Description
Pretrained BertForTokenClassification model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP.bert_large_portuguese_ner_enamex is a Portuguese model originally trained by marcosgg.
How to use
documentAssembler = DocumentAssembler() \
.setInputCol("text") \
.setOutputCol("documents")
tokenClassifier = BertForTokenClassification.pretrained("bert_large_portuguese_ner_enamex","pt") \
.setInputCols(["documents","token"]) \
.setOutputCol("ner")
pipeline = Pipeline().setStages([documentAssembler, tokenClassifier])
pipelineModel = pipeline.fit(data)
pipelineDF = pipelineModel.transform(data)
val documentAssembler = new DocumentAssembler()
.setInputCol("text")
.setOutputCol("embeddings")
val tokenClassifier = BertForTokenClassification
.pretrained("bert_large_portuguese_ner_enamex", "pt")
.setInputCols(Array("documents","token"))
.setOutputCol("ner")
val pipeline = new Pipeline().setStages(Array(documentAssembler, tokenClassifier))
val pipelineModel = pipeline.fit(data)
val pipelineDF = pipelineModel.transform(data)
Model Information
| Model Name: | bert_large_portuguese_ner_enamex |
| Compatibility: | Spark NLP 5.2.0+ |
| License: | Open Source |
| Edition: | Official |
| Input Labels: | [documents, token] |
| Output Labels: | [ner] |
| Language: | pt |
| Size: | 1.2 GB |
References
https://huggingface.co/marcosgg/bert-large-pt-ner-enamex