Hugging Face's logo

projecte-aina
/
roberta-large-ca-v2

Fill-Mask
Transformers
PyTorch
Catalan
roberta
catalan
masked-lm
RoBERTa-large-ca-v2
CaText
Catalan Textual Corpus
Inference Endpoints
Model card Files Community
1
roberta-large-ca-v2 / README.md
gonzalez-agirre's picture
gonzalez-agirre
Model uploaded
6efed62
preview code
|
raw
history blame
9.52 kB
metadata
language:
  - ca
license: apache-2.0
tags:
  - catalan
  - masked-lm
  - RoBERTa-large-ca
  - CaText
  - Catalan Textual Corpus
widget:
  - text: El Català és una llengua molt <mask>.
  - text: Salvador Dalí va viure a <mask>.
  - text: La Costa Brava té les millors <mask> d'Espanya.
  - text: El cacaolat és un batut de <mask>.
  - text: <mask> és la capital de la Garrotxa.
  - text: Vaig al <mask> a buscar bolets.
  - text: Antoni Gaudí vas ser un <mask> molt important per la ciutat.
  - text: Catalunya és una referència en <mask> a nivell europeu.

Catalan BERTa (roberta-large-ca) large model

Table of Contents

Click to expand

Model description

The roberta-large-ca is a transformer-based masked language model for the Catalan language. It is based on the RoBERTA large model and has been trained on a medium-size corpus collected from publicly available corpora and crawlers.

Intended Uses and Limitations

roberta-large-ca model is ready-to-use only for masked language modeling to perform the Fill Mask task (try the inference API or read the next section). However, it is intended to be fine-tuned on non-generative downstream tasks such as Question Answering, Text Classification, or Named Entity Recognition.

How to Use

Here is how to use this model:

from transformers import AutoModelForMaskedLM
from transformers import AutoTokenizer, FillMaskPipeline
from pprint import pprint
tokenizer_hf = AutoTokenizer.from_pretrained('projecte-aina/roberta-large-ca')
model = AutoModelForMaskedLM.from_pretrained('projecte-aina/roberta-large-ca')
model.eval()
pipeline = FillMaskPipeline(model, tokenizer_hf)
text = f"Em dic <mask>."
res_hf = pipeline(text)
pprint([r['token_str'] for r in res_hf])

Training

Training data

The training corpus consists of several corpora gathered from web crawling and public corpora.

Corpus Size in GB
Catalan Crawling 13.00
Wikipedia 1.10
DOGC 0.78
Catalan Open Subtitles 0.02
Catalan Oscar 4.00
CaWaC 3.60
Cat. General Crawling 2.50
Cat. Goverment Crawling 0.24
ACN 0.42
Padicat 0.63
RacoCatalá 8.10
Nació Digital 0.42
Vilaweb 0.06
Tweets 0.02

Training Procedure

The training corpus has been tokenized using a byte version of Byte-Pair Encoding (BPE) used in the original RoBERTA model with a vocabulary size of 52,000 tokens. The RoBERTa-large pretraining consists of a masked language model training that follows the approach employed for the RoBERTa large model with the same hyperparameters as in the original work. The training lasted a total of 96 hours with 32 NVIDIA V100 GPUs of 16GB DDRAM.

Evaluation

CLUB Benchmark

The BERTa-large model has been fine-tuned on the downstream tasks of the Catalan Language Understanding Evaluation benchmark (CLUB), that has been created along with the model.

It contains the following tasks and their related datasets:

  1. Named Entity Recognition (NER)

    NER (AnCora): extracted named entities from the original Ancora version, filtering out some unconventional ones, like book titles, and transcribed them into a standard CONLL-IOB format

  2. Part-of-Speech Tagging (POS)

    POS (AnCora): from the Universal Dependencies treebank of the well-known Ancora corpus.

  3. Text Classification (TC)

    TeCla: consisting of 137k news pieces from the Catalan News Agency (ACN) corpus, with 30 labels.

  4. Textual Entailment (TE)

    TE-ca: consisting of 21,163 pairs of premises and hypotheses, annotated according to the inference relation they have (implication, contradiction, or neutral), extracted from the Catalan Textual Corpus.

  5. Semantic Textual Similarity (STS)

    STS-ca: consisting of more than 3000 sentence pairs, annotated with the semantic similarity between them, scraped from the Catalan Textual Corpus.

  6. Question Answering (QA):

    VilaQuAD: contains 6,282 pairs of questions and answers, outsourced from 2095 Catalan language articles from VilaWeb newswire text.

    ViquiQuAD: consisting of more than 15,000 questions outsourced from Catalan Wikipedia randomly chosen from a set of 596 articles that were originally written in Catalan.

    CatalanQA: an aggregation of 2 previous datasets (VilaQuAD and ViquiQuAD), 21,427 pairs of Q/A balanced by type of question, containing one question and one answer per context, although the contexts can repeat multiple times.

    XQuAD-ca: the Catalan translation of XQuAD, a multilingual collection of manual translations of 1,190 question-answer pairs from English Wikipedia used only as a test set.

Here are the train/dev/test splits of the datasets:

Task (Dataset) Total Train Dev Test
NER (Ancora) 13,581 10,628 1,427 1,526
POS (Ancora) 16,678 13,123 1,709 1,846
STS (STS-ca) 3,073 2,073 500 500
TC (TeCla) 137,775 110,203 13,786 13,786
TE (TE-ca) 21,163 16,930 2,116 2,117
QA (VilaQuAD) 6,282 3,882 1,200 1,200
QA (ViquiQuAD) 14,239 11,255 1,492 1,429
QA (CatalanQA) 21,427 17,135 2,157 2,135

Evaluation Results

Task NER (F1) POS (F1) STS-ca (Comb) TeCla (Acc.) TEca (Acc.) VilaQuAD (F1/EM) ViquiQuAD (F1/EM) CatalanQA (F1/EM) XQuAD-ca 1 (F1/EM)
RoBERTa-large-ca 89.82 99.02 83.41 75.46 83.61 89.34/75.50 89.20/75.77 90.72/79.06 73.79/55.34
RoBERTa-base-ca-v2 89.29 98.96 79.07 74.26 83.14 87.74/72.58 88.72/75.91 89.50/76.63 73.64/55.42
BERTa 89.76 98.96 80.19 73.65 79.26 85.93/70.58 87.12/73.11 89.17/77.14 69.20/51.47
mBERT 86.87 98.83 74.26 69.90 74.63 82.78/67.33 86.89/73.53 86.90/74.19 68.79/50.80
XLM-RoBERTa 86.31 98.89 61.61 70.14 33.30 86.29/71.83 86.88/73.11 88.17/75.93 72.55/54.16

1 : Trained on CatalanQA, tested on XQuAD-ca.

Licensing Information

Apache License, Version 2.0

Citation Information

If you use any of these resources (datasets or models) in your work, please cite our latest paper:

@inproceedings{armengol-estape-etal-2021-multilingual,
    title = "Are Multilingual Models the Best Choice for Moderately Under-resourced Languages? {A} Comprehensive Assessment for {C}atalan",
    author = "Armengol-Estap{\'e}, Jordi  and
      Carrino, Casimiro Pio  and
      Rodriguez-Penagos, Carlos  and
      de Gibert Bonet, Ona  and
      Armentano-Oller, Carme  and
      Gonzalez-Agirre, Aitor  and
      Melero, Maite  and
      Villegas, Marta",
    booktitle = "Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021",
    month = aug,
    year = "2021",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2021.findings-acl.437",
    doi = "10.18653/v1/2021.findings-acl.437",
    pages = "4933--4946",
}

Funding

This work was funded by the Departament de la Vicepresidència i de Polítiques Digitals i Territori de la Generalitat de Catalunya within the framework of Projecte AINA.

Contributions

[N/A]