license: apache-2.0
language:
- es
- ca
Aina Project's Spanish-Catalan machine translation model
Table of Contents
- Model Description
- Intended Uses and Limitations
- How to Use
- Training
- Evaluation
- Additional Information
Model description
This model was trained from scratch using the Fairseq toolkit on a combination of Catalan-Spanish datasets, up to 92 million sentences. Additionally, the model was evaluated on several public datasets pertaining to 5 different domains, specifically: general, adminstrative, technology, biomedical, and news.
Intended uses and limitations
You can use this model for machine translation from Spanish 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/mt-aina-es-ca", revision="main")
tokenizer=pyonmttok.Tokenizer(mode="none", sp_model_path = model_dir + "/spm.model")
tokenized=tokenizer.tokenize("Bienvenido al Proyecto Aina!")
translator = ctranslate2.Translator(model_dir)
translated = translator.translate_batch([tokenized[0]])
print(tokenizer.detokenize(translated[0][0]['tokens']))
Training
Training data
The model was trained on a combination of the following datasets:
| Dataset | Sentences | Tokens |
|---|---|---|
| DOCG v2 | 8.472.786 | 188.929.206 |
| El Periodico | 6.483.106 | 145.591.906 |
| EuroParl | 1.876.669 | 49.212.670 |
| WikiMatrix | 1.421.077 | 34.902.039 |
| Wikimedia | 335.955 | 8.682.025 |
| QED | 71.867 | 1.079.705 |
| TED2020 v1 | 52.177 | 836.882 |
| CCMatrix v1 | 56.103.820 | 1.064.182.320 |
| MultiCCAligned v1 | 2.433.418 | 48.294.144 |
| ParaCrawl | 15.327.808 | 334.199.408 |
| Total | 92.578.683 | 1.875.910.305 |
Training procedure
Data preparation
All datasets were concatenated and filtered using the mBERT Gencata parallel filter and cleaned using the clean-corpus-n.pl script from moses, allowing sentences between 5 and 150 words.
Before training, the punctuation was normalized using a modified version of the join-single-file.py script from SoftCatalà.
Tokenization
All data was 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 8.000 updates. Weights were saved every 1000 updates and reported results are the average of the last 6 checkpoints.
Evaluation
Variables and metrics
We use the BLEU score for evaluation on following test sets: Flores-101, TaCon, United Nations, Cybersecurity, wmt19 biomedical test set, wmt13 news test set
Evaluation results
Below are the evaluation results on the machine translation from Spanish to Catalan compared to Softcatalà and Google Translate:
| Test set | SoftCatalà | Google Translate | mt-aina-es-ca |
|---|---|---|---|
| Spanish Constitution | 63,6 | 61,7 | 63,0 |
| United Nations | 73,8 | 74,8 | 74,9 |
| Flores 101 dev | 22 | 23,1 | 22,5 |
| Flores 101 devtest | 22,7 | 23,6 | 23,1 |
| Cybersecurity | 61,4 | 69,5 | 67,3 |
| wmt 19 biomedical | 60,2 | 59,7 | 60,6 |
| wmt 13 news | 21,3 | 22,4 | 22,0 |
| Average | 46,4 | 47,8 | 47,6 |
Additional information
Author
Language Technologies Unit (LangTech) at the Barcelona Supercomputing Center.
Contact information
For further information, please send an email to [email protected].
Copyright
Language Technologies Unit at Barcelona Supercomputing Center (2023).
Licensing Information
This work is licensed under an 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 models published in this repository are intended for a generalist purpose and are available to third parties. These models may have bias and/or any other undesirable distortions. When third parties, deploy or provide systems and/or services to other parties using any of these models (or using systems based on these models) or become users of the models, they should note that it is their responsibility to mitigate the risks arising from their 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 models (BSC – Barcelona Supercomputing Center) be liable for any results arising from the use made by third parties of these models.