Edit on GitHub
  • John Snow Labs
  • Apr 26, 2022

Ukrainian Part of Speech Tagger (Base)

  • bert
  • pos
  • part_of_speech
  • uk
  • open_source

Description

Pretrained Part of Speech model, uploaded to Hugging Face, adapted and imported into Spark NLP. bert-base-slavic-cyrillic-upos is a Ukrainian model orginally trained by KoichiYasuoka.

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_pos_bert_base_slavic_cyrillic_upos","uk") \
.setInputCols(["sentence", "token"]) \
.setOutputCol("pos")

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

data = spark.createDataFrame([["Я люблю Spark NLP"]]).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_pos_bert_base_slavic_cyrillic_upos","uk") 
.setInputCols(Array("sentence", "token")) 
.setOutputCol("pos")

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

val data = Seq("Я люблю Spark NLP").toDF("text")

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

import nlu
nlu.load("uk.pos.bert_base_slavic_cyrillic_upos").predict("""Я люблю Spark NLP""")

Model Information

Model Name: bert_pos_bert_base_slavic_cyrillic_upos
Compatibility: Spark NLP 3.4.2+
License: Open Source
Edition: Official
Input Labels: [document, token]
Output Labels: [ner]
Language: uk
Size: 668.2 MB
Case sensitive: true
Max sentence length: 128

References

  • https://huggingface.co/KoichiYasuoka/bert-base-slavic-cyrillic-upos
  • https://universaldependencies.org/be/
  • https://universaldependencies.org/bg/
  • https://universaldependencies.org/ru/
  • https://universaldependencies.org/treebanks/sr_set/
  • https://universaldependencies.org/treebanks/uk_iu/
  • https://universaldependencies.org/u/pos/
  • https://github.com/KoichiYasuoka/esupar