rasyosef / bert-tiny-amharic

bert-tiny-amharic

This model has the same architecture as bert-tiny and was pretrained from scratch using the Amharic subsets of the oscar , mc4 , and amharic-sentences-corpus datasets, on a total of 290 million tokens . The tokenizer was trained from scratch on the same text corpus, and had a vocabulary size of 28k.

It achieves the following results on the evaluation set:

• Loss: 4.27
• Perplexity: 71.52

This model has just 4.18M parameters.

How to use

You can use this model directly with a pipeline for masked language modeling:

>>> from transformers import pipeline
>>> unmasker = pipeline('fill-mask', model='rasyosef/bert-tiny-amharic')
>>> unmasker("ከሀገራቸው ከኢትዮጵያ ከወጡ ግማሽ ምዕተ [MASK] ተቆጥሯል።")

[{'score': 0.5629344582557678,
  'token': 9617,
  'token_str': 'ዓመታት',
  'sequence': 'ከሀገራቸው ከኢትዮጵያ ከወጡ ግማሽ ምዕተ ዓመታት ተቆጥሯል ።'},
 {'score': 0.3049253523349762,
  'token': 9345,
  'token_str': 'ዓመት',
  'sequence': 'ከሀገራቸው ከኢትዮጵያ ከወጡ ግማሽ ምዕተ ዓመት ተቆጥሯል ።'},
 {'score': 0.0681595504283905,
  'token': 10898,
  'token_str': 'አመታት',
  'sequence': 'ከሀገራቸው ከኢትዮጵያ ከወጡ ግማሽ ምዕተ አመታት ተቆጥሯል ።'},
 {'score': 0.028840897604823112,
  'token': 9913,
  'token_str': 'አመት',
  'sequence': 'ከሀገራቸው ከኢትዮጵያ ከወጡ ግማሽ ምዕተ አመት ተቆጥሯል ።'},
 {'score': 0.008974998258054256,
  'token': 15098,
  'token_str': 'ዘመናት',
  'sequence': 'ከሀገራቸው ከኢትዮጵያ ከወጡ ግማሽ ምዕተ ዘመናት ተቆጥሯል ።'}]

Finetuning

This model was finetuned and evaluated on the following Amharic NLP tasks

• Sentiment Classification
  • Dataset: amharic-sentiment
  • Code: https://github.com/rasyosef/amharic-sentiment-classification
• Named Entity Recognition
  • Dataset: amharic-named-entity-recognition
  • Code: https://github.com/rasyosef/amharic-named-entity-recognition

Finetuned Model Performance

The reported F1 scores are macro averages.