rasyosef/bert-medium-amharic-finetuned-ner

This is a fine-tuned version of the bert-medium-amharic model on the amharic-named-entity-recognition dataset and is ready to use for named entity recognition (NER).

It achieves the following results on the evaluation set:

• Precision: 0.65
• Recall: 0.73
• F1: 0.69

How to use

You can use this model directly with a pipeline for token classification:

from transformers import pipeline
checkpoint = "rasyosef/bert-medium-amharic-finetuned-ner"
token_classifier = pipeline("token-classification", model=checkpoint, aggregation_strategy="simple")
token_classifier("አትሌት ኃይሌ ገ/ሥላሴ ኒውዮርክ ውስጥ በሚደረገው የተባበሩት መንግሥታት ድርጅት ልዩ የሰላም ስብሰባ ላይ እንዲገኝ ተጋበዘ።")

Output:

[{'entity_group': 'TTL',
  'score': 0.9841112,
  'word': 'አትሌት',
  'start': 0,
  'end': 4},
 {'entity_group': 'PER',
  'score': 0.99379075,
  'word': 'ኃይሌ ገ / ሥላሴ',
  'start': 5,
  'end': 14},
 {'entity_group': 'LOC',
  'score': 0.8818362,
  'word': 'ኒውዮርክ',
  'start': 15,
  'end': 20},
 {'entity_group': 'ORG',
  'score': 0.99056435,
  'word': 'የተባበሩት መንግሥታት ድርጅት',
  'start': 32,
  'end': 50}]

Code

https://github.com/rasyosef/amharic-named-entity-recognition