Description
DeBertaForTokenClassification can load DeBERTA Models v2 and v3 with a token classification head on top (a linear layer on top of the hidden-states output) e.g. for Named-Entity-Recognition (NER) tasks.
deberta_v3_xsmall_token_classifier_conll03 is a fine-tuned DeBERTa model that is ready to be used for Token Classification task such as Named Entity Recognition and it achieves state-of-the-art performance.
We used TFDebertaV2ForTokenClassification to train this model and used DeBertaForTokenClassification annotator in Spark NLP 🚀 for prediction at scale! This model has been trained to recognize four types of entities: location (LOC), organizations (ORG), person (PER), and Miscellaneous (MISC).
How to use
document_assembler = DocumentAssembler()\
.setInputCol("text")\
.setOutputCol("document")
tokenizer = Tokenizer()\
.setInputCols(['document'])\
.setOutputCol('token')
tokenClassifier = DeBertaForTokenClassification.pretrained("deberta_v3_xsmall_token_classifier_conll03", "en")\
.setInputCols(["document", "token"])\
.setOutputCol("ner")\
.setCaseSensitive(True)\
.setMaxSentenceLength(512)
# since output column is IOB/IOB2 style, NerConverter can extract entities
ner_converter = NerConverter()\
.setInputCols(['document', 'token', 'ner'])\
.setOutputCol('entities')
pipeline = Pipeline(stages=[
document_assembler,
tokenizer,
tokenClassifier,
ner_converter
])
example = spark.createDataFrame([['I really liked that movie!']]).toDF("text")
result = pipeline.fit(example).transform(example)
val document_assembler = new DocumentAssembler()
.setInputCol("text")
.setOutputCol("document")
val tokenizer = new Tokenizer()
.setInputCols("document")
.setOutputCol("token")
val tokenClassifier = DeBertaForTokenClassification.pretrained("deberta_v3_xsmall_token_classifier_conll03", "en")
.setInputCols("document", "token")
.setOutputCol("ner")
.setCaseSensitive(true)
.setMaxSentenceLength(512)
// since output column is IOB/IOB2 style, NerConverter can extract entities
val ner_converter = NerConverter()
.setInputCols("document", "token", "ner")
.setOutputCol("entities")
val pipeline = new Pipeline().setStages(Array(document_assembler, tokenizer, tokenClassifier, ner_converter))
val example = Seq("I really liked that movie!").toDS.toDF("text")
val result = pipeline.fit(example).transform(example)
import nlu
nlu.load("en.ner.debertav3_xsmall.conll03").predict("""I really liked that movie!""")
Model Information
| Model Name: | deberta_v3_xsmall_token_classifier_conll03 |
| Compatibility: | Spark NLP 3.4.4+ |
| License: | Open Source |
| Edition: | Official |
| Input Labels: | [token, document] |
| Output Labels: | [ner] |
| Language: | en |
| Size: | 224.9 MB |
| Case sensitive: | true |
| Max sentence length: | 512 |