English XLMRobertaForTokenClassification Large Uncased model (from tner)

Description

Pretrained XLMRobertaForTokenClassification model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. xlm-roberta-large-uncased-all-english is a English model originally trained by tner.

Predicted Entities

actor, time, corporation, ordinal number, cardinal number, restaurant, director, rna, geopolitical area, rating, product, percent, protein, plot, dna, disease, cell line, law, soundtrack, quote, date, other, origin, amenity, chemical, event, cuisine, dish, work of art, cell type, location, genre, language, quantity, award, character name, facility, organization, relationship, opinion, group, money, person

Live Demo Open in Colab Download Copy S3 URI

How to use

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

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

token_classifier = XlmRoBertaForTokenClassification.pretrained("xlmroberta_ner_large_uncased_all_english","en") \
    .setInputCols(["document", "token"]) \
    .setOutputCol("ner")

ner_converter = NerConverter()\
    .setInputCols(["document", "token", "ner"])\
    .setOutputCol("ner_chunk")

pipeline = Pipeline(stages=[documentAssembler, tokenizer, token_classifier, ner_converter])

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

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

val documentAssembler = new DocumentAssembler()
      .setInputCols(Array("text"))
      .setOutputCols(Array("document"))

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

val token_classifier = XlmRoBertaForTokenClassification.pretrained("xlmroberta_ner_large_uncased_all_english","en")
    .setInputCols(Array("document", "token"))
    .setOutputCol("ner")

val ner_converter = new NerConverter()
    .setInputCols(Array("document", "token', "ner"))
    .setOutputCol("ner_chunk")

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

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

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

import nlu
nlu.load("en.ner.xlmr_roberta.all_english.uncased_large.by_tner").predict("""PUT YOUR STRING HERE""")

Model Information

References

• https://huggingface.co/tner/xlm-roberta-large-uncased-all-english
• https://github.com/asahi417/tner