Financial Korean Electra Embeddings (Base)

Description

Pretrained Financial Korean Electra Embeddings model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. finance-koelectra-base-generator is a Korean model orginally trained by krevas. This is a Base model.

Download Copy S3 URI

How to use

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

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

data = spark.createDataFrame([["나는 Spark NLP를 좋아합니다"]]).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_finance_koelectra_base_generator","ko") 
    .setInputCols(Array("document", "token")) 
    .setOutputCol("embeddings")

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

val data = Seq("나는 Spark NLP를 좋아합니다").toDF("text")

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

Model Information

Model Name: electra_embeddings_finance_koelectra_base_generator
Compatibility: Spark NLP 3.4.4+
License: Open Source
Edition: Official
Input Labels: [sentence, token]
Output Labels: [embeddings]
Language: ko
Size: 129.8 MB
Case sensitive: true

References

  • https://huggingface.co/krevas/finance-koelectra-base-generator
  • https://openreview.net/forum?id=r1xMH1BtvB
  • https://github.com/google-research/electra