ type: translation
1088
+ args: ssw-spa
1089
+ dataset:
1090
+ name: ntrex128
1091
+ type: ntrex128
1092
+ args: ssw-spa
1093
+ metrics:
1094
+ - name: BLEU
1095
+ type: bleu
1096
+ value: 11.4
1097
+ - name: chr-F
1098
+ type: chrf
1099
+ value: 0.35392
1100
+ - task:
1101
+ name: Translation swa-deu
1102
+ type: translation
1103
+ args: swa-deu
1104
+ dataset:
1105
+ name: ntrex128
1106
+ type: ntrex128
1107
+ args: swa-deu
1108
+ metrics:
1109
+ - name: BLEU
1110
+ type: bleu
1111
+ value: 11.2
1112
+ - name: chr-F
1113
+ type: chrf
1114
+ value: 0.39475
1115
+ - task:
1116
+ name: Translation swa-eng
1117
+ type: translation
1118
+ args: swa-eng
1119
+ dataset:
1120
+ name: ntrex128
1121
+ type: ntrex128
1122
+ args: swa-eng
1123
+ metrics:
1124
+ - name: BLEU
1125
+ type: bleu
1126
+ value: 30.4
1127
+ - name: chr-F
1128
+ type: chrf
1129
+ value: 0.54492
1130
+ - task:
1131
+ name: Translation swa-fra
1132
+ type: translation
1133
+ args: swa-fra
1134
+ dataset:
1135
+ name: ntrex128
1136
+ type: ntrex128
1137
+ args: swa-fra
1138
+ metrics:
1139
+ - name: BLEU
1140
+ type: bleu
1141
+ value: 15.6
1142
+ - name: chr-F
1143
+ type: chrf
1144
+ value: 0.43008
1145
+ - task:
1146
+ name: Translation swa-por
1147
+ type: translation
1148
+ args: swa-por
1149
+ dataset:
1150
+ name: ntrex128
1151
+ type: ntrex128
1152
+ args: swa-por
1153
+ metrics:
1154
+ - name: BLEU
1155
+ type: bleu
1156
+ value: 15.4
1157
+ - name: chr-F
1158
+ type: chrf
1159
+ value: 0.42343
1160
+ - task:
1161
+ name: Translation swa-spa
1162
+ type: translation
1163
+ args: swa-spa
1164
+ dataset:
1165
+ name: ntrex128
1166
+ type: ntrex128
1167
+ args: swa-spa
1168
+ metrics:
1169
+ - name: BLEU
1170
+ type: bleu
1171
+ value: 18.9
1172
+ - name: chr-F
1173
+ type: chrf
1174
+ value: 0.44892
1175
+ - task:
1176
+ name: Translation tsn-eng
1177
+ type: translation
1178
+ args: tsn-eng
1179
+ dataset:
1180
+ name: ntrex128
1181
+ type: ntrex128
1182
+ args: tsn-eng
1183
+ metrics:
1184
+ - name: BLEU
1185
+ type: bleu
1186
+ value: 20.1
1187
+ - name: chr-F
1188
+ type: chrf
1189
+ value: 0.44944
1190
+ - task:
1191
+ name: Translation tsn-fra
1192
+ type: translation
1193
+ args: tsn-fra
1194
+ dataset:
1195
+ name: ntrex128
1196
+ type: ntrex128
1197
+ args: tsn-fra
1198
+ metrics:
1199
+ - name: BLEU
1200
+ type: bleu
1201
+ value: 11.5
1202
+ - name: chr-F
1203
+ type: chrf
1204
+ value: 0.36584
1205
+ - task:
1206
+ name: Translation tsn-por
1207
+ type: translation
1208
+ args: tsn-por
1209
+ dataset:
1210
+ name: ntrex128
1211
+ type: ntrex128
1212
+ args: tsn-por
1213
+ metrics:
1214
+ - name: BLEU
1215
+ type: bleu
1216
+ value: 11.1
1217
+ - name: chr-F
1218
+ type: chrf
1219
+ value: 0.35774
1220
+ - task:
1221
+ name: Translation tsn-spa
1222
+ type: translation
1223
+ args: tsn-spa
1224
+ dataset:
1225
+ name: ntrex128
1226
+ type: ntrex128
1227
+ args: tsn-spa
1228
+ metrics:
1229
+ - name: BLEU
1230
+ type: bleu
1231
+ value: 13.8
1232
+ - name: chr-F
1233
+ type: chrf
1234
+ value: 0.38149
1235
+ - task:
1236
+ name: Translation ven-eng
1237
+ type: translation
1238
+ args: ven-eng
1239
+ dataset:
1240
+ name: ntrex128
1241
+ type: ntrex128
1242
+ args: ven-eng
1243
+ metrics:
1244
+ - name: BLEU
1245
+ type: bleu
1246
+ value: 12.7
1247
+ - name: chr-F
1248
+ type: chrf
1249
+ value: 0.36848
1250
+ - task:
1251
+ name: Translation xho-eng
1252
+ type: translation
1253
+ args: xho-eng
1254
+ dataset:
1255
+ name: ntrex128
1256
+ type: ntrex128
1257
+ args: xho-eng
1258
+ metrics:
1259
+ - name: BLEU
1260
+ type: bleu
1261
+ value: 21.8
1262
+ - name: chr-F
1263
+ type: chrf
1264
+ value: 0.46636
1265
+ - task:
1266
+ name: Translation xho-fra
1267
+ type: translation
1268
+ args: xho-fra
1269
+ dataset:
1270
+ name: ntrex128
1271
+ type: ntrex128
1272
+ args: xho-fra
1273
+ metrics:
1274
+ - name: BLEU
1275
+ type: bleu
1276
+ value: 11.6
1277
+ - name: chr-F
1278
+ type: chrf
1279
+ value: 0.36905
1280
+ - task:
1281
+ name: Translation xho-por
1282
+ type: translation
1283
+ args: xho-por
1284
+ dataset:
1285
+ name: ntrex128
1286
+ type: ntrex128
1287
+ args: xho-por
1288
+ metrics:
1289
+ - name: BLEU
1290
+ type: bleu
1291
+ value: 10.6
1292
+ - name: chr-F
1293
+ type: chrf
1294
+ value: 0.35687
1295
+ - task:
1296
+ name: Translation xho-spa
1297
+ type: translation
1298
+ args: xho-spa
1299
+ dataset:
1300
+ name: ntrex128
1301
+ type: ntrex128
1302
+ args: xho-spa
1303
+ metrics:
1304
+ - name: BLEU
1305
+ type: bleu
1306
+ value: 13.3
1307
+ - name: chr-F
1308
+ type: chrf
1309
+ value: 0.38176
1310
+ - task:
1311
+ name: Translation yor-eng
1312
+ type: translation
1313
+ args: yor-eng
1314
+ dataset:
1315
+ name: ntrex128
1316
+ type: ntrex128
1317
+ args: yor-eng
1318
+ metrics:
1319
+ - name: BLEU
1320
+ type: bleu
1321
+ value: 11.2
1322
+ - name: chr-F
1323
+ type: chrf
1324
+ value: 0.34615
1325
+ - task:
1326
+ name: Translation zul-eng
1327
+ type: translation
1328
+ args: zul-eng
1329
+ dataset:
1330
+ name: ntrex128
1331
+ type: ntrex128
1332
+ args: zul-eng
1333
+ metrics:
1334
+ - name: BLEU
1335
+ type: bleu
1336
+ value: 21.9
1337
+ - name: chr-F
1338
+ type: chrf
1339
+ value: 0.45848
1340
+ - task:
1341
+ name: Translation zul-fra
1342
+ type: translation
1343
+ args: zul-fra
1344
+ dataset:
1345
+ name: ntrex128
1346
+ type: ntrex128
1347
+ args: zul-fra
1348
+ metrics:
1349
+ - name: BLEU
1350
+ type: bleu
1351
+ value: 11.4
1352
+ - name: chr-F
1353
+ type: chrf
1354
+ value: 0.36203
1355
+ - task:
1356
+ name: Translation zul-por
1357
+ type: translation
1358
+ args: zul-por
1359
+ dataset:
1360
+ name: ntrex128
1361
+ type: ntrex128
1362
+ args: zul-por
1363
+ metrics:
1364
+ - name: BLEU
1365
+ type: bleu
1366
+ value: 10.8
1367
+ - name: chr-F
1368
+ type: chrf
1369
+ value: 0.35081
1370
+ - task:
1371
+ name: Translation zul-spa
1372
+ type: translation
1373
+ args: zul-spa
1374
+ dataset:
1375
+ name: ntrex128
1376
+ type: ntrex128
1377
+ args: zul-spa
1378
+ metrics:
1379
+ - name: BLEU
1380
+ type: bleu
1381
+ value: 13.0
1382
+ - name: chr-F
1383
+ type: chrf
1384
+ value: 0.37270
1385
+ - task:
1386
+ name: Translation multi-multi
1387
+ type: translation
1388
+ args: multi-multi
1389
+ dataset:
1390
+ name: tatoeba-test-v2020-07-28-v2023-09-26
1391
+ type: tatoeba_mt
1392
+ args: multi-multi
1393
+ metrics:
1394
+ - name: BLEU
1395
+ type: bleu
1396
+ value: 27.3
1397
+ - name: chr-F
1398
+ type: chrf
1399
+ value: 0.44751
1400
+ - task:
1401
+ name: Translation run-eng
1402
+ type: translation
1403
+ args: run-eng
1404
+ dataset:
1405
+ name: tatoeba-test-v2021-08-07
1406
+ type: tatoeba_mt
1407
+ args: run-eng
1408
+ metrics:
1409
+ - name: BLEU
1410
+ type: bleu
1411
+ value: 34.9
1412
+ - name: chr-F
1413
+ type: chrf
1414
+ value: 0.49949
1415
+ - task:
1416
+ name: Translation run-fra
1417
+ type: translation
1418
+ args: run-fra
1419
+ dataset:
1420
+ name: tatoeba-test-v2021-08-07
1421
+ type: tatoeba_mt
1422
+ args: run-fra
1423
+ metrics:
1424
+ - name: BLEU
1425
+ type: bleu
1426
+ value: 22.4
1427
+ - name: chr-F
1428
+ type: chrf
1429
+ value: 0.41431
1430
+ - task:
1431
+ name: Translation swa-eng
1432
+ type: translation
1433
+ args: swa-eng
1434
+ dataset:
1435
+ name: tatoeba-test-v2021-08-07
1436
+ type: tatoeba_mt
1437
+ args: swa-eng
1438
+ metrics:
1439
+ - name: BLEU
1440
+ type: bleu
1441
+ value: 41.5
1442
+ - name: chr-F
1443
+ type: chrf
1444
+ value: 0.57031
1445
+ - task:
1446
+ name: Translation xho-eng
1447
+ type: translation
1448
+ args: xho-eng
1449
+ dataset:
1450
+ name: tatoeba-test-v2021-03-30
1451
+ type: tatoeba_mt
1452
+ args: xho-eng
1453
+ metrics:
1454
+ - name: BLEU
1455
+ type: bleu
1456
+ value: 36.4
1457
+ - name: chr-F
1458
+ type: chrf
1459
+ value: 0.51250
1460
+ - task:
1461
+ name: Translation zul-eng
1462
+ type: translation
1463
+ args: zul-eng
1464
+ dataset:
1465
+ name: tico19-test
1466
+ type: tico19-test
1467
+ args: zul-eng
1468
+ metrics:
1469
+ - name: BLEU
1470
+ type: bleu
1471
+ value: 25.2
1472
+ - name: chr-F
1473
+ type: chrf
1474
+ value: 0.48762
1475
+ - task:
1476
+ name: Translation zul-fra
1477
+ type: translation
1478
+ args: zul-fra
1479
+ dataset:
1480
+ name: tico19-test
1481
+ type: tico19-test
1482
+ args: zul-fra
1483
+ metrics:
1484
+ - name: BLEU
1485
+ type: bleu
1486
+ value: 13.7
1487
+ - name: chr-F
1488
+ type: chrf
1489
+ value: 0.37823
1490
+ - task:
1491
+ name: Translation zul-por
1492
+ type: translation
1493
+ args: zul-por
1494
+ dataset:
1495
+ name: tico19-test
1496
+ type: tico19-test
1497
+ args: zul-por
1498
+ metrics:
1499
+ - name: BLEU
1500
+ type: bleu
1501
+ value: 14.0
1502
+ - name: chr-F
1503
+ type: chrf
1504
+ value: 0.38586
1505
+ - task:
1506
+ name: Translation zul-spa
1507
+ type: translation
1508
+ args: zul-spa
1509
+ dataset:
1510
+ name: tico19-test
1511
+ type: tico19-test
1512
+ args: zul-spa
1513
+ metrics:
1514
+ - name: BLEU
1515
+ type: bleu
1516
+ value: 15.9
1517
+ - name: chr-F
1518
+ type: chrf
1519
+ value: 0.40041
1520
+ ---
1521
+ # opus-mt-tc-bible-big-alv-deu_eng_fra_por_spa
1522
+
1523
+ ## Table of Contents
1524
+ - [Model Details](#model-details)
1525
+ - [Uses](#uses)
1526
+ - [Risks, Limitations and Biases](#risks-limitations-and-biases)
1527
+ - [How to Get Started With the Model](#how-to-get-started-with-the-model)
1528
+ - [Training](#training)
1529
+ - [Evaluation](#evaluation)
1530
+ - [Citation Information](#citation-information)
1531
+ - [Acknowledgements](#acknowledgements)
1532
+
1533
+ ## Model Details
1534
+
1535
+ Neural machine translation model for translating from Atlantic-Congo languages (alv) to unknown (deu+eng+fra+por+spa).
1536
+
1537
+ This model is part of the [OPUS-MT project](https://github.com/Helsinki-NLP/Opus-MT), an effort to make neural machine translation models widely available and accessible for many languages in the world. All models are originally trained using the amazing framework of [Marian NMT](https://marian-nmt.github.io/), an efficient NMT implementation written in pure C++. The models have been converted to pyTorch using the transformers library by huggingface. Training data is taken from [OPUS](https://opus.nlpl.eu/) and training pipelines use the procedures of [OPUS-MT-train](https://github.com/Helsinki-NLP/Opus-MT-train).
1538
+ **Model Description:**
1539
+ - **Developed by:** Language Technology Research Group at the University of Helsinki
1540
+ - **Model Type:** Translation (transformer-big)
1541
+ - **Release**: 2024-05-30
1542
+ - **License:** Apache-2.0
1543
+ - **Language(s):**
1544
+ - Source Language(s): abi acd ade adj aka akp ann anv atg avn bas bav bba beh bem bfd bfo bim biv bkv blh bmq bmv bom bov box bqj bss btt bud bwu cce cjk cko cme csk cwe cwt dag dga dgi dig dop dug dyi dyo efi ewe fal fon fuc ful gej gkn gng gog gud gur guw gux gwr hag hay heh her ibo ife iri izr jbu jmc kam kbp kdc kdl kdn ken keu kez kia kik kin kki kkj kma kmb kon ksb ktj kua kub kus kyf las lee lef lem lia lin lip lob lon lua lug luy maw mcp mcu mda mfq mgo mnf mnh mor mos muh myk myx mzk mzm mzw nbl ncu nde ndo ndz nfr nhu nim nin nmz nnb nnh nnw nso ntm ntr nuj nwb nya nyf nyn nyo nyy nzi oku old ozm pai pbl pkb rim run sag seh sig sil sld sna snw sot soy spp ssw suk swa swc swh sxb tbz tem thk tik tlj toh toi tpm tsn tso tsw tum twi umb vag ven vmw vun wmw wob wol xho xog xon xrb xsm xuo yam yaz yor zul
1545
+ - Target Language(s): deu eng fra por spa
1546
+ - Valid Target Language Labels: >>deu<< >>eng<< >>fra<< >>por<< >>spa<< >>xxx<<
1547
+ - **Original Model**: [opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-30.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/alv-deu+eng+fra+por+spa/opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-30.zip)
1548
+ - **Resources for more information:**
1549
+ - [OPUS-MT dashboard](https://opus.nlpl.eu/dashboard/index.php?pkg=opusmt&test=all&scoreslang=all&chart=standard&model=Tatoeba-MT-models/alv-deu%2Beng%2Bfra%2Bpor%2Bspa/opusTCv20230926max50%2Bbt%2Bjhubc_transformer-big_2024-05-30)
1550
+ - [OPUS-MT-train GitHub Repo](https://github.com/Helsinki-NLP/OPUS-MT-train)
1551
+ - [More information about MarianNMT models in the transformers library](https://huggingface.co/docs/transformers/model_doc/marian)
1552
+ - [Tatoeba Translation Challenge](https://github.com/Helsinki-NLP/Tatoeba-Challenge/)
1553
+ - [HPLT bilingual data v1 (as part of the Tatoeba Translation Challenge dataset)](https://hplt-project.org/datasets/v1)
1554
+ - [A massively parallel Bible corpus](https://aclanthology.org/L14-1215/)
1555
+
1556
+ This is a multilingual translation model with multiple target languages. A sentence initial language token is required in the form of `>>id<<` (id = valid target language ID), e.g. `>>deu<<`
1557
+
1558
+ ## Uses
1559
+
1560
+ This model can be used for translation and text-to-text generation.
1561
+
1562
+ ## Risks, Limitations and Biases
1563
+
1564
+ **CONTENT WARNING: Readers should be aware that the model is trained on various public data sets that may contain content that is disturbing, offensive, and can propagate historical and current stereotypes.**
1565
+
1566
+ Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)).
1567
+
1568
+ ## How to Get Started With the Model
1569
+
1570
+ A short example code:
1571
+
1572
+ ```python
1573
+ from transformers import MarianMTModel, MarianTokenizer
1574
+
1575
+ src_text = [
1576
+ ">>deu<< Replace this with text in an accepted source language.",
1577
+ ">>spa<< This is the second sentence."
1578
+ ]
1579
+
1580
+ model_name = "pytorch-models/opus-mt-tc-bible-big-alv-deu_eng_fra_por_spa"
1581
+ tokenizer = MarianTokenizer.from_pretrained(model_name)
1582
+ model = MarianMTModel.from_pretrained(model_name)
1583
+ translated = model.generate(**tokenizer(src_text, return_tensors="pt", padding=True))
1584
+
1585
+ for t in translated:
1586
+ print( tokenizer.decode(t, skip_special_tokens=True) )
1587
+ ```
1588
+
1589
+ You can also use OPUS-MT models with the transformers pipelines, for example:
1590
+
1591
+ ```python
1592
+ from transformers import pipeline
1593
+ pipe = pipeline("translation", model="Helsinki-NLP/opus-mt-tc-bible-big-alv-deu_eng_fra_por_spa")
1594
+ print(pipe(">>deu<< Replace this with text in an accepted source language."))
1595
+ ```
1596
+
1597
+ ## Training
1598
+
1599
+ - **Data**: opusTCv20230926max50+bt+jhubc ([source](https://github.com/Helsinki-NLP/Tatoeba-Challenge))
1600
+ - **Pre-processing**: SentencePiece (spm32k,spm32k)
1601
+ - **Model Type:** transformer-big
1602
+ - **Original MarianNMT Model**: [opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-30.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/alv-deu+eng+fra+por+spa/opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-30.zip)
1603
+ - **Training Scripts**: [GitHub Repo](https://github.com/Helsinki-NLP/OPUS-MT-train)
1604
+
1605
+ ## Evaluation
1606
+
1607
+ * [Model scores at the OPUS-MT dashboard](https://opus.nlpl.eu/dashboard/index.php?pkg=opusmt&test=all&scoreslang=all&chart=standard&model=Tatoeba-MT-models/alv-deu%2Beng%2Bfra%2Bpor%2Bspa/opusTCv20230926max50%2Bbt%2Bjhubc_transformer-big_2024-05-30)
1608
+ * test set translations: [opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-29.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/alv-deu+eng+fra+por+spa/opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-29.test.txt)
1609
+ * test set scores: [opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-29.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/alv-deu+eng+fra+por+spa/opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-29.eval.txt)
1610
+ * benchmark results: [benchmark_results.txt](benchmark_results.txt)
1611
+ * benchmark output: [benchmark_translations.zip](benchmark_translations.zip)
1612
+
1613
+ | langpair | testset | chr-F | BLEU | #sent | #words |
1614
+ |----------|---------|-------|-------|-------|--------|
1615
+ | run-eng | tatoeba-test-v2021-08-07 | 0.49949 | 34.9 | 1703 | 10041 |
1616
+ | run-fra | tatoeba-test-v2021-08-07 | 0.41431 | 22.4 | 1274 | 7479 |
1617
+ | swa-eng | tatoeba-test-v2021-08-07 | 0.57031 | 41.5 | 387 | 2508 |
1618
+ | swh-por | flores101-devtest | 0.40847 | 14.7 | 1012 | 26519 |
1619
+ | kin-eng | flores200-devtest | 0.41964 | 18.1 | 1012 | 24721 |
1620
+ | nso-eng | flores200-devtest | 0.45662 | 22.3 | 1012 | 24721 |
1621
+ | sna-eng | flores200-devtest | 0.41974 | 17.2 | 1012 | 24721 |
1622
+ | sot-eng | flores200-devtest | 0.45415 | 20.7 | 1012 | 24721 |
1623
+ | swh-eng | flores200-devtest | 0.54048 | 29.1 | 1012 | 24721 |
1624
+ | swh-fra | flores200-devtest | 0.44837 | 18.2 | 1012 | 28343 |
1625
+ | swh-por | flores200-devtest | 0.44062 | 17.6 | 1012 | 26519 |
1626
+ | tsn-eng | flores200-devtest | 0.40410 | 15.3 | 1012 | 24721 |
1627
+ | tso-eng | flores200-devtest | 0.41504 | 17.6 | 1012 | 24721 |
1628
+ | xho-eng | flores200-devtest | 0.47667 | 23.7 | 1012 | 24721 |
1629
+ | zul-eng | flores200-devtest | 0.47798 | 23.4 | 1012 | 24721 |
1630
+ | ibo-eng | ntrex128 | 0.42002 | 17.4 | 1997 | 47673 |
1631
+ | kin-eng | ntrex128 | 0.42892 | 16.9 | 1997 | 47673 |
1632
+ | nso-eng | ntrex128 | 0.42278 | 17.0 | 1997 | 47673 |
1633
+ | nya-eng | ntrex128 | 0.42702 | 19.2 | 1997 | 47673 |
1634
+ | ssw-eng | ntrex128 | 0.43041 | 18.0 | 1997 | 47673 |
1635
+ | swa-eng | ntrex128 | 0.54492 | 30.4 | 1997 | 47673 |
1636
+ | swa-fra | ntrex128 | 0.43008 | 15.6 | 1997 | 53481 |
1637
+ | swa-por | ntrex128 | 0.42343 | 15.4 | 1997 | 51631 |
1638
+ | swa-spa | ntrex128 | 0.44892 | 18.9 | 1997 | 54107 |
1639
+ | tsn-eng | ntrex128 | 0.44944 | 20.1 | 1997 | 47673 |
1640
+ | xho-eng | ntrex128 | 0.46636 | 21.8 | 1997 | 47673 |
1641
+ | zul-eng | ntrex128 | 0.45848 | 21.9 | 1997 | 47673 |
1642
+ | zul-eng | tico19-test | 0.48762 | 25.2 | 2100 | 56804 |
1643
+ | zul-spa | tico19-test | 0.40041 | 15.9 | 2100 | 66563 |
1644
+
1645
+ ## Citation Information
1646
+
1647
+ * Publications: [Democratizing neural machine translation with OPUS-MT](https://doi.org/10.1007/s10579-023-09704-w) and [OPUS-MT – Building open translation services for the World](https://aclanthology.org/2020.eamt-1.61/) and [The Tatoeba Translation Challenge – Realistic Data Sets for Low Resource and Multilingual MT](https://aclanthology.org/2020.wmt-1.139/) (Please, cite if you use this model.)
1648
+
1649
+ ```bibtex
1650
+ @article{tiedemann2023democratizing,
1651
+ title={Democratizing neural machine translation with {OPUS-MT}},
1652
+ author={Tiedemann, J{\"o}rg and Aulamo, Mikko and Bakshandaeva, Daria and Boggia, Michele and Gr{\"o}nroos, Stig-Arne and Nieminen, Tommi and Raganato, Alessandro and Scherrer, Yves and Vazquez, Raul and Virpioja, Sami},
1653
+ journal={Language Resources and Evaluation},
1654
+ number={58},
1655
+ pages={713--755},
1656
+ year={2023},
1657
+ publisher={Springer Nature},
1658
+ issn={1574-0218},
1659
+ doi={10.1007/s10579-023-09704-w}
1660
+ }
1661
+
1662
+ @inproceedings{tiedemann-thottingal-2020-opus,
1663
+ title = "{OPUS}-{MT} {--} Building open translation services for the World",
1664
+ author = {Tiedemann, J{\"o}rg and Thottingal, Santhosh},
1665
+ booktitle = "Proceedings of the 22nd Annual Conference of the European Association for Machine Translation",
1666
+ month = nov,
1667
+ year = "2020",
1668
+ address = "Lisboa, Portugal",
1669
+ publisher = "European Association for Machine Translation",
1670
+ url = "https://aclanthology.org/2020.eamt-1.61",
1671
+ pages = "479--480",
1672
+ }
1673
+
1674
+ @inproceedings{tiedemann-2020-tatoeba,
1675
+ title = "The Tatoeba Translation Challenge {--} Realistic Data Sets for Low Resource and Multilingual {MT}",
1676
+ author = {Tiedemann, J{\"o}rg},
1677
+ booktitle = "Proceedings of the Fifth Conference on Machine Translation",
1678
+ month = nov,
1679
+ year = "2020",
1680
+ address = "Online",
1681
+ publisher = "Association for Computational Linguistics",
1682
+ url = "https://aclanthology.org/2020.wmt-1.139",
1683
+ pages = "1174--1182",
1684
+ }
1685
+ ```
1686
+
1687
+ ## Acknowledgements
1688
+
1689
+ The work is supported by the [HPLT project](https://hplt-project.org/), funded by the European Union’s Horizon Europe research and innovation programme under grant agreement No 101070350. We are also grateful for the generous computational resources and IT infrastructure provided by [CSC -- IT Center for Science](https://www.csc.fi/), Finland, and the [EuroHPC supercomputer LUMI](https://www.lumi-supercomputer.eu/).
1690
+
1691
+ ## Model conversion info
1692
+
1693
+ * transformers version: 4.45.1
1694
+ * OPUS-MT git hash: a0ea3b3
1695
+ * port time: Mon Oct 7 16:22:19 EEST 2024
1696
+ * port machine: LM0-400-22516.local
benchmark_results.txt ADDED
@@ -0,0 +1,296 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ multi-multi tatoeba-test-v2020-07-28-v2023-09-26 0.44751 27.3 6455 38460
2
+ ful-eng flores101-devtest 0.21713 3.5 1012 24721
3
+ ful-fra flores101-devtest 0.19514 2.1 1012 28343
4
+ ibo-deu flores101-devtest 0.29329 4.8 1012 25094
5
+ ibo-eng flores101-devtest 0.36320 12.5 1012 24721
6
+ kam-eng flores101-devtest 0.23819 5.3 1012 24721
7
+ kam-por flores101-devtest 0.20757 3.3 1012 26519
8
+ lin-deu flores101-devtest 0.28572 5.2 1012 25094
9
+ lin-fra flores101-devtest 0.31914 8.9 1012 28343
10
+ lin-por flores101-devtest 0.31177 8.4 1012 26519
11
+ lin-spa flores101-devtest 0.29320 6.6 1012 29199
12
+ lug-deu flores101-devtest 0.24065 3.3 1012 25094
13
+ nso-fra flores101-devtest 0.29772 7.8 1012 28343
14
+ nya-eng flores101-devtest 0.36765 13.2 1012 24721
15
+ nya-fra flores101-devtest 0.30619 8.0 1012 28343
16
+ sna-fra flores101-devtest 0.34168 9.6 1012 28343
17
+ sna-spa flores101-devtest 0.30187 6.8 1012 29199
18
+ swh-por flores101-devtest 0.40847 14.7 1012 26519
19
+ umb-deu flores101-devtest 0.20753 1.4 1012 25094
20
+ umb-eng flores101-devtest 0.22787 3.3 1012 24721
21
+ wol-fra flores101-devtest 0.23494 3.9 1012 28343
22
+ wol-por flores101-devtest 0.23358 3.7 1012 26519
23
+ xho-por flores101-devtest 0.33906 10.1 1012 26519
24
+ yor-fra flores101-devtest 0.24198 4.0 1012 28343
25
+ zul-fra flores101-devtest 0.35968 11.2 1012 28343
26
+ aka-deu flores200-devtest 0.21103 2.0 1012 25094
27
+ aka-eng flores200-devtest 0.25750 5.6 1012 24721
28
+ aka-fra flores200-devtest 0.21075 3.1 1012 28343
29
+ aka-por flores200-devtest 0.20728 3.2 1012 26519
30
+ aka-spa flores200-devtest 0.21494 3.4 1012 29199
31
+ bem-deu flores200-devtest 0.29324 4.7 1012 25094
32
+ bem-eng flores200-devtest 0.37071 13.1 1012 24721
33
+ bem-fra flores200-devtest 0.31561 8.6 1012 28343
34
+ bem-por flores200-devtest 0.30790 8.1 1012 26519
35
+ bem-spa flores200-devtest 0.29661 6.9 1012 29199
36
+ cjk-deu flores200-devtest 0.21803 1.8 1012 25094
37
+ cjk-eng flores200-devtest 0.23825 3.9 1012 24721
38
+ cjk-fra flores200-devtest 0.21484 2.8 1012 28343
39
+ cjk-por flores200-devtest 0.22811 3.3 1012 26519
40
+ cjk-spa flores200-devtest 0.23353 3.2 1012 29199
41
+ ewe-deu flores200-devtest 0.20417 1.9 1012 25094
42
+ ewe-eng flores200-devtest 0.24329 4.6 1012 24721
43
+ ewe-fra flores200-devtest 0.19872 2.7 1012 28343
44
+ ewe-por flores200-devtest 0.19574 2.4 1012 26519
45
+ ewe-spa flores200-devtest 0.20390 2.8 1012 29199
46
+ fon-deu flores200-devtest 0.21065 1.7 1012 25094
47
+ fon-eng flores200-devtest 0.23791 3.8 1012 24721
48
+ fon-fra flores200-devtest 0.20637 2.8 1012 28343
49
+ fon-por flores200-devtest 0.20322 2.5 1012 26519
50
+ fon-spa flores200-devtest 0.21140 2.8 1012 29199
51
+ ibo-deu flores200-devtest 0.31201 5.9 1012 25094
52
+ ibo-eng flores200-devtest 0.38994 14.6 1012 24721
53
+ ibo-fra flores200-devtest 0.33654 9.3 1012 28343
54
+ ibo-por flores200-devtest 0.32381 8.7 1012 26519
55
+ ibo-spa flores200-devtest 0.30903 6.8 1012 29199
56
+ kam-deu flores200-devtest 0.22623 2.4 1012 25094
57
+ kam-eng flores200-devtest 0.25176 5.9 1012 24721
58
+ kam-fra flores200-devtest 0.22149 3.8 1012 28343
59
+ kam-por flores200-devtest 0.22020 4.0 1012 26519
60
+ kam-spa flores200-devtest 0.22602 3.6 1012 29199
61
+ kbp-deu flores200-devtest 0.22271 2.3 1012 25094
62
+ kbp-eng flores200-devtest 0.26127 5.1 1012 24721
63
+ kbp-fra flores200-devtest 0.22128 3.4 1012 28343
64
+ kbp-por flores200-devtest 0.21208 3.1 1012 26519
65
+ kbp-spa flores200-devtest 0.22121 3.3 1012 29199
66
+ kik-deu flores200-devtest 0.27542 3.4 1012 25094
67
+ kik-eng flores200-devtest 0.33821 9.8 1012 24721
68
+ kik-fra flores200-devtest 0.29550 6.5 1012 28343
69
+ kik-por flores200-devtest 0.28161 5.5 1012 26519
70
+ kik-spa flores200-devtest 0.28074 5.1 1012 29199
71
+ kin-deu flores200-devtest 0.30940 6.3 1012 25094
72
+ kin-eng flores200-devtest 0.41964 18.1 1012 24721
73
+ kin-fra flores200-devtest 0.34887 10.7 1012 28343
74
+ kin-por flores200-devtest 0.32719 9.6 1012 26519
75
+ kin-spa flores200-devtest 0.31594 7.7 1012 29199
76
+ kmb-deu flores200-devtest 0.22967 2.2 1012 25094
77
+ kmb-eng flores200-devtest 0.25883 5.2 1012 24721
78
+ kmb-fra flores200-devtest 0.22897 3.9 1012 28343
79
+ kmb-por flores200-devtest 0.23387 4.5 1012 26519
80
+ kmb-spa flores200-devtest 0.24068 4.1 1012 29199
81
+ kon-deu flores200-devtest 0.29087 5.1 1012 25094
82
+ kon-eng flores200-devtest 0.34262 11.3 1012 24721
83
+ kon-fra flores200-devtest 0.31994 8.4 1012 28343
84
+ kon-por flores200-devtest 0.30547 7.8 1012 26519
85
+ kon-spa flores200-devtest 0.29534 6.4 1012 29199
86
+ lin-deu flores200-devtest 0.30708 6.4 1012 25094
87
+ lin-eng flores200-devtest 0.37728 14.2 1012 24721
88
+ lin-fra flores200-devtest 0.35052 11.0 1012 28343
89
+ lin-por flores200-devtest 0.33695 9.9 1012 26519
90
+ lin-spa flores200-devtest 0.31627 7.8 1012 29199
91
+ lua-deu flores200-devtest 0.23914 2.4 1012 25094
92
+ lua-eng flores200-devtest 0.26673 5.7 1012 24721
93
+ lua-fra flores200-devtest 0.23716 4.0 1012 28343
94
+ lua-por flores200-devtest 0.23412 4.0 1012 26519
95
+ lua-spa flores200-devtest 0.24285 3.9 1012 29199
96
+ lug-deu flores200-devtest 0.26004 4.3 1012 25094
97
+ lug-eng flores200-devtest 0.31805 10.6 1012 24721
98
+ lug-fra flores200-devtest 0.27167 6.5 1012 28343
99
+ lug-por flores200-devtest 0.25710 5.6 1012 26519
100
+ lug-spa flores200-devtest 0.25848 4.5 1012 29199
101
+ mos-deu flores200-devtest 0.19759 1.4 1012 25094
102
+ mos-eng flores200-devtest 0.23490 4.2 1012 24721
103
+ mos-fra flores200-devtest 0.19502 2.3 1012 28343
104
+ mos-por flores200-devtest 0.19102 2.6 1012 26519
105
+ mos-spa flores200-devtest 0.19918 2.4 1012 29199
106
+ nso-deu flores200-devtest 0.32125 7.4 1012 25094
107
+ nso-eng flores200-devtest 0.45662 22.3 1012 24721
108
+ nso-fra flores200-devtest 0.33732 10.5 1012 28343
109
+ nso-por flores200-devtest 0.31523 9.4 1012 26519
110
+ nso-spa flores200-devtest 0.29414 7.0 1012 29199
111
+ nya-deu flores200-devtest 0.30179 5.7 1012 25094
112
+ nya-eng flores200-devtest 0.39887 15.8 1012 24721
113
+ nya-fra flores200-devtest 0.33930 9.9 1012 28343
114
+ nya-por flores200-devtest 0.31301 8.6 1012 26519
115
+ nya-spa flores200-devtest 0.30478 7.2 1012 29199
116
+ run-deu flores200-devtest 0.31691 6.9 1012 25094
117
+ run-eng flores200-devtest 0.39846 15.5 1012 24721
118
+ run-fra flores200-devtest 0.34845 10.8 1012 28343
119
+ run-por flores200-devtest 0.33836 10.2 1012 26519
120
+ run-spa flores200-devtest 0.32116 8.0 1012 29199
121
+ sag-deu flores200-devtest 0.22221 2.4 1012 25094
122
+ sag-eng flores200-devtest 0.26423 5.4 1012 24721
123
+ sag-fra flores200-devtest 0.22575 3.8 1012 28343
124
+ sag-por flores200-devtest 0.21856 3.4 1012 26519
125
+ sag-spa flores200-devtest 0.22854 3.8 1012 29199
126
+ sna-deu flores200-devtest 0.32378 6.9 1012 25094
127
+ sna-eng flores200-devtest 0.41974 17.2 1012 24721
128
+ sna-fra flores200-devtest 0.36443 11.3 1012 28343
129
+ sna-por flores200-devtest 0.34260 10.2 1012 26519
130
+ sna-spa flores200-devtest 0.32210 8.0 1012 29199
131
+ sot-deu flores200-devtest 0.32197 6.9 1012 25094
132
+ sot-eng flores200-devtest 0.45415 20.7 1012 24721
133
+ sot-fra flores200-devtest 0.34608 10.7 1012 28343
134
+ sot-por flores200-devtest 0.32424 9.5 1012 26519
135
+ sot-spa flores200-devtest 0.30357 7.2 1012 29199
136
+ ssw-deu flores200-devtest 0.30791 5.8 1012 25094
137
+ ssw-eng flores200-devtest 0.39768 16.1 1012 24721
138
+ ssw-fra flores200-devtest 0.33603 9.8 1012 28343
139
+ ssw-por flores200-devtest 0.32413 9.2 1012 26519
140
+ ssw-spa flores200-devtest 0.30217 6.8 1012 29199
141
+ swh-deu flores200-devtest 0.38892 10.9 1012 25094
142
+ swh-eng flores200-devtest 0.54048 29.1 1012 24721
143
+ swh-fra flores200-devtest 0.44837 18.2 1012 28343
144
+ swh-por flores200-devtest 0.44062 17.6 1012 26519
145
+ swh-spa flores200-devtest 0.38855 11.6 1012 29199
146
+ tsn-deu flores200-devtest 0.31644 6.3 1012 25094
147
+ tsn-eng flores200-devtest 0.40410 15.3 1012 24721
148
+ tsn-fra flores200-devtest 0.34284 10.3 1012 28343
149
+ tsn-por flores200-devtest 0.33165 9.3 1012 26519
150
+ tsn-spa flores200-devtest 0.31377 7.7 1012 29199
151
+ tso-deu flores200-devtest 0.30073 6.3 1012 25094
152
+ tso-eng flores200-devtest 0.41504 17.6 1012 24721
153
+ tso-fra flores200-devtest 0.33502 10.1 1012 28343
154
+ tso-por flores200-devtest 0.30741 8.8 1012 26519
155
+ tso-spa flores200-devtest 0.29220 6.9 1012 29199
156
+ tum-deu flores200-devtest 0.27088 3.2 1012 25094
157
+ tum-eng flores200-devtest 0.31625 7.3 1012 24721
158
+ tum-fra flores200-devtest 0.28093 5.5 1012 28343
159
+ tum-por flores200-devtest 0.27543 5.3 1012 26519
160
+ tum-spa flores200-devtest 0.27894 4.9 1012 29199
161
+ twi-deu flores200-devtest 0.21184 2.4 1012 25094
162
+ twi-eng flores200-devtest 0.25960 5.9 1012 24721
163
+ twi-fra flores200-devtest 0.20738 3.1 1012 28343
164
+ twi-por flores200-devtest 0.20581 3.2 1012 26519
165
+ twi-spa flores200-devtest 0.21265 3.1 1012 29199
166
+ umb-deu flores200-devtest 0.22686 2.2 1012 25094
167
+ umb-eng flores200-devtest 0.24875 4.5 1012 24721
168
+ umb-fra flores200-devtest 0.22385 3.6 1012 28343
169
+ umb-por flores200-devtest 0.23949 4.0 1012 26519
170
+ umb-spa flores200-devtest 0.23796 3.5 1012 29199
171
+ wol-deu flores200-devtest 0.25851 3.1 1012 25094
172
+ wol-eng flores200-devtest 0.27455 6.8 1012 24721
173
+ wol-fra flores200-devtest 0.25659 5.0 1012 28343
174
+ wol-por flores200-devtest 0.25442 4.8 1012 26519
175
+ wol-spa flores200-devtest 0.26073 4.3 1012 29199
176
+ xho-deu flores200-devtest 0.34702 8.7 1012 25094
177
+ xho-eng flores200-devtest 0.47667 23.7 1012 24721
178
+ xho-fra flores200-devtest 0.39392 14.1 1012 28343
179
+ xho-por flores200-devtest 0.37032 12.3 1012 26519
180
+ xho-spa flores200-devtest 0.34071 8.9 1012 29199
181
+ yor-deu flores200-devtest 0.26350 3.8 1012 25094
182
+ yor-eng flores200-devtest 0.30560 8.5 1012 24721
183
+ yor-fra flores200-devtest 0.26611 5.3 1012 28343
184
+ yor-por flores200-devtest 0.26354 5.3 1012 26519
185
+ yor-spa flores200-devtest 0.26991 4.4 1012 29199
186
+ zul-deu flores200-devtest 0.34382 8.3 1012 25094
187
+ zul-eng flores200-devtest 0.47798 23.4 1012 24721
188
+ zul-fra flores200-devtest 0.39504 13.9 1012 28343
189
+ zul-por flores200-devtest 0.36947 12.3 1012 26519
190
+ zul-spa flores200-devtest 0.33858 9.0 1012 29199
191
+ bem-deu ntrex128 0.30672 5.9 1997 48761
192
+ bem-eng ntrex128 0.38068 14.9 1997 47673
193
+ bem-fra ntrex128 0.31600 8.2 1997 53481
194
+ bem-por ntrex128 0.31404 8.3 1997 51631
195
+ bem-spa ntrex128 0.33394 10.2 1997 54107
196
+ ewe-deu ntrex128 0.21779 3.0 1997 48761
197
+ ewe-eng ntrex128 0.26902 6.6 1997 47673
198
+ ewe-fra ntrex128 0.21230 3.6 1997 53481
199
+ ewe-por ntrex128 0.20871 3.5 1997 51631
200
+ ewe-spa ntrex128 0.22529 4.2 1997 54107
201
+ fuc-deu ntrex128 0.22996 2.9 1997 48761
202
+ fuc-eng ntrex128 0.26465 7.0 1997 47673
203
+ fuc-fra ntrex128 0.23921 4.4 1997 53481
204
+ fuc-por ntrex128 0.23450 4.5 1997 51631
205
+ fuc-spa ntrex128 0.25677 5.3 1997 54107
206
+ ibo-deu ntrex128 0.33283 7.3 1997 48761
207
+ ibo-eng ntrex128 0.42002 17.4 1997 47673
208
+ ibo-fra ntrex128 0.34993 10.3 1997 53481
209
+ ibo-por ntrex128 0.33898 9.2 1997 51631
210
+ ibo-spa ntrex128 0.36108 11.7 1997 54107
211
+ kin-deu ntrex128 0.33121 7.2 1997 48761
212
+ kin-eng ntrex128 0.42892 16.9 1997 47673
213
+ kin-fra ntrex128 0.35842 10.7 1997 53481
214
+ kin-por ntrex128 0.34399 10.0 1997 51631
215
+ kin-spa ntrex128 0.37224 12.5 1997 54107
216
+ nde-deu ntrex128 0.32255 6.0 1997 48761
217
+ nde-eng ntrex128 0.39640 13.8 1997 47673
218
+ nde-fra ntrex128 0.33712 8.6 1997 53481
219
+ nde-por ntrex128 0.32696 8.4 1997 51631
220
+ nde-spa ntrex128 0.34638 10.0 1997 54107
221
+ nso-deu ntrex128 0.32696 7.1 1997 48761
222
+ nso-eng ntrex128 0.42278 17.0 1997 47673
223
+ nso-fra ntrex128 0.34162 9.8 1997 53481
224
+ nso-por ntrex128 0.32895 9.2 1997 51631
225
+ nso-spa ntrex128 0.35027 11.3 1997 54107
226
+ nya-deu ntrex128 0.32971 7.6 1997 48761
227
+ nya-eng ntrex128 0.42702 19.2 1997 47673
228
+ nya-fra ntrex128 0.35503 10.7 1997 53481
229
+ nya-por ntrex128 0.33912 10.4 1997 51631
230
+ nya-spa ntrex128 0.36355 12.5 1997 54107
231
+ ssw-deu ntrex128 0.32656 6.6 1997 48761
232
+ ssw-eng ntrex128 0.43041 18.0 1997 47673
233
+ ssw-fra ntrex128 0.34293 9.8 1997 53481
234
+ ssw-por ntrex128 0.33562 9.3 1997 51631
235
+ ssw-spa ntrex128 0.35392 11.4 1997 54107
236
+ swa-deu ntrex128 0.39475 11.2 1997 48761
237
+ swa-eng ntrex128 0.54492 30.4 1997 47673
238
+ swa-fra ntrex128 0.43008 15.6 1997 53481
239
+ swa-por ntrex128 0.42343 15.4 1997 51631
240
+ swa-spa ntrex128 0.44892 18.9 1997 54107
241
+ tsn-deu ntrex128 0.34744 7.9 1997 48761
242
+ tsn-eng ntrex128 0.44944 20.1 1997 47673
243
+ tsn-fra ntrex128 0.36584 11.5 1997 53481
244
+ tsn-por ntrex128 0.35774 11.1 1997 51631
245
+ tsn-spa ntrex128 0.38149 13.8 1997 54107
246
+ ven-deu ntrex128 0.29846 5.0 1997 48761
247
+ ven-eng ntrex128 0.36848 12.7 1997 47673
248
+ ven-fra ntrex128 0.31439 8.0 1997 53481
249
+ ven-por ntrex128 0.30542 7.3 1997 51631
250
+ ven-spa ntrex128 0.32722 9.4 1997 54107
251
+ wol-deu ntrex128 0.25962 3.6 1997 48761
252
+ wol-eng ntrex128 0.27538 7.2 1997 47673
253
+ wol-fra ntrex128 0.24251 4.4 1997 53481
254
+ wol-por ntrex128 0.25264 4.7 1997 51631
255
+ wol-spa ntrex128 0.27877 6.0 1997 54107
256
+ xho-deu ntrex128 0.34694 8.2 1997 48761
257
+ xho-eng ntrex128 0.46636 21.8 1997 47673
258
+ xho-fra ntrex128 0.36905 11.6 1997 53481
259
+ xho-por ntrex128 0.35687 10.6 1997 51631
260
+ xho-spa ntrex128 0.38176 13.3 1997 54107
261
+ yor-deu ntrex128 0.28804 4.3 1997 48761
262
+ yor-eng ntrex128 0.34615 11.2 1997 47673
263
+ yor-fra ntrex128 0.29559 6.6 1997 53481
264
+ yor-por ntrex128 0.29273 6.1 1997 51631
265
+ yor-spa ntrex128 0.31894 7.8 1997 54107
266
+ zul-deu ntrex128 0.34049 8.4 1997 48761
267
+ zul-eng ntrex128 0.45848 21.9 1997 47673
268
+ zul-fra ntrex128 0.36203 11.4 1997 53481
269
+ zul-por ntrex128 0.35081 10.8 1997 51631
270
+ zul-spa ntrex128 0.37270 13.0 1997 54107
271
+ run-fra tatoeba-test-v2021-03-30 0.37924 18.4 1278 7496
272
+ xho-eng tatoeba-test-v2021-03-30 0.51250 36.4 222 1467
273
+ run-deu tatoeba-test-v2021-08-07 0.37181 19.1 1752 10562
274
+ run-eng tatoeba-test-v2021-08-07 0.49949 34.9 1703 10041
275
+ run-fra tatoeba-test-v2021-08-07 0.41431 22.4 1274 7479
276
+ run-spa tatoeba-test-v2021-08-07 0.39311 19.3 963 5167
277
+ swa-eng tatoeba-test-v2021-08-07 0.57031 41.5 387 2508
278
+ kin-eng tico19-test 0.16333 1.1 2100 56323
279
+ kin-fra tico19-test 0.20797 2.3 2100 64661
280
+ kin-por tico19-test 0.20686 1.9 2100 62729
281
+ kin-spa tico19-test 0.20848 2.0 2100 66563
282
+ lin-fra tico19-test 0.20671 2.3 2100 64661
283
+ lin-por tico19-test 0.20516 2.0 2100 62729
284
+ lin-spa tico19-test 0.20385 2.0 2100 66563
285
+ lug-eng tico19-test 0.15561 1.1 2100 56323
286
+ lug-fra tico19-test 0.19865 1.8 2100 64661
287
+ lug-por tico19-test 0.20548 2.0 2100 62729
288
+ lug-spa tico19-test 0.20545 1.9 2100 66563
289
+ swa-eng tico19-test 0.20835 2.0 2100 56315
290
+ swa-fra tico19-test 0.25275 3.3 2100 64661
291
+ swa-por tico19-test 0.23967 2.1 2100 62729
292
+ swa-spa tico19-test 0.24260 2.1 2100 66563
293
+ zul-eng tico19-test 0.48762 25.2 2100 56804
294
+ zul-fra tico19-test 0.37823 13.7 2100 64661
295
+ zul-por tico19-test 0.38586 14.0 2100 62729
296
+ zul-spa tico19-test 0.40041 15.9 2100 66563
benchmark_translations.zip ADDED
File without changes
config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "pytorch-models/opus-mt-tc-bible-big-alv-deu_eng_fra_por_spa",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "relu",
5
+ "architectures": [
6
+ "MarianMTModel"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "bos_token_id": 0,
10
+ "classifier_dropout": 0.0,
11
+ "d_model": 1024,
12
+ "decoder_attention_heads": 16,
13
+ "decoder_ffn_dim": 4096,
14
+ "decoder_layerdrop": 0.0,
15
+ "decoder_layers": 6,
16
+ "decoder_start_token_id": 61871,
17
+ "decoder_vocab_size": 61872,
18
+ "dropout": 0.1,
19
+ "encoder_attention_heads": 16,
20
+ "encoder_ffn_dim": 4096,
21
+ "encoder_layerdrop": 0.0,
22
+ "encoder_layers": 6,
23
+ "eos_token_id": 236,
24
+ "forced_eos_token_id": null,
25
+ "init_std": 0.02,
26
+ "is_encoder_decoder": true,
27
+ "max_length": null,
28
+ "max_position_embeddings": 1024,
29
+ "model_type": "marian",
30
+ "normalize_embedding": false,
31
+ "num_beams": null,
32
+ "num_hidden_layers": 6,
33
+ "pad_token_id": 61871,
34
+ "scale_embedding": true,
35
+ "share_encoder_decoder_embeddings": true,
36
+ "static_position_embeddings": true,
37
+ "torch_dtype": "float32",
38
+ "transformers_version": "4.45.1",
39
+ "use_cache": true,
40
+ "vocab_size": 61872
41
+ }
generation_config.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bad_words_ids": [
4
+ [
5
+ 61871
6
+ ]
7
+ ],
8
+ "bos_token_id": 0,
9
+ "decoder_start_token_id": 61871,
10
+ "eos_token_id": 236,
11
+ "forced_eos_token_id": 236,
12
+ "max_length": 512,
13
+ "num_beams": 4,
14
+ "pad_token_id": 61871,
15
+ "transformers_version": "4.45.1"
16
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73e39e03d4e581b629f63916c16b6f34defc47cba7fac40280604fe84e035a08
3
+ size 959134320
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f26fd801c46459a211936d1d2691005255ff1dd30f128e69c394c671ed01abc
3
+ size 959185541
source.spm ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:512274fc5f10a67cc547763ad2f499e5ecaa9d055e5f4887256142b25621b951
3
+ size 760897
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
target.spm ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc935647403de50358df1994b1b8e4687268ae34ffb201a8a4526f3a505635d2
3
+ size 820799
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"source_lang": "alv", "target_lang": "deu+eng+fra+por+spa", "unk_token": "<unk>", "eos_token": "</s>", "pad_token": "<pad>", "model_max_length": 512, "sp_model_kwargs": {}, "separate_vocabs": false, "special_tokens_map_file": null, "name_or_path": "marian-models/opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-30/alv-deu+eng+fra+por+spa", "tokenizer_class": "MarianTokenizer"}
vocab.json ADDED
The diff for this file is too large to render. See raw diff