bert-base-multilingual-cased-finetuned-amharic is a Amharic BERT model obtained by replacing mBERT vocabulary by amharic vocabulary because the language was not supported, and fine-tuning bert-base-multilingual-cased model on Amharic language texts. It provides better performance than the multilingual Amharic on named entity recognition datasets.
Specifically, this model is a bert-base-multilingual-cased model that was fine-tuned on Amharic corpus using Amharic vocabulary.
You can use this model with Transformers pipeline for masked token prediction.
from transformers import pipeline unmasker = pipeline('fill-mask', model='Davlan/bert-base-multilingual-cased-finetuned-amharic') unmasker("የአሜሪካ የአፍሪካ ቀንድ ልዩ መልዕክተኛ ጄፈሪ ፌልትማን በአራት አገራት የሚያደጉትን [MASK] መጀመራቸውን የአሜሪካ የውጪ ጉዳይ ሚንስቴር አስታወቀ።")
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.
This model was fine-tuned on Amharic CC-100
This model was trained on a single NVIDIA V100 GPU
|Dataset||mBERT F1||am_bert F1|
By David Adelani
- Downloads last month