XLM-RoBERTa Base for Hausa (xlm_roberta_base_finetuned_hausa)


xlm_roberta_base_finetuned_hausa is a ** Hausa RoBERTa** model obtained by fine-tuning xlm-roberta-base model on Hausa language texts. It provides better performance than the XLM-RoBERTa on named entity recognition datasets.

Specifically, this model is an xlm-roberta-base model that was fine-tuned on the Hausa corpus.

Predicted Entities


How to use

embeddings = XlmRoBertaEmbeddings.pretrained("xlm_roberta_base_finetuned_hausa", "ha") \
      .setInputCols("sentence", "token") \
val embeddings = XlmRoBertaEmbeddings.pretrained("xlm_roberta_base_finetuned_hausa", "ha)
      .setInputCols("sentence", "token")

Model Information

Model Name: xlm_roberta_base_finetuned_hausa
Compatibility: Spark NLP 3.3.0+
License: Open Source
Edition: Official
Input Labels: [token, sentence]
Output Labels: [embeddings]
Language: ha
Case sensitive: true

Data Source



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

Dataset| XLM-R F1 | ha_roberta F1
[MasakhaNER](https://github.com/masakhane-io/masakhane-ner) | 86.10 | 91.47
[VOA Hausa Textclass](https://huggingface.co/datasets/hausa_voa_topics) | |