metadata
license: apache-2.0
datasets:
- Helsinki-NLP/tatoeba
- openlanguagedata/flores_plus
language:
- en
- ar
metrics:
- bleu
- comet
- chrf
pipeline_tag: translation
OPUS-MT-tiny-ara-eng
Distilled model from a Tatoeba-MT Teacher: OPUS-MT-models/ar-en/opus-2019-12-18, which has been trained on the Tatoeba dataset.
We used the OpusDistillery to train new a new student with the tiny architecture, with a regular transformer decoder. For training data, we used Tatoeba. The configuration file fed into OpusDistillery can be found here.
How to run
from transformers import MarianMTModel, MarianTokenizer
model_name = "Helsinki-NLP/opus-mt_tiny_ara-eng"
tokenizer = MarianTokenizer.from_pretrained(model_name)
model = MarianMTModel.from_pretrained(model_name)
tok = tokenizer('أضاف قائلاً، "لدينا الآن فئران تبلغ من العمر 4 أشهر التي كانت تعاني في السابق من مرض السكري، ولكنها لم تعد تعاني منه الآن.', return_tensors="pt").input_ids
output = model.generate(tok)[0]
tokenizer.decode(output, skip_special_tokens=True)
Benchmarks
Teacher
| testset | BLEU | chr-F | COMET |
|---|---|---|---|
| Flores+ | 34.0 | 61.0 | 0.8307 |
Student
| testset | BLEU | chr-F | COMET |
|---|---|---|---|
| Flores+ | 31.1 | 59.1 | 0.8174 |