Update README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: sem
|
| 3 |
+
tags:
|
| 4 |
+
- translation
|
| 5 |
+
|
| 6 |
+
license: apache-2.0
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
### sem-sem
|
| 10 |
+
|
| 11 |
+
* source languages: sem
|
| 12 |
+
* target languages: sem
|
| 13 |
+
* OPUS readme: [sem-sem](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/sem-sem/README.md)
|
| 14 |
+
|
| 15 |
+
* dataset: opus
|
| 16 |
+
* model: transformer
|
| 17 |
+
* source language(s): apc ara arq arz heb mlt
|
| 18 |
+
* target language(s): apc ara arq arz heb mlt
|
| 19 |
+
* model: transformer
|
| 20 |
+
* pre-processing: normalization + SentencePiece (spm32k,spm32k)
|
| 21 |
+
* a sentence initial language token is required in the form of `>>id<<` (id = valid target language ID)
|
| 22 |
+
* download original weights: [opus-2020-07-27.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/sem-sem/opus-2020-07-27.zip)
|
| 23 |
+
* test set translations: [opus-2020-07-27.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/sem-sem/opus-2020-07-27.test.txt)
|
| 24 |
+
* test set scores: [opus-2020-07-27.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/sem-sem/opus-2020-07-27.eval.txt)
|
| 25 |
+
|
| 26 |
+
## Benchmarks
|
| 27 |
+
|
| 28 |
+
| testset | BLEU | chr-F |
|
| 29 |
+
|-----------------------|-------|-------|
|
| 30 |
+
| Tatoeba-test.ara-ara.ara.ara | 4.2 | 0.200 |
|
| 31 |
+
| Tatoeba-test.ara-heb.ara.heb | 34.0 | 0.542 |
|
| 32 |
+
| Tatoeba-test.ara-mlt.ara.mlt | 16.6 | 0.513 |
|
| 33 |
+
| Tatoeba-test.heb-ara.heb.ara | 18.8 | 0.477 |
|
| 34 |
+
| Tatoeba-test.mlt-ara.mlt.ara | 20.7 | 0.388 |
|
| 35 |
+
| Tatoeba-test.multi.multi | 27.1 | 0.507 |
|
| 36 |
+
|