Description
Pretrained Question Answering model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. bert-large-japanese-wikipedia-ud-head is a Japanese model originally trained by KoichiYasuoka.
How to use
documentAssembler = MultiDocumentAssembler() \
.setInputCols(["question", "context"]) \
.setOutputCols(["document_question", "document_context"])
spanClassifier = BertForQuestionAnswering.pretrained("bert_qa_large_japanese_wikipedia_ud_head","ja") \
.setInputCols(["document_question", "document_context"]) \
.setOutputCol("answer")\
.setCaseSensitive(True)
pipeline = Pipeline(stages=[documentAssembler, spanClassifier])
data = spark.createDataFrame([["私の名前は何ですか？", "私の名前はクララで、私はバークレーに住んでいます。"]]).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 = BertForQuestionAnswering.pretrained("bert_qa_large_japanese_wikipedia_ud_head","ja")
.setInputCols(Array("document", "token"))
.setOutputCol("answer")
.setCaseSensitive(true)
val pipeline = new Pipeline().setStages(Array(documentAssembler, spanClassifier))
val data = Seq("私の名前は何ですか？", "私の名前はクララで、私はバークレーに住んでいます。").toDF("question", "context")
val result = pipeline.fit(data).transform(data)
import nlu
nlu.load("ja.answer_question.wikipedia.bert.large").predict("""私の名前は何ですか？|||"私の名前はクララで、私はバークレーに住んでいます。""")
Model Information
| Model Name: | bert_qa_large_japanese_wikipedia_ud_head |
| Compatibility: | Spark NLP 4.0.0+ |
| License: | Open Source |
| Edition: | Official |
| Input Labels: | [document_question, document_context] |
| Output Labels: | [answer] |
| Language: | ja |
| Size: | 1.2 GB |
| Case sensitive: | true |
| Max sentence length: | 512 |
References
- https://huggingface.co/KoichiYasuoka/bert-large-japanese-wikipedia-ud-head
- https://github.com/UniversalDependencies/UD_Japanese-GSDLUW