Dutch RobertaForMaskedLM Base Cased model (from pdelobelle)

Description

Pretrained RobertaForMaskedLM model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. robbert-v2-dutch-base is a Dutch model originally trained by pdelobelle.

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_pdelobelle_robbert_v2_dutch_base","nl") \
    .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_pdelobelle_robbert_v2_dutch_base","nl")
    .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("nl.embed.roberta.conll.v2_base").predict("""I love Spark NLP""")

Model Information

References

• https://huggingface.co/pdelobelle/robbert-v2-dutch-base
• https://github.com/iPieter/RobBERT
• https://scholar.google.com/scholar?oi=bibs&hl=en&cites=7180110604335112086
• https://www.aclweb.org/anthology/2021.wassa-1.27/
• https://arxiv.org/pdf/2001.06286.pdf
• https://biblio.ugent.be/publication/8704637/file/8704638.pdf
• https://arxiv.org/pdf/2001.06286.pdf
• https://arxiv.org/pdf/2001.06286.pdf
• https://arxiv.org/pdf/2004.02814.pdf
• https://github.com/proycon/deepfrog
• https://arxiv.org/pdf/2001.06286.pdf
• https://github.com/proycon/deepfrog
• https://arxiv.org/pdf/2001.06286.pdf
• https://arxiv.org/pdf/2010.13652.pdf
• https://www.cambridge.org/core/journals/natural-language-engineering/article/abs/automatic-classification-of-participant-roles-in-cyberbullying-can-we-detect-victims-bullies-and-bystanders-in-social-media-text/A2079C2C738C29428E666810B8903342
• https://gitlab.com/spelfouten/dutch-simpletransformers/
• https://arxiv.org/pdf/2101.05716.pdf
• https://medium.com/broadhorizon-cmotions/nlp-with-r-part-5-state-of-the-art-in-nlp-transformers-bert-3449e3cd7494
• https://people.cs.kuleuven.be/~pieter.delobelle/robbert/
• https://arxiv.org/abs/2001.06286
• https://github.com/iPieter/RobBERT
• https://arxiv.org/abs/1907.11692
• https://github.com/pytorch/fairseq/tree/master/examples/roberta
• https://people.cs.kuleuven.be/~pieter.delobelle/robbert/
• https://arxiv.org/abs/2001.06286
• https://github.com/iPieter/RobBERT
• https://github.com/benjaminvdb/110kDBRD
• https://www.statmt.org/europarl/
• https://arxiv.org/abs/2001.02943
• https://universaldependencies.org/treebanks/nl_lassysmall/index.html
• https://www.clips.uantwerpen.be/conll2002/ner/
• https://oscar-corpus.com/
• https://github.com/pytorch/fairseq/tree/master/examples/roberta
• https://github.com/pytorch/fairseq/tree/master/examples/roberta
• https://arxiv.org/abs/2001.06286
• https://github.com/iPieter/RobBERT#how-to-replicate-our-paper-experiments
• https://arxiv.org/abs/1909.11942
• https://camembert-model.fr/
• https://en.wikipedia.org/wiki/Robbert
• https://muppet.fandom.com/wiki/Bert
• https://github.com/iPieter/RobBERT/blob/master/res/robbert_logo.png
• https://people.cs.kuleuven.be/~pieter.delobelle
• https://thomaswinters.be
• https://people.cs.kuleuven.be/~bettina.berendt/