system HF staff commited on
Commit
f7b9177
1 Parent(s): 0415646

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +97 -0
README.md ADDED
@@ -0,0 +1,97 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: en
3
+ tags:
4
+ - translation
5
+
6
+ license: apache-2.0
7
+ ---
8
+
9
+ ### eng-zle
10
+
11
+ * source group: English
12
+ * target group: East Slavic languages
13
+ * OPUS readme: [eng-zle](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/eng-zle/README.md)
14
+
15
+ * model: transformer
16
+ * source language(s): eng
17
+ * target language(s): bel bel_Latn orv_Cyrl rue rus ukr
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-2020-08-02.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/eng-zle/opus2m-2020-08-02.zip)
22
+ * test set translations: [opus2m-2020-08-02.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/eng-zle/opus2m-2020-08-02.test.txt)
23
+ * test set scores: [opus2m-2020-08-02.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/eng-zle/opus2m-2020-08-02.eval.txt)
24
+
25
+ ## Benchmarks
26
+
27
+ | testset | BLEU | chr-F |
28
+ |-----------------------|-------|-------|
29
+ | newstest2012-engrus.eng.rus | 27.4 | 0.550 |
30
+ | newstest2013-engrus.eng.rus | 21.4 | 0.493 |
31
+ | newstest2015-enru-engrus.eng.rus | 24.2 | 0.534 |
32
+ | newstest2016-enru-engrus.eng.rus | 23.3 | 0.518 |
33
+ | newstest2017-enru-engrus.eng.rus | 25.3 | 0.541 |
34
+ | newstest2018-enru-engrus.eng.rus | 22.4 | 0.527 |
35
+ | newstest2019-enru-engrus.eng.rus | 24.1 | 0.505 |
36
+ | Tatoeba-test.eng-bel.eng.bel | 20.8 | 0.471 |
37
+ | Tatoeba-test.eng.multi | 37.2 | 0.580 |
38
+ | Tatoeba-test.eng-orv.eng.orv | 0.6 | 0.130 |
39
+ | Tatoeba-test.eng-rue.eng.rue | 1.4 | 0.168 |
40
+ | Tatoeba-test.eng-rus.eng.rus | 41.3 | 0.616 |
41
+ | Tatoeba-test.eng-ukr.eng.ukr | 38.7 | 0.596 |
42
+
43
+
44
+ ### System Info:
45
+ - hf_name: eng-zle
46
+
47
+ - source_languages: eng
48
+
49
+ - target_languages: zle
50
+
51
+ - opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/eng-zle/README.md
52
+
53
+ - original_repo: Tatoeba-Challenge
54
+
55
+ - tags: ['translation']
56
+
57
+ - prepro: normalization + SentencePiece (spm32k,spm32k)
58
+
59
+ - url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/eng-zle/opus2m-2020-08-02.zip
60
+
61
+ - url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/eng-zle/opus2m-2020-08-02.test.txt
62
+
63
+ - src_alpha3: eng
64
+
65
+ - tgt_alpha3: zle
66
+
67
+ - short_pair: en-zle
68
+
69
+ - chrF2_score: 0.58
70
+
71
+ - bleu: 37.2
72
+
73
+ - brevity_penalty: 0.9890000000000001
74
+
75
+ - ref_len: 63493.0
76
+
77
+ - src_name: English
78
+
79
+ - tgt_name: East Slavic languages
80
+
81
+ - train_date: 2020-08-02
82
+
83
+ - src_alpha2: en
84
+
85
+ - tgt_alpha2: zle
86
+
87
+ - prefer_old: False
88
+
89
+ - long_pair: eng-zle
90
+
91
+ - helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535
92
+
93
+ - transformers_git_sha: 46e9f53347bbe9e989f0335f98465f30886d8173
94
+
95
+ - port_machine: brutasse
96
+
97
+ - port_time: 2020-08-18-01:48