system HF staff commited on
Commit
b409427
·
1 Parent(s): 8e9b7ee

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +105 -0
README.md ADDED
@@ -0,0 +1,105 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: en
3
+ tags:
4
+ - translation
5
+
6
+ license: apache-2.0
7
+ ---
8
+
9
+ ### eng-trk
10
+
11
+ * source group: English
12
+ * target group: Turkic languages
13
+ * OPUS readme: [eng-trk](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/eng-trk/README.md)
14
+
15
+ * model: transformer
16
+ * source language(s): eng
17
+ * target language(s): aze_Latn bak chv crh crh_Latn kaz_Cyrl kaz_Latn kir_Cyrl kjh kum ota_Arab ota_Latn sah tat tat_Arab tat_Latn tuk tuk_Latn tur tyv uig_Arab uig_Cyrl uzb_Cyrl uzb_Latn
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-trk/opus2m-2020-08-01.zip)
22
+ * test set translations: [opus2m-2020-08-01.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/eng-trk/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-trk/opus2m-2020-08-01.eval.txt)
24
+
25
+ ## Benchmarks
26
+
27
+ | testset | BLEU | chr-F |
28
+ |-----------------------|-------|-------|
29
+ | newsdev2016-entr-engtur.eng.tur | 10.1 | 0.437 |
30
+ | newstest2016-entr-engtur.eng.tur | 9.2 | 0.410 |
31
+ | newstest2017-entr-engtur.eng.tur | 9.0 | 0.410 |
32
+ | newstest2018-entr-engtur.eng.tur | 9.2 | 0.413 |
33
+ | Tatoeba-test.eng-aze.eng.aze | 26.8 | 0.577 |
34
+ | Tatoeba-test.eng-bak.eng.bak | 7.6 | 0.308 |
35
+ | Tatoeba-test.eng-chv.eng.chv | 4.3 | 0.270 |
36
+ | Tatoeba-test.eng-crh.eng.crh | 8.1 | 0.330 |
37
+ | Tatoeba-test.eng-kaz.eng.kaz | 11.1 | 0.359 |
38
+ | Tatoeba-test.eng-kir.eng.kir | 28.6 | 0.524 |
39
+ | Tatoeba-test.eng-kjh.eng.kjh | 1.0 | 0.041 |
40
+ | Tatoeba-test.eng-kum.eng.kum | 2.2 | 0.075 |
41
+ | Tatoeba-test.eng.multi | 19.9 | 0.455 |
42
+ | Tatoeba-test.eng-ota.eng.ota | 0.5 | 0.065 |
43
+ | Tatoeba-test.eng-sah.eng.sah | 0.7 | 0.030 |
44
+ | Tatoeba-test.eng-tat.eng.tat | 9.7 | 0.316 |
45
+ | Tatoeba-test.eng-tuk.eng.tuk | 5.9 | 0.317 |
46
+ | Tatoeba-test.eng-tur.eng.tur | 34.6 | 0.623 |
47
+ | Tatoeba-test.eng-tyv.eng.tyv | 5.4 | 0.210 |
48
+ | Tatoeba-test.eng-uig.eng.uig | 0.1 | 0.155 |
49
+ | Tatoeba-test.eng-uzb.eng.uzb | 3.4 | 0.275 |
50
+
51
+
52
+ ### System Info:
53
+ - hf_name: eng-trk
54
+
55
+ - source_languages: eng
56
+
57
+ - target_languages: trk
58
+
59
+ - opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/eng-trk/README.md
60
+
61
+ - original_repo: Tatoeba-Challenge
62
+
63
+ - tags: ['translation']
64
+
65
+ - prepro: normalization + SentencePiece (spm32k,spm32k)
66
+
67
+ - url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/eng-trk/opus2m-2020-08-01.zip
68
+
69
+ - url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/eng-trk/opus2m-2020-08-01.test.txt
70
+
71
+ - src_alpha3: eng
72
+
73
+ - tgt_alpha3: trk
74
+
75
+ - short_pair: en-trk
76
+
77
+ - chrF2_score: 0.455
78
+
79
+ - bleu: 19.9
80
+
81
+ - brevity_penalty: 1.0
82
+
83
+ - ref_len: 57072.0
84
+
85
+ - src_name: English
86
+
87
+ - tgt_name: Turkic languages
88
+
89
+ - train_date: 2020-08-01
90
+
91
+ - src_alpha2: en
92
+
93
+ - tgt_alpha2: trk
94
+
95
+ - prefer_old: False
96
+
97
+ - long_pair: eng-trk
98
+
99
+ - helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535
100
+
101
+ - transformers_git_sha: 46e9f53347bbe9e989f0335f98465f30886d8173
102
+
103
+ - port_machine: brutasse
104
+
105
+ - port_time: 2020-08-18-01:48