Edit model card

IndoBERT-Lite-SQuAD base fine-tuned on Full Translated SQuAD v2

IndoBERT-Lite trained by Indo Benchmark and fine-tuned on Translated SQuAD 2.0 for Q&A downstream task.

Model in action

Fast usage with pipelines:

from transformers import BertTokenizerFast, pipeline

tokenizer = BertTokenizerFast.from_pretrained(
    'Wikidepia/indobert-lite-squad'
)
qa_pipeline = pipeline(
    "question-answering",
    model="Wikidepia/indobert-lite-squad",
    tokenizer=tokenizer
)

qa_pipeline({
    'context': "Setelah menghabiskan waktu satu tahun di Praha, Einstein tinggal di Swiss antara tahun 1895 dan 1914, melepas kewarganegaraan Jermannya pada tahun 1896, dan lulus sarjana dari sekolah politeknik federal Swiss (kelak Eidgenössische Technische Hochschule, ETH) di Zürich pada tahun 1900.",
    'question': "Kapan Einstein melepas kewarganegaraan Jerman?"
})

Output:

{
  "score": 0.9169162511825562,
  "start": 147,
  "end": 151,
  "answer": "1896"
}

README copied from mrm8488's repository

Downloads last month
18
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.