mirror of
https://github.com/Helsinki-NLP/OPUS-MT-train.git
synced 2024-09-11 20:27:19 +03:00
| .. | ||
| README.md | ||
opus-2019-12-18.zip
- dataset: opus
- model: transformer-align
- pre-processing: normalization + SentencePiece
- download: opus-2019-12-18.zip
- test set translations: opus-2019-12-18.test.txt
- test set scores: opus-2019-12-18.eval.txt
Benchmarks
| testset | BLEU | chr-F |
|---|---|---|
| Tatoeba.mr.en | 38.2 | 0.515 |
opus+bt-2020-05-23.zip
- dataset: opus+bt
- model: transformer-align
- source language(s): mr
- target language(s): en
- model: transformer-align
- pre-processing: normalization + SentencePiece (spm4k,spm4k)
- download: opus+bt-2020-05-23.zip
- test set translations: opus+bt-2020-05-23.test.txt
- test set scores: opus+bt-2020-05-23.eval.txt
Training data: opus+bt
- mr-en: bible-uedin (60446) GNOME (86) JW300 (245434) KDE4 (5356) QED (1541) Ubuntu (2354) wikimedia (1596)
- mr-en: total size = 316813
- unused dev/test data is added to training data
- total size (opus+bt): 355237
Validation data
-
en-mr: Tatoeba
-
devset = top 5000 lines of Tatoeba.src.shuffled!
-
testset = next 5000 lines of Tatoeba.src.shuffled!
-
remaining lines are added to traindata
Benchmarks
| testset | BLEU | chr-F |
|---|---|---|
| Tatoeba.mr.en | 47.8 | 0.582 |