Modelpublic

Helsinki-NLP/opus-mt_tiny_spa-eng

sourceHugging Faceapache-2.0updated 5mo agoView on Hugging Face

1likes1kdownloads

Model Card

OPUS-MT-tiny-spa-eng

Distilled model from a Tatoeba-MT Teacher: Tatoeba-MT-models/cat+oci+spa-eng/opusTCv20210807+bt_transformer-big_2022-03-13, 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

python

from transformers import MarianMTModel, MarianTokenizer
model_name = "Helsinki-NLP/opus-mt_tiny_spa-eng"
tokenizer = MarianTokenizer.from_pretrained(model_name)
model = MarianMTModel.from_pretrained(model_name)
tok = tokenizer("La esfinge aparece como escenario de fondo y narradora de una larga historia.", 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+	29.8	59.7	0.8510
Bouquet	43.1	64.4	0.851

Student

testset	BLEU	chr-F	COMET
Flores+	25.0	55.2	0.8300
Bouquet	37.4	60.2	0.8510

Marian models

We also provide Marian-compatible versions of this model. To use them, compile Marian and run decoding with marian-decoder, for example:

bash

marian-decoder \
  -i input.txt \
  -c final.model.npz.best-perplexity.npz.decoder.yml \
  -m final.model.npz.best-perplexity.npz \
  -v vocab.spm vocab.spm