Chinese BertForTokenClassification Base Cased model (from ckiplab)

Description

Pretrained BertForTokenClassification model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. bert-base-chinese-ner is a Chinese model originally trained by ckiplab.

Predicted Entities

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

Live Demo Open in Colab 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_base_chinese_ner","zh") \
    .setInputCols(["sentence", "token"]) \
    .setOutputCol("ner")

pipeline = Pipeline(stages=[documentAssembler, sentenceDetector, 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 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_base_chinese_ner","zh") 
    .setInputCols(Array("sentence", "token")) 
    .setOutputCol("ner")

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

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

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

import nlu
nlu.load("zh.ner.bert.base").predict("""PUT YOUR STRING HERE""")

Model Information

References

• https://huggingface.co/ckiplab/bert-base-chinese-ner
• https://github.com/ckiplab/ckip-transformers
• https://ckip.iis.sinica.edu.tw
• https://muyang.pro