File size: 464 Bytes
4a5c795
 
 
 
 
 
 
3e56dc1
4a5c795
 
 
 
a5f0822
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
---
license: apache-2.0
datasets:
- squad
language:
- pt
pipeline_tag: sentence-similarity
library_name: sentence-transformers
---

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