DistilBERTZero-Shot Classification Base - distilbert_base_zero_shot_classifier_turkish_cased_allnli

Description

This model is intended to be used for zero-shot text classification, especially in Trukish. It is fine-tuned on MNLI by using DistilBERT Base Uncased model.

DistilBertForZeroShotClassification using a ModelForSequenceClassification trained on NLI (natural language inference) tasks. Equivalent of DistilBertForSequenceClassification 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 TFDistilBertForSequenceClassification to train this model and used DistilBertForZeroShotClassification annotator in Spark NLP 🚀 for prediction at scale!

Predicted Entities

Live Demo Open in Colab Download Copy S3 URI

How to use

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

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

zeroShotClassifier = DistilBertForZeroShotClassification \
.pretrained('distilbert_base_zero_shot_classifier_turkish_cased_allnli', 'en') \
.setInputCols(['token', 'document']) \
.setOutputCol('class') \
.setCaseSensitive(True) \
.setMaxSentenceLength(512) \
.setCandidateLabels(["olumsuz", "olumlu"])

pipeline = Pipeline(stages=[
document_assembler,
tokenizer,
zeroShotClassifier
])
example = spark.createDataFrame([['Senaryo çok saçmaydı, beğendim diyemem.']]).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 = DistilBertForZeroShotClassification.pretrained("distilbert_base_zero_shot_classifier_turkish_cased_allnli", "en")
.setInputCols("document", "token")
.setOutputCol("class")
.setCaseSensitive(true)
.setMaxSentenceLength(512)
.setCandidateLabels(Array("olumsuz", "olumlu"))

val pipeline = new Pipeline().setStages(Array(document_assembler, tokenizer, zeroShotClassifier))
val example = Seq("Senaryo çok saçmaydı, beğendim diyemem.").toDS.toDF("text")
val result = pipeline.fit(example).transform(example)

Model Information