XLM-RoBERTa Base for Amharic (xlm_roberta_base_finetuned_amharic)

Description

xlm_roberta_base_finetuned_amharic is a Amharic RoBERTa model obtained by fine-tuning xlm-roberta-base model on Amharic 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 Amharic corpus.

Predicted Entities

Live Demo Open in Colab Download Copy S3 URI

How to use

embeddings = XlmRoBertaEmbeddings.pretrained("xlm_roberta_base_finetuned_amharic", "am") \
      .setInputCols("sentence", "token") \
      .setOutputCol("embeddings")

val embeddings = XlmRoBertaEmbeddings.pretrained("xlm_roberta_base_finetuned_amharic", "am")
      .setInputCols("sentence", "token")
      .setOutputCol("embeddings")

import nlu
nlu.load("am.embed.xlm_roberta").predict("""Put your text here.""")

Model Information

Data Source

https://huggingface.co/Davlan/xlm-roberta-base-finetuned-amharic

Benchmarking

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

Dataset| XLM-R F1 | am_roberta F1
-|-|-
[MasakhaNER](https://github.com/masakhane-io/masakhane-ner) | 70.96 | 77.97