Question Answering
Transformers
PyTorch
Bulgarian
bert
torch
Edit model card

BERT BASE (cased) finetuned on Bulgarian squad data

Pretrained model on Bulgarian language using a masked language modeling (MLM) objective. It was introduced in this paper and first released in this repository. This model is cased: it does make a difference between bulgarian and Bulgarian. The training data is Bulgarian text from OSCAR, Chitanka and Wikipedia.

It was finetuned on private squad Bulgarian data.

Then, it was compressed via progressive module replacing.

How to use

Here is how to use this model in PyTorch:

>>> from transformers import pipeline
>>> 
>>> model = pipeline(
>>>     'question-answering',
>>>     model='rmihaylov/bert-base-squad-theseus-bg',
>>>     tokenizer='rmihaylov/bert-base-squad-theseus-bg',
>>>     device=0,
>>>     revision=None)
>>>
>>> question = "С какво се проследява пандемията?"
>>> context = "Епидемията гасне, обяви при обявяването на данните тази сутрин Тодор Кантарджиев, член на Националния оперативен щаб. Той направи този извод на база на данните от математическите модели, с които се проследява развитието на заразата. Те показват, че т. нар. ефективно репродуктивно число е вече в границите 0.6-1. Тоест, 10 души заразяват 8, те на свой ред 6 и така нататък. "

>>> output = model(**{'question': question, 'context': context})
>>> print(output)

{'score': 0.85157310962677, 'start': 162, 'end': 186, 'answer': ' математическите модели,'}
Downloads last month
7
Inference Examples
Inference API (serverless) has been turned off for this model.

Datasets used to train rmihaylov/bert-base-squad-theseus-bg