Russian DistilBertForQuestionAnswering model (from AndrewChar)

Description

Pretrained Question Answering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. model-QA-5-epoch-RU is a Russian model originally trained by AndrewChar.

Download Copy S3 URI

How to use

documentAssembler = MultiDocumentAssembler() \
.setInputCols(["question", "context"]) \
.setOutputCols(["document_question", "document_context"])

spanClassifier = DistilBertForQuestionAnswering.pretrained("distilbert_qa_model_QA_5_epoch_RU","ru") \
.setInputCols(["document_question", "document_context"]) \
.setOutputCol("answer")\
.setCaseSensitive(True)

pipeline = Pipeline(stages=[documentAssembler, spanClassifier])

data = spark.createDataFrame([["Как меня зовут?", "Меня зовут Клара, и я живу в Беркли."]]).toDF("question", "context")

result = pipeline.fit(data).transform(data)
val documentAssembler = new MultiDocumentAssembler() 
.setInputCols(Array("question", "context")) 
.setOutputCols(Array("document_question", "document_context"))

val spanClassifer = DistilBertForQuestionAnswering.pretrained("distilbert_qa_model_QA_5_epoch_RU","ru") 
.setInputCols(Array("document", "token")) 
.setOutputCol("answer")
.setCaseSensitive(true)

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

val data = Seq("Как меня зовут?", "Меня зовут Клара, и я живу в Беркли.").toDF("question", "context")

val result = pipeline.fit(data).transform(data)
import nlu
nlu.load("ru.answer_question.distil_bert").predict("""Как меня зовут?|||"Меня зовут Клара, и я живу в Беркли.""")

Model Information

Model Name: distilbert_qa_model_QA_5_epoch_RU
Compatibility: Spark NLP 4.0.0+
License: Open Source
Edition: Official
Input Labels: [document_question, document_context]
Output Labels: [answer]
Language: ru
Size: 505.7 MB
Case sensitive: true
Max sentence length: 512

References

  • https://huggingface.co/AndrewChar/model-QA-5-epoch-RU