Spanish Bert Embeddings (from IIC)

Description

Pretrained Bert Embeddings model, uploaded to Hugging Face, adapted and imported into Spark NLP. dpr-spanish-passage_encoder-squades-base is a Spanish model orginally trained by IIC.

Download

How to use

documentAssembler = DocumentAssembler() \
    .setInputCol("text") \
    .setOutputCol("document")

tokenizer = Tokenizer() \
    .setInputCols("document") \
    .setOutputCol("token")
  
embeddings = BertEmbeddings.pretrained("bert_embeddings_dpr_spanish_passage_encoder_squades_base","es") \
    .setInputCols(["document", "token"]) \
    .setOutputCol("embeddings")
    
pipeline = Pipeline(stages=[documentAssembler, tokenizer, embeddings])

data = spark.createDataFrame([["Me encanta chispa nlp"]]).toDF("text")

result = pipeline.fit(data).transform(data)
val documentAssembler = new DocumentAssembler() 
      .setInputCol("text") 
      .setOutputCol("document")
 
val tokenizer = new Tokenizer() 
    .setInputCols(Array("document"))
    .setOutputCol("token")

val embeddings = BertEmbeddings.pretrained("bert_embeddings_dpr_spanish_passage_encoder_squades_base","es") 
    .setInputCols(Array("document", "token")) 
    .setOutputCol("embeddings")

val pipeline = new Pipeline().setStages(Array(documentAssembler, tokenizer, embeddings))

val data = Seq("Me encanta chispa nlp").toDF("text")

val result = pipeline.fit(data).transform(data)

Model Information

Model Name: bert_embeddings_dpr_spanish_passage_encoder_squades_base
Compatibility: Spark NLP 3.4.2+
License: Open Source
Edition: Official
Input Labels: [sentence, token]
Output Labels: [bert]
Language: es
Size: 412.4 MB
Case sensitive: true

References

  • https://huggingface.co/IIC/dpr-spanish-passage_encoder-squades-base
  • https://arxiv.org/abs/2004.04906
  • https://github.com/facebookresearch/DPR
  • https://paperswithcode.com/sota?task=text+similarity&dataset=squad_es