system HF staff commited on
Commit
72d0831
1 Parent(s): d99c9c7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +92 -0
README.md ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: bat
3
+ tags:
4
+ - translation
5
+
6
+ license: apache-2.0
7
+ ---
8
+
9
+ ### bat-eng
10
+
11
+ * source group: Baltic languages
12
+ * target group: English
13
+ * OPUS readme: [bat-eng](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/bat-eng/README.md)
14
+
15
+ * model: transformer
16
+ * source language(s): lav lit ltg prg_Latn sgs
17
+ * target language(s): eng
18
+ * model: transformer
19
+ * pre-processing: normalization + SentencePiece (spm32k,spm32k)
20
+ * download original weights: [opus2m-2020-07-31.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/bat-eng/opus2m-2020-07-31.zip)
21
+ * test set translations: [opus2m-2020-07-31.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/bat-eng/opus2m-2020-07-31.test.txt)
22
+ * test set scores: [opus2m-2020-07-31.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/bat-eng/opus2m-2020-07-31.eval.txt)
23
+
24
+ ## Benchmarks
25
+
26
+ | testset | BLEU | chr-F |
27
+ |-----------------------|-------|-------|
28
+ | newsdev2017-enlv-laveng.lav.eng | 27.5 | 0.566 |
29
+ | newsdev2019-enlt-liteng.lit.eng | 27.8 | 0.557 |
30
+ | newstest2017-enlv-laveng.lav.eng | 21.1 | 0.512 |
31
+ | newstest2019-lten-liteng.lit.eng | 30.2 | 0.592 |
32
+ | Tatoeba-test.lav-eng.lav.eng | 51.5 | 0.687 |
33
+ | Tatoeba-test.lit-eng.lit.eng | 55.1 | 0.703 |
34
+ | Tatoeba-test.multi.eng | 50.6 | 0.662 |
35
+ | Tatoeba-test.prg-eng.prg.eng | 1.0 | 0.159 |
36
+ | Tatoeba-test.sgs-eng.sgs.eng | 16.5 | 0.265 |
37
+
38
+
39
+ ### System Info:
40
+ - hf_name: bat-eng
41
+
42
+ - source_languages: bat
43
+
44
+ - target_languages: eng
45
+
46
+ - opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/bat-eng/README.md
47
+
48
+ - original_repo: Tatoeba-Challenge
49
+
50
+ - tags: ['translation']
51
+
52
+ - prepro: normalization + SentencePiece (spm32k,spm32k)
53
+
54
+ - url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/bat-eng/opus2m-2020-07-31.zip
55
+
56
+ - url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/bat-eng/opus2m-2020-07-31.test.txt
57
+
58
+ - src_alpha3: bat
59
+
60
+ - tgt_alpha3: eng
61
+
62
+ - short_pair: bat-en
63
+
64
+ - chrF2_score: 0.662
65
+
66
+ - bleu: 50.6
67
+
68
+ - brevity_penalty: 0.9890000000000001
69
+
70
+ - ref_len: 30772.0
71
+
72
+ - src_name: Baltic languages
73
+
74
+ - tgt_name: English
75
+
76
+ - train_date: 2020-07-31
77
+
78
+ - src_alpha2: bat
79
+
80
+ - tgt_alpha2: en
81
+
82
+ - prefer_old: False
83
+
84
+ - long_pair: bat-eng
85
+
86
+ - helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535
87
+
88
+ - transformers_git_sha: 46e9f53347bbe9e989f0335f98465f30886d8173
89
+
90
+ - port_machine: brutasse
91
+
92
+ - port_time: 2020-08-18-01:48