Portuguese BertForTokenClassification Cased model (from dominguesm)

Description

Pretrained BertForSequenceClassification model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. bert-restore-punctuation-ptbr is a Portuguese model originally trained by dominguesm.

Predicted Entities

,O, ,U, OO, :O, ;O, .O, ?O, ?U, OU, !U, !O, -O, :U, .U, 'O

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 = BertForTokenClassification.pretrained("bert_token_classifier_restore_punctuation_ptbr","pt") \
    .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 = BertForTokenClassification.pretrained("bert_token_classifier_restore_punctuation_ptbr","pt")
    .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)

import nlu
nlu.load("pt.ner.bert.by_dominguesm").predict("""PUT YOUR STRING HERE""")

Model Information

References

• https://huggingface.co/dominguesm/bert-restore-punctuation-ptbr
• https://wandb.ai/dominguesm/RestorePunctuationPTBR
• https://github.com/DominguesM/respunct
• https://github.com/esdurmus/Wikilingua
• https://paperswithcode.com/sota?task=named-entity-recognition&dataset=wiki_lingua