system HF staff commited on
Commit
b00f6e6
1 Parent(s): 6889f0c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +107 -0
README.md ADDED
@@ -0,0 +1,107 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: inc
3
+ tags:
4
+ - translation
5
+
6
+ license: apache-2.0
7
+ ---
8
+
9
+ ### inc-eng
10
+
11
+ * source group: Indic languages
12
+ * target group: English
13
+ * OPUS readme: [inc-eng](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/inc-eng/README.md)
14
+
15
+ * model: transformer
16
+ * source language(s): asm awa ben bho gom guj hif_Latn hin mai mar npi ori pan_Guru pnb rom san_Deva sin snd_Arab urd
17
+ * target language(s): eng
18
+ * model: transformer
19
+ * pre-processing: normalization + SentencePiece (spm32k,spm32k)
20
+ * download original weights: [opus2m-2020-08-01.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/inc-eng/opus2m-2020-08-01.zip)
21
+ * test set translations: [opus2m-2020-08-01.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/inc-eng/opus2m-2020-08-01.test.txt)
22
+ * test set scores: [opus2m-2020-08-01.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/inc-eng/opus2m-2020-08-01.eval.txt)
23
+
24
+ ## Benchmarks
25
+
26
+ | testset | BLEU | chr-F |
27
+ |-----------------------|-------|-------|
28
+ | newsdev2014-hineng.hin.eng | 8.9 | 0.341 |
29
+ | newsdev2019-engu-gujeng.guj.eng | 8.7 | 0.321 |
30
+ | newstest2014-hien-hineng.hin.eng | 13.1 | 0.396 |
31
+ | newstest2019-guen-gujeng.guj.eng | 6.5 | 0.290 |
32
+ | Tatoeba-test.asm-eng.asm.eng | 18.1 | 0.363 |
33
+ | Tatoeba-test.awa-eng.awa.eng | 6.2 | 0.222 |
34
+ | Tatoeba-test.ben-eng.ben.eng | 44.7 | 0.595 |
35
+ | Tatoeba-test.bho-eng.bho.eng | 29.4 | 0.458 |
36
+ | Tatoeba-test.guj-eng.guj.eng | 19.3 | 0.383 |
37
+ | Tatoeba-test.hif-eng.hif.eng | 3.7 | 0.220 |
38
+ | Tatoeba-test.hin-eng.hin.eng | 38.6 | 0.564 |
39
+ | Tatoeba-test.kok-eng.kok.eng | 6.6 | 0.287 |
40
+ | Tatoeba-test.lah-eng.lah.eng | 16.0 | 0.272 |
41
+ | Tatoeba-test.mai-eng.mai.eng | 75.6 | 0.796 |
42
+ | Tatoeba-test.mar-eng.mar.eng | 25.9 | 0.497 |
43
+ | Tatoeba-test.multi.eng | 29.0 | 0.502 |
44
+ | Tatoeba-test.nep-eng.nep.eng | 4.5 | 0.198 |
45
+ | Tatoeba-test.ori-eng.ori.eng | 5.0 | 0.226 |
46
+ | Tatoeba-test.pan-eng.pan.eng | 17.4 | 0.375 |
47
+ | Tatoeba-test.rom-eng.rom.eng | 1.7 | 0.174 |
48
+ | Tatoeba-test.san-eng.san.eng | 5.0 | 0.173 |
49
+ | Tatoeba-test.sin-eng.sin.eng | 31.2 | 0.511 |
50
+ | Tatoeba-test.snd-eng.snd.eng | 45.7 | 0.670 |
51
+ | Tatoeba-test.urd-eng.urd.eng | 25.6 | 0.456 |
52
+
53
+
54
+ ### System Info:
55
+ - hf_name: inc-eng
56
+
57
+ - source_languages: inc
58
+
59
+ - target_languages: eng
60
+
61
+ - opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/inc-eng/README.md
62
+
63
+ - original_repo: Tatoeba-Challenge
64
+
65
+ - tags: ['translation']
66
+
67
+ - prepro: normalization + SentencePiece (spm32k,spm32k)
68
+
69
+ - url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/inc-eng/opus2m-2020-08-01.zip
70
+
71
+ - url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/inc-eng/opus2m-2020-08-01.test.txt
72
+
73
+ - src_alpha3: inc
74
+
75
+ - tgt_alpha3: eng
76
+
77
+ - short_pair: inc-en
78
+
79
+ - chrF2_score: 0.502
80
+
81
+ - bleu: 29.0
82
+
83
+ - brevity_penalty: 1.0
84
+
85
+ - ref_len: 64706.0
86
+
87
+ - src_name: Indic languages
88
+
89
+ - tgt_name: English
90
+
91
+ - train_date: 2020-08-01
92
+
93
+ - src_alpha2: inc
94
+
95
+ - tgt_alpha2: en
96
+
97
+ - prefer_old: False
98
+
99
+ - long_pair: inc-eng
100
+
101
+ - helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535
102
+
103
+ - transformers_git_sha: 46e9f53347bbe9e989f0335f98465f30886d8173
104
+
105
+ - port_machine: brutasse
106
+
107
+ - port_time: 2020-08-18-01:48