Edit on GitHub
  • John Snow Labs
  • May 09, 2022

Persian Named Entity Recognition (from HooshvareLab)

  • bert
  • ner
  • token_classification
  • fa
  • open_source

Description

Pretrained Named Entity Recognition model, uploaded to Hugging Face, adapted and imported into Spark NLP. bert-fa-base-uncased-ner-arman is a Persian model orginally trained by HooshvareLab.

Predicted Entities

fac, pers, pro, event, org, loc

Download Copy S3 URI

How to use

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

sentenceDetector = SentenceDetectorDLModel.pretrained("sentence_detector_dl", "xx")\
       .setInputCols(["document"])\
       .setOutputCol("sentence")

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

tokenClassifier = BertForTokenClassification.pretrained("bert_ner_bert_fa_base_uncased_ner_arman","fa") \
    .setInputCols(["sentence", "token"]) \
    .setOutputCol("ner")

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

data = spark.createDataFrame([["من عاشق جرقه nlp هستم"]]).toDF("text")

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

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

val sentenceDetector = SentenceDetectorDLModel.pretrained("sentence_detector_dl", "xx")
       .setInputCols(Array("document"))
       .setOutputCol("sentence")

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

val tokenClassifier = BertForTokenClassification.pretrained("bert_ner_bert_fa_base_uncased_ner_arman","fa") 
    .setInputCols(Array("sentence", "token")) 
    .setOutputCol("ner")

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

val data = Seq("من عاشق جرقه nlp هستم").toDF("text")

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

Model Information

Model Name: bert_ner_bert_fa_base_uncased_ner_arman
Compatibility: Spark NLP 3.4.2+
License: Open Source
Edition: Official
Input Labels: [document, token]
Output Labels: [ner]
Language: fa
Size: 607.1 MB
Case sensitive: true
Max sentence length: 128

References

  • https://huggingface.co/HooshvareLab/bert-fa-base-uncased-ner-arman
  • https://github.com/hooshvare/parsbert
  • https://github.com/HaniehP/PersianNER
  • https://github.com/hooshvare/parsbert-ner/blob/master/persian-ner-pipeline.ipynb
  • https://colab.research.google.com/github/hooshvare/parsbert-ner/blob/master/persian-ner-pipeline.ipynb
  • https://github.com/hooshvare/parsbert/issues