Description
Pretrained RoBertaForZeroShotClassification model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP.nli_minilm2_l6_h768 is a English model originally trained by cross-encoder.
How to use
documentAssembler = DocumentAssembler() \
.setInputCol('text') \
.setOutputCol('document')
tokenizer = Tokenizer() \
.setInputCols(['document']) \
.setOutputCol('token')
zeroShotClassifier = RoBertaForZeroShotClassification.pretrained("nli_minilm2_l6_h768","en") \
.setInputCols(["document","token"]) \
.setOutputCol("class")
pipeline = Pipeline().setStages([documentAssembler, tokenizer, zeroShotClassifier])
data = spark.createDataFrame([["I love spark-nlp"]]).toDF("text")
pipelineModel = pipeline.fit(data)
pipelineDF = pipelineModel.transform(data)
val documentAssembler = new DocumentAssembler()
.setInputCols("text")
.setOutputCols("document")
val tokenizer = new Tokenizer()
.setInputCols("document")
.setOutputCol("token")
val zeroShotClassifier = RoBertaForZeroShotClassification.pretrained("nli_minilm2_l6_h768", "en")
.setInputCols(Array("documents","token"))
.setOutputCol("class")
val pipeline = new Pipeline().setStages(Array(documentAssembler, tokenizer, zeroShotClassifier))
val data = Seq("I love spark-nlp").toDS.toDF("text")
val pipelineModel = pipeline.fit(data)
val pipelineDF = pipelineModel.transform(data)
Model Information
| Model Name: | nli_minilm2_l6_h768 |
| Compatibility: | Spark NLP 5.5.1+ |
| License: | Open Source |
| Edition: | Official |
| Input Labels: | [document, token] |
| Output Labels: | [class] |
| Language: | en |
| Size: | 306.5 MB |
References
https://huggingface.co/cross-encoder/nli-MiniLM2-L6-H768