Helsinki-NLP/opus-mt_tiny_cat-eng

OPUS-MT-tiny-cat-eng

Distilled model from a Tatoeba-MT Teacher: Tatoeba-MT-models/roa-eng/opusTCv20230926max50+bt+jhubc_transformer-big_2024-08-17, 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_cat-eng"
tokenizer = MarianTokenizer.from_pretrained(model_name)
model = MarianMTModel.from_pretrained(model_name)
tok = tokenizer("El concepte prové de la Xina, on la flor del cirerer era la més apreciada.", 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+	46.7	70.5	0.8574

Student

testset	BLEU	chr-F	COMET
Flores+	42.5	67.7	0.8637