NGC | Catalog
CatalogModelsNMT Multilingual En De/Es/Fr Transformer12x2

NMT Multilingual En De/Es/Fr Transformer12x2

Logo for NMT Multilingual En De/Es/Fr Transformer12x2
Description
Multilingual Neural Machine Translation model to translate from English to German/Spanish/French
Publisher
NVIDIA
Latest Version
1.2.0
Modified
April 4, 2023
Size
1003.25 MB

Model Overview

This model can be used for translating text in source language (En) to a text in target language (De/Es/Fr).

Model Architecture

The model is based on Transformer "Big" architecture originally presented in "Attention Is All You Need" paper [1]. In this particular instance, the model has 12 layers in the encoder and 2 layers in the decoder. It is using SentencePiece tokenizer [2].

Training

These models were trained on a collection of many publicly available datasets comprising of millions of parallel sentences. The NeMo toolkit [5] was used for training this model over roughly 700k steps.

Datasets

While training this model, we used the following datasets:

German

Spanish

French

Tokenizer Construction

We used the SentencePiece tokenizer [2] with shared encoder and decoder BPE tokenizers.

Performance

The accuracy of translation models are often measured using BLEU scores [3]. The model achieves the following sacreBLEU [4] scores on WMT test sets

De
WMT13 - 28.3
WMT14 - 29.8

Es
WMT12 - 39.8
WMT13 - 35.8

Fr
WMT13 - 34.7
WMT14 - 40.7

How to Use this Model

The model is available for use in the NeMo toolkit [5], and can be used as a pre-trained checkpoint for inference or for fine-tuning on another dataset.

Automatically load the model from NGC

import nemo
import nemo.collections.nlp as nemo_nlp
nmt_model = nemo_nlp.models.machine_translation.MTEncDecModel.from_pretrained(model_name="mnmt_en_deesfr_transformer12x2")

Translating text with this model

python [NEMO_GIT_FOLDER]/examples/nlp/machine_translation/nmt_transformer_infer.py --model=mnmt_en_deesfr_transformer12x2.nemo --srctext=[TEXT_IN_SRC_LANGUAGE] --tgtout=[WHERE_TO_SAVE_TRANSLATIONS] --target_lang [TARGET_LANGUAGE] --source_lang en

where [TARGET_LANGUAGE] can be 'de' or 'es' or 'fr'

Input

This translate method of the NMT model accepts a list of de-tokenized strings.

Output

The translate method outputs a list of de-tokenized strings in the target language.

Limitations

No known limitations at this time.

References

[1] Vaswani, Ashish, et al. "Attention is all you need." arXiv preprint arXiv:1706.03762 (2017).

[2] https://github.com/google/sentencepiece

[3] https://en.wikipedia.org/wiki/BLEU

[4] https://github.com/mjpost/sacreBLEU

[5] NVIDIA NeMo Toolkit

Licence

License to use this model is covered by the NGC TERMS OF USE unless another License/Terms Of Use/EULA is clearly specified. By downloading the public and release version of the model, you accept the terms and conditions of the NGC TERMS OF USE.