Urdu RoBERTa Embeddings

Description

Pretrained RoBERTa Embeddings model, uploaded to Hugging Face, adapted and imported into Spark NLP. roberta-urdu-small is a Urdu model orginally trained by urduhack.

Download

How to use

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

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

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

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

data = spark.createDataFrame([["مجھے سپارک این ایل پی سے محبت ہے"]]).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_roberta_urdu_small","ur") 
.setInputCols(Array("document", "token")) 
.setOutputCol("embeddings")

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

val data = Seq("مجھے سپارک این ایل پی سے محبت ہے").toDF("text")

val result = pipeline.fit(data).transform(data)
import nlu
nlu.load("ur.embed.roberta_urdu_small").predict("""مجھے سپارک این ایل پی سے محبت ہے""")

Model Information

Model Name: roberta_embeddings_roberta_urdu_small
Compatibility: Spark NLP 3.4.2+
License: Open Source
Edition: Official
Input Labels: [sentence, token]
Output Labels: [bert]
Language: ur
Size: 473.9 MB
Case sensitive: true

References

  • https://huggingface.co/urduhack/roberta-urdu-small
  • https://github.com/urduhack/urduhack/blob/master/LICENSE
  • https://github.com/urduhack/urduhack