Edit on GitHub
  • John Snow Labs
  • Apr 05, 2023

BERT Zero-Shot Classification Base - XLNI (bert_base_cased_zero_shot_classifier_xnli)

  • bert
  • zero_shot
  • xnli
  • en
  • english
  • base
  • open_source
  • tensorflow

Description

This model is intended to be used for zero-shot text classification, especially in English. It is fine-tuned on XNLI by using BERT Base Case model.

BertForZeroShotClassification using a ModelForSequenceClassification trained on NLI (natural language inference) tasks. Equivalent of BertForSequenceClassification models, but these models don’t require a hardcoded number of potential classes, they can be chosen at runtime. It usually means it’s slower but it is much more flexible.

We used TFBertForSequenceClassification to train this model and used BertForZeroShotClassification annotator in Spark NLP 🚀 for prediction at scale!

Predicted Entities

Download Copy S3 URI

How to use

document_assembler = DocumentAssembler() \
.setInputCol('text') \
.setOutputCol('document')

tokenizer = Tokenizer() \
.setInputCols(['document']) \
.setOutputCol('token')

zeroShotClassifier = BertForZeroShotClassification \
.pretrained('bert_base_cased_zero_shot_classifier_xnli', 'en') \
.setInputCols(['token', 'document']) \
.setOutputCol('class') \
.setCaseSensitive(True) \
.setMaxSentenceLength(512) \
.setCandidateLabels(["urgent", "mobile", "travel", "movie", "music", "sport", "weather", "technology"])

pipeline = Pipeline(stages=[
document_assembler,
tokenizer,
zeroShotClassifier
])

example = spark.createDataFrame([['I have a problem with my iphone that needs to be resolved asap!!']]).toDF("text")
result = pipeline.fit(example).transform(example)

val document_assembler = DocumentAssembler()
.setInputCol("text")
.setOutputCol("document")

val tokenizer = Tokenizer()
.setInputCols("document")
.setOutputCol("token")

val zeroShotClassifier = BertForSequenceClassification.pretrained("bert_base_cased_zero_shot_classifier_xnli", "en")
.setInputCols("document", "token")
.setOutputCol("class")
.setCaseSensitive(true)
.setMaxSentenceLength(512)
.setCandidateLabels(Array("urgent", "mobile", "travel", "movie", "music", "sport", "weather", "technology"))

val pipeline = new Pipeline().setStages(Array(document_assembler, tokenizer, zeroShotClassifier))

val example = Seq("I have a problem with my iphone that needs to be resolved asap!!").toDS.toDF("text")

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

Model Information

Model Name: bert_base_cased_zero_shot_classifier_xnli
Compatibility: Spark NLP 4.4.0+
License: Open Source
Edition: Official
Input Labels: [token, document]
Output Labels: [class]
Language: en
Size: 406.5 MB
Case sensitive: true
Max sentence length: 512

References

https://huggingface.co/datasets/xnli