English Bert Embeddings (from lordtt13)

Description

Pretrained Bert Embeddings model, uploaded to Hugging Face, adapted and imported into Spark NLP. COVID-SciBERT is a English model orginally trained by lordtt13.

Download

How to use

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

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

data = spark.createDataFrame([["I love 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("bert_embeddings_COVID_SciBERT","en") 
    .setInputCols(Array("document", "token")) 
    .setOutputCol("embeddings")

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

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

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

Model Information

Model Name: bert_embeddings_COVID_SciBERT
Compatibility: Spark NLP 3.4.2+
License: Open Source
Edition: Official
Input Labels: [sentence, token]
Output Labels: [bert]
Language: en
Size: 415.3 MB
Case sensitive: true

References

  • https://huggingface.co/lordtt13/COVID-SciBERT
  • https://arxiv.org/abs/1903.10676
  • https://www.kaggle.com/allen-institute-for-ai/CORD-19-research-challenge
  • https://github.com/lordtt13/word-embeddings/blob/master/COVID-19%20Research%20Data/COVID-SciBERT.ipynb
  • https://github.com/lordtt13
  • https://www.linkedin.com/in/tanmay-thakur-6bb5a9154/