Add multilingual to the language tag

#1
by lbourdois - opened
Files changed (1) hide show
  1. README.md +69 -144
README.md CHANGED
@@ -7,168 +7,93 @@ language:
7
  - sh
8
  - sl
9
  - sr
10
- language_bcp47:
11
- - sr_Cyrl
12
- - sr_Latn
13
-
14
  tags:
15
  - translation
16
  - opus-mt-tc
17
-
18
- license: cc-by-4.0
 
19
  model-index:
20
  - name: opus-mt-tc-big-zls-de
21
  results:
22
  - task:
23
- name: Translation bul-deu
24
  type: translation
25
- args: bul-deu
26
  dataset:
27
  name: flores101-devtest
28
  type: flores_101
29
  args: bul deu devtest
30
  metrics:
31
- - name: BLEU
32
- type: bleu
33
- value: 28.4
34
- - name: chr-F
35
- type: chrf
36
- value: 0.57688
37
- - task:
38
- name: Translation hrv-deu
39
- type: translation
40
- args: hrv-deu
41
- dataset:
42
- name: flores101-devtest
43
- type: flores_101
44
- args: hrv deu devtest
45
- metrics:
46
- - name: BLEU
47
- type: bleu
48
- value: 27.4
49
- - name: chr-F
50
- type: chrf
51
- value: 0.56674
52
- - task:
53
- name: Translation mkd-deu
54
- type: translation
55
- args: mkd-deu
56
- dataset:
57
- name: flores101-devtest
58
- type: flores_101
59
- args: mkd deu devtest
60
- metrics:
61
- - name: BLEU
62
- type: bleu
63
- value: 29.3
64
- - name: chr-F
65
- type: chrf
66
- value: 0.57688
67
- - task:
68
- name: Translation slv-deu
69
- type: translation
70
- args: slv-deu
71
- dataset:
72
- name: flores101-devtest
73
- type: flores_101
74
- args: slv deu devtest
75
- metrics:
76
- - name: BLEU
77
- type: bleu
78
- value: 26.7
79
- - name: chr-F
80
- type: chrf
81
- value: 0.56258
82
  - task:
83
- name: Translation srp_Cyrl-deu
84
  type: translation
85
- args: srp_Cyrl-deu
86
- dataset:
87
- name: flores101-devtest
88
- type: flores_101
89
- args: srp_Cyrl deu devtest
90
- metrics:
91
- - name: BLEU
92
- type: bleu
93
- value: 30.7
94
- - name: chr-F
95
- type: chrf
96
- value: 0.59271
97
- - task:
98
  name: Translation bul-deu
99
- type: translation
100
- args: bul-deu
101
  dataset:
102
  name: tatoeba-test-v2021-08-07
103
  type: tatoeba_mt
104
  args: bul-deu
105
  metrics:
106
- - name: BLEU
107
- type: bleu
108
- value: 54.5
109
- - name: chr-F
110
- type: chrf
111
- value: 0.71220
112
- - task:
113
- name: Translation hbs-deu
114
- type: translation
115
- args: hbs-deu
116
- dataset:
117
- name: tatoeba-test-v2021-08-07
118
- type: tatoeba_mt
119
- args: hbs-deu
120
- metrics:
121
- - name: BLEU
122
- type: bleu
123
- value: 54.8
124
- - name: chr-F
125
- type: chrf
126
- value: 0.71283
127
- - task:
128
- name: Translation hrv-deu
129
- type: translation
130
- args: hrv-deu
131
- dataset:
132
- name: tatoeba-test-v2021-08-07
133
- type: tatoeba_mt
134
- args: hrv-deu
135
- metrics:
136
- - name: BLEU
137
- type: bleu
138
- value: 53.1
139
- - name: chr-F
140
- type: chrf
141
- value: 0.69448
142
- - task:
143
- name: Translation slv-deu
144
- type: translation
145
- args: slv-deu
146
- dataset:
147
- name: tatoeba-test-v2021-08-07
148
- type: tatoeba_mt
149
- args: slv-deu
150
- metrics:
151
- - name: BLEU
152
- type: bleu
153
- value: 21.1
154
- - name: chr-F
155
- type: chrf
156
- value: 0.36339
157
- - task:
158
- name: Translation srp_Latn-deu
159
- type: translation
160
- args: srp_Latn-deu
161
- dataset:
162
- name: tatoeba-test-v2021-08-07
163
- type: tatoeba_mt
164
- args: srp_Latn-deu
165
- metrics:
166
- - name: BLEU
167
- type: bleu
168
- value: 56.0
169
- - name: chr-F
170
- type: chrf
171
- value: 0.72489
172
  ---
173
  # opus-mt-tc-big-zls-de
174
 
@@ -236,7 +161,7 @@ for t in translated:
236
 
237
  # expected output:
238
  # Sind Sie Student?
239
- # Zwei Dinge sollten Kinder von ihren Eltern bekommen: Wurzeln und Flügel.
240
  ```
241
 
242
  You can also use OPUS-MT models with the transformers pipelines, for example:
@@ -279,7 +204,7 @@ print(pipe("Jesi li ti student?"))
279
 
280
  ## Citation Information
281
 
282
- * Publications: [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.)
283
 
284
  ```
285
  @inproceedings{tiedemann-thottingal-2020-opus,
@@ -309,7 +234,7 @@ print(pipe("Jesi li ti student?"))
309
 
310
  ## Acknowledgements
311
 
312
- The work is supported by the [European Language Grid](https://www.european-language-grid.eu/) as [pilot project 2866](https://live.european-language-grid.eu/catalogue/#/resource/projects/2866), by the [FoTran project](https://www.helsinki.fi/en/researchgroups/natural-language-understanding-with-cross-lingual-grounding), funded by the European Research Council (ERC) under the European Unions Horizon 2020 research and innovation programme (grant agreement No 771113), and the [MeMAD project](https://memad.eu/), funded by the European Unions Horizon 2020 Research and Innovation Programme under grant agreement No 780069. We are also grateful for the generous computational resources and IT infrastructure provided by [CSC -- IT Center for Science](https://www.csc.fi/), Finland.
313
 
314
  ## Model conversion info
315
 
 
7
  - sh
8
  - sl
9
  - sr
10
+ - multilingual
11
+ license: cc-by-4.0
 
 
12
  tags:
13
  - translation
14
  - opus-mt-tc
15
+ language_bcp47:
16
+ - sr_Cyrl
17
+ - sr_Latn
18
  model-index:
19
  - name: opus-mt-tc-big-zls-de
20
  results:
21
  - task:
 
22
  type: translation
23
+ name: Translation bul-deu
24
  dataset:
25
  name: flores101-devtest
26
  type: flores_101
27
  args: bul deu devtest
28
  metrics:
29
+ - type: bleu
30
+ value: 28.4
31
+ name: BLEU
32
+ - type: chrf
33
+ value: 0.57688
34
+ name: chr-F
35
+ - type: bleu
36
+ value: 27.4
37
+ name: BLEU
38
+ - type: chrf
39
+ value: 0.56674
40
+ name: chr-F
41
+ - type: bleu
42
+ value: 29.3
43
+ name: BLEU
44
+ - type: chrf
45
+ value: 0.57688
46
+ name: chr-F
47
+ - type: bleu
48
+ value: 26.7
49
+ name: BLEU
50
+ - type: chrf
51
+ value: 0.56258
52
+ name: chr-F
53
+ - type: bleu
54
+ value: 30.7
55
+ name: BLEU
56
+ - type: chrf
57
+ value: 0.59271
58
+ name: chr-F
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
59
  - task:
 
60
  type: translation
 
 
 
 
 
 
 
 
 
 
 
 
 
61
  name: Translation bul-deu
 
 
62
  dataset:
63
  name: tatoeba-test-v2021-08-07
64
  type: tatoeba_mt
65
  args: bul-deu
66
  metrics:
67
+ - type: bleu
68
+ value: 54.5
69
+ name: BLEU
70
+ - type: chrf
71
+ value: 0.7122
72
+ name: chr-F
73
+ - type: bleu
74
+ value: 54.8
75
+ name: BLEU
76
+ - type: chrf
77
+ value: 0.71283
78
+ name: chr-F
79
+ - type: bleu
80
+ value: 53.1
81
+ name: BLEU
82
+ - type: chrf
83
+ value: 0.69448
84
+ name: chr-F
85
+ - type: bleu
86
+ value: 21.1
87
+ name: BLEU
88
+ - type: chrf
89
+ value: 0.36339
90
+ name: chr-F
91
+ - type: bleu
92
+ value: 56.0
93
+ name: BLEU
94
+ - type: chrf
95
+ value: 0.72489
96
+ name: chr-F
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
97
  ---
98
  # opus-mt-tc-big-zls-de
99
 
 
161
 
162
  # expected output:
163
  # Sind Sie Student?
164
+ # Zwei Dinge sollten Kinder von ihren Eltern bekommen: Wurzeln und Fl�gel.
165
  ```
166
 
167
  You can also use OPUS-MT models with the transformers pipelines, for example:
 
204
 
205
  ## Citation Information
206
 
207
+ * Publications: [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.)
208
 
209
  ```
210
  @inproceedings{tiedemann-thottingal-2020-opus,
 
234
 
235
  ## Acknowledgements
236
 
237
+ The work is supported by the [European Language Grid](https://www.european-language-grid.eu/) as [pilot project 2866](https://live.european-language-grid.eu/catalogue/#/resource/projects/2866), by the [FoTran project](https://www.helsinki.fi/en/researchgroups/natural-language-understanding-with-cross-lingual-grounding), funded by the European Research Council (ERC) under the European Unions Horizon 2020 research and innovation programme (grant agreement No 771113), and the [MeMAD project](https://memad.eu/), funded by the European Unions Horizon 2020 Research and Innovation Programme under grant agreement No 780069. We are also grateful for the generous computational resources and IT infrastructure provided by [CSC -- IT Center for Science](https://www.csc.fi/), Finland.
238
 
239
  ## Model conversion info
240