Hugging Face's logo

language: yo datasets: - Menyo-20k - Yoruba Embedding corpus

bert-base-multilingual-cased-finetuned-yoruba

Model description

bert-base-multilingual-cased-finetuned-yoruba is a Yoruba BERT model obtained by fine-tuning bert-base-multilingual-cased model on Yorùbá language texts. It provides better performance than the multilingual BERT on text classification and named entity recognition datasets.

Specifically, this model is a bert-base-multilingual-cased model that was fine-tuned on Yorùbá corpus.

Intended uses & limitations

How to use

You can use this model with Transformers pipeline for masked token prediction.

from transformers import pipeline
>>> from transformers import pipeline
>>> unmasker = pipeline('fill-mask', model='Davlan/bert-base-multilingual-cased-finetuned-yoruba')
>>> unmasker("Arẹmọ Phillip to jẹ ọkọ [MASK] Elizabeth to ti wa lori aisan ti dagbere faye lẹni ọdun mọkandilọgọrun")

Limitations and bias

This model is limited by its training dataset of entity-annotated news articles from a specific span of time. This may not generalize well for all use cases in different domains.

Training data

This model was fine-tuned on on Bible, JW300, Menyo-20k, Yoruba Embedding corpus and CC-Aligned, Wikipedia, news corpora (BBC Yoruba, VON Yoruba, Asejere, Alaroye), and other small datasets curated from friends.

Training procedure

This model was trained on a single NVIDIA V100 GPU

Eval results on Test set (F-score)

Dataset	F1-score
Yoruba GV NER	75.34
MasakhaNER	80.82
BBC Yoruba	80.66

BibTeX entry and citation info

By David Adelani