English RobertaForMaskedLM Base Cased model (from model-attribution-challenge)

Description

Pretrained RobertaForMaskedLM model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. roberta-base is a English model originally trained by model-attribution-challenge.

Live Demo Open in Colab Download Copy S3 URI

How to use

documentAssembler = DocumentAssembler() \
    .setInputCol("text") \
    .setOutputCol("document")

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

roberta_loaded = RoBertaEmbeddings.pretrained("roberta_embeddings_model_attribution_challenge_base","en") \
    .setInputCols(["document", "token"]) \
    .setOutputCol("embeddings") \
    .setCaseSensitive(True)

pipeline = Pipeline(stages=[documentAssembler, tokenizer, roberta_loaded])

data = spark.createDataFrame([["I love Spark NLP"]]).toDF("text")

result = pipeline.fit(data).transform(data)

val documentAssembler = new DocumentAssembler()
    .setInputCol("text")
    .setOutputCol("document")

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

val roberta_loaded = RoBertaEmbeddings.pretrained("roberta_embeddings_model_attribution_challenge_base","en")
    .setInputCols(Array("document", "token"))
    .setOutputCol("embeddings")
    .setCaseSensitive(true)

val pipeline = new Pipeline().setStages(Array(documentAssembler, tokenizer, roberta_loaded))

val data = Seq("I love Spark NLP").toDS.toDF("text")

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

import nlu
nlu.load("en.embed.roberta.base.by_model_attribution_challenge").predict("""I love Spark NLP""")

Model Information

References

• https://huggingface.co/model-attribution-challenge/roberta-base
• https://arxiv.org/abs/1907.11692
• https://github.com/pytorch/fairseq/tree/master/examples/roberta
• https://yknzhu.wixsite.com/mbweb
• https://en.wikipedia.org/wiki/English_Wikipedia
• https://commoncrawl.org/2016/10/news-dataset-available/
• https://github.com/jcpeterson/openwebtext
• https://arxiv.org/abs/1806.02847