Sundanese RoBERTa Embeddings (from w11wo)

Description

Pretrained RoBERTa Embeddings model, uploaded to Hugging Face, adapted and imported into Spark NLP. sundanese-roberta-base is a Sundanese model orginally trained by w11wo.

Live Demo Open in Colab Download Copy S3 URICopied!

How to use

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

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

embeddings = RoBertaEmbeddings.pretrained("roberta_embeddings_sundanese_roberta_base","su") \
.setInputCols(["document", "token"]) \
.setOutputCol("embeddings")

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

data = spark.createDataFrame([["Abdi bogoh Spark NLP"]]).toDF("text")

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

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

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

val embeddings = RoBertaEmbeddings.pretrained("roberta_embeddings_sundanese_roberta_base","su") 
.setInputCols(Array("document", "token")) 
.setOutputCol("embeddings")

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

val data = Seq("Abdi bogoh Spark NLP").toDF("text")

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

import nlu
nlu.load("su.embed.sundanese_roberta_base").predict("""Abdi bogoh Spark NLP""")

Model Information

References

• https://huggingface.co/w11wo/sundanese-roberta-base
• https://arxiv.org/abs/1907.11692
• https://hf.co/datasets/oscar
• https://hf.co/datasets/mc4
• https://hf.co/datasets/cc100
• https://su.wikipedia.org/
• https://hf.co/w11wo/sundanese-roberta-base/tree/main
• https://hf.co/w11wo/sundanese-roberta-base/tensorboard
• https://w11wo.github.io/