Spanish Electra Uncased Embeddings (Oscar dataset)

Description

Pretrained Electra Embeddings model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. electricidad-base-generator is a Spanish model orginally trained by mrm8488.

Live Demo Open in Colab Download Copy S3 URI

How to use

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

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

data = spark.createDataFrame([["Amo Spark 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("electra_embeddings_electricidad_base_generator","es") 
    .setInputCols(Array("document", "token")) 
    .setOutputCol("embeddings")

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

val data = Seq("Amo Spark NLP").toDF("text")

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

import nlu
nlu.load("es.embed.electra.base").predict("""Amo Spark NLP""")

Model Information

References

• https://huggingface.co/mrm8488/electricidad-base-generator
• https://imgur.com/uxAvBfh
• https://oscar-corpus.com/
• https://openreview.net/pdf?id=r1xMH1BtvB
• https://arxiv.org/pdf/1406.2661.pdf
• https://rajpurkar.github.io/SQuAD-explorer/
• https://openreview.net/pdf?id=r1xMH1BtvB
• https://twitter.com/julien_c
• https://twitter.com/mrm8488