This is a sentence_bert based model fine tuned using portuguese SQuAD dataset with batchs of:

  • positive samples: question - sentence containing answer
  • hard negative samples question which answers is not in the sentence, but is in the context of SQuAD dataset.

This model generate embeddings from text to use with semantic search

Downloads last month
10
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.

Dataset used to train vabatista/sbert-squad-pt-br