projecte-aina/aina-translator-fr-ca

Projecte Aina’s French-Catalan machine translation model

Model description

This model was trained from scratch using the Fairseq toolkit on a combination of Catalan-French datasets, which after filtering and cleaning comprised 18.634.844 sentence pairs. The model is evaluated on the Flores and NTREX evaluation sets.

Intended uses and limitations

You can use this model for machine translation from French to Catalan.

How to use

Usage

Required libraries:

pip install ctranslate2 pyonmttok

Translate a sentence using python

import ctranslate2
import pyonmttok
from huggingface_hub import snapshot_download
model_dir = snapshot_download(repo_id="projecte-aina/aina-translator-fr-ca", revision="main")

tokenizer=pyonmttok.Tokenizer(mode="none", sp_model_path = model_dir + "/spm.model")
tokenized=tokenizer.tokenize("Bienvenue au projet Aina!")

translator = ctranslate2.Translator(model_dir)
translated = translator.translate_batch([tokenized[0]])
print(tokenizer.detokenize(translated[0][0]['tokens']))

Limitations and bias

At the time of submission, no measures have been taken to estimate the bias and toxicity embedded in the model. However, we are well aware that our models may be biased. We intend to conduct research in these areas in the future, and if completed, this model card will be updated.

Training

Training data

The model was trained on a combination of the following datasets:

Dataset	Sentences	Sentences after Cleaning
CCMatrix	24.386.198	16.305.758
Multi CCAligned	1.954.475	1.442.584
WikiMatrix	490.871	437.665
GNOME	12.962	1.686
KDE 4	163.143	111.750
QED	65.336	52.797
TED 2020	51.833	44.101
Open Subtitles	392.159	225.786
Total	27.532.050	18.634.844

Training procedure

Data preparation

All datasets are deduplicated and filtered to remove any sentence pairs with a cosine similarity of less than 0.75. This is done using sentence embeddings calculated using LaBSE. The filtered datasets are then concatenated to form a final corpus of 6.159.631 and before training the punctuation is normalized using a modified version of the join-single-file.py script from SoftCatalà.

Tokenization

All data is tokenized using sentencepiece, with 50 thousand token sentencepiece model learned from the combination of all filtered training data. This model is included.

Hyperparameters

The model is based on the Transformer-XLarge proposed by Subramanian et al. The following hyperparamenters were set on the Fairseq toolkit:

Hyperparameter	Value
Architecture	transformer_vaswani_wmt_en_de_bi
Embedding size	1024
Feedforward size	4096
Number of heads	16
Encoder layers	24
Decoder layers	6
Normalize before attention	True
--share-decoder-input-output-embed	True
--share-all-embeddings	True
Effective batch size	96.000
Optimizer	adam
Adam betas	(0.9, 0.980)
Clip norm	0.0
Learning rate	1e-3
Lr. schedurer	inverse sqrt
Warmup updates	4000
Dropout	0.1
Label smoothing	0.1

The model was trained using shards of 10 million sentences, for a total of 8.548 updates. Weights were saved every 1000 updates and reported results are the average of the last 8 checkpoints.

Evaluation

Variable and metrics

We use the BLEU score for evaluation on the Flores-101 and NTREX test sets.

Evaluation results

Below are the evaluation results on the machine translation from French to Catalan compared to Softcatalà and Google Translate:

Test set	SoftCatalà	Google Translate	aina-translator-fr-ca
Flores 101 dev	30,9	37,0	33,0
Flores 101 devtest	31,3	37,1	34,4
NTREX	24,5	30,5	27,0
Average	28,9	34,9	31,5

Additional information

Author

The Language Technologies Unit from Barcelona Supercomputing Center.

Contact

For further information, please send an email to langtech@bsc.es.

Copyright

Copyright(c) 2023 by Language Technologies Unit, Barcelona Supercomputing Center.

License

Apache License, Version 2.0

Funding

This work has been promoted and financed by the Generalitat de Catalunya through the Aina project.

Disclaimer

Click to expand

The model published in this repository is intended for a generalist purpose and is available to third parties under a permissive Apache License, Version 2.0.

Be aware that the model may have biases and/or any other undesirable distortions.

When third parties deploy or provide systems and/or services to other parties using this model (or any system based on it) or become users of the model, they should note that it is their responsibility to mitigate the risks arising from its use and, in any event, to comply with applicable regulations, including regulations regarding the use of Artificial Intelligence.

In no event shall the owner and creator of the model (Barcelona Supercomputing Center) be liable for any results arising from the use made by third parties.