Xlm Roberta Base Finetuned Hausa

🚀 xlm-roberta-base-finetuned-hausa

The xlm-roberta-base-finetuned-hausa is a Hausa RoBERTa model that offers better performance in text classification and named - entity recognition compared to the XLM - RoBERTa.

🚀 Quick Start

This model can be used with Transformers pipeline for masked token prediction. You can follow the code example below to start using it.

✨ Features

• Enhanced Performance: It provides better performance than the XLM - RoBERTa on text classification and named entity recognition datasets.
• Fine - tuned on Hausa: This model is a xlm - roberta - base model fine - tuned on Hausa corpus.

💻 Usage Examples

Basic Usage

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

>>> from transformers import pipeline
>>> unmasker = pipeline('fill - mask', model='Davlan/xlm - roberta - base - finetuned - hausa')
>>> unmasker("Shugaban <mask> Muhammadu Buhari ya amince da shawarar da ma’aikatar sufuri karkashin jagoranci")
                    
[{'sequence': '<s> Shugaban kasa Muhammadu Buhari ya amince da shawarar da ma’aikatar sufuri karkashin jagoranci</s>', 
'score': 0.8104371428489685, 
'token': 29762, 
'token_str': '▁kasa'}, 
{'sequence': '<s> Shugaban Najeriya Muhammadu Buhari ya amince da shawarar da ma’aikatar sufuri karkashin jagoranci</s>', 'score': 0.17371904850006104, 
'token': 49173, 
'token_str': '▁Najeriya'}, 
{'sequence': '<s> Shugaban kasar Muhammadu Buhari ya amince da shawarar da ma’aikatar sufuri karkashin jagoranci</s>', 'score': 0.006917025428265333, 
'token': 21221, 
'token_str': '▁kasar'}, 
{'sequence': '<s> Shugaban Nigeria Muhammadu Buhari ya amince da shawarar da ma’aikatar sufuri karkashin jagoranci</s>', 'score': 0.005785710643976927, 
'token': 72620, 
'token_str': '▁Nigeria'}, 
{'sequence': '<s> Shugaban Kasar Muhammadu Buhari ya amince da shawarar da ma’aikatar sufuri karkashin jagoranci</s>', 'score': 0.0010596115607768297, 
'token': 170255, 
'token_str': '▁Kasar'}]

📚 Documentation

Intended uses & limitations

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 [Hausa CC - 100](http://data.statmt.org/cc - 100/)

Training procedure

This model was trained on a single NVIDIA V100 GPU

Eval results on Test set (F - score, average over 5 runs)

BibTeX entry and citation info

By David Adelani