Latin RobertaForTokenClassification Large Cased model (from tner)

Description

Pretrained RobertaForTokenClassification model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. roberta-large-ontonotes5 is a Latin model originally trained by tner.

Predicted Entities

NORP, FAC, QUANTITY, LOC, EVENT, CARDINAL, LANGUAGE, GPE, ORG, TIME, PERSON, WORK_OF_ART, DATE, PRODUCT, PERCENT, LAW, ORDINAL, MONEY

Live Demo Open in Colab Download Copy S3 URI

How to use

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

tokenizer = Tokenizer() \
    .setInputCols("document") \
    .setOutputCol("token")

tokenClassifier = RobertaForTokenClassification.pretrained("roberta_token_classifier_large_ontonotes5","la") \
    .setInputCols(["document", "token"]) \
    .setOutputCol("ner")

pipeline = Pipeline(stages=[documentAssembler, tokenizer, tokenClassifier])

data = spark.createDataFrame([["PUT YOUR STRING HERE"]]).toDF("text")

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

val documentAssembler = new DocumentAssembler()
    .setInputCol("text")
    .setOutputCol("document")

val tokenizer = new Tokenizer()
    .setInputCols("document")
    .setOutputCol("token")

val tokenClassifier = RobertaForTokenClassification.pretrained("roberta_token_classifier_large_ontonotes5","la")
    .setInputCols(Array("document", "token"))
    .setOutputCol("ner")

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

val data = Seq("PUT YOUR STRING HERE").toDS.toDF("text")

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

Model Information

References

• https://huggingface.co/tner/roberta-large-ontonotes5
• https://github.com/asahi417/tner
• https://github.com/asahi417/tner
• https://aclanthology.org/2021.eacl-demos.7/
• https://paperswithcode.com/sota?task=Token+Classification&dataset=tner%2Fontonotes5