Description
Pretrained DistilBertForTokenClassification model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. icelandic-ner-distilbert is a Icelandic model originally trained by m3hrdadfi.
Predicted Entities
Money, Date, Time, Percent, Miscellaneous, Location, Person, Organization
How to use
documentAssembler = DocumentAssembler() \
.setInputCol("text") \
.setOutputCol("document")
tokenizer = Tokenizer() \
.setInputCols("document") \
.setOutputCol("token")
tokenClassifier = DistilBertForTokenClassification.pretrained("distilbert_token_classifier_icelandic_ner_distilbert","is") \
.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 = DistilBertForTokenClassification.pretrained("distilbert_token_classifier_icelandic_ner_distilbert","is")
.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)
Model Information
| Model Name: | distilbert_token_classifier_icelandic_ner_distilbert |
| Compatibility: | Spark NLP 4.3.1+ |
| License: | Open Source |
| Edition: | Official |
| Input Labels: | [document, token] |
| Output Labels: | [ner] |
| Language: | is |
| Size: | 505.8 MB |
| Case sensitive: | true |
| Max sentence length: | 128 |
References
- https://huggingface.co/m3hrdadfi/icelandic-ner-distilbert
- http://hdl.handle.net/20.500.12537/42
- https://en.ru.is/
- https://github.com/m3hrdadfi/icelandic-ner/issues