Marathi Bert Embeddings (from l3cube-pune)


Pretrained Bert Embeddings model, uploaded to Hugging Face, adapted and imported into Spark NLP. marathi-bert is a Marathi model orginally trained by l3cube-pune.


How to use

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

tokenizer = Tokenizer() \
    .setInputCols("document") \
embeddings = BertEmbeddings.pretrained("bert_embeddings_marathi_bert","mr") \
    .setInputCols(["document", "token"]) \
pipeline = Pipeline(stages=[documentAssembler, tokenizer, embeddings])

data = spark.createDataFrame([["मला स्पार्क एनएलपी आवडते"]]).toDF("text")

result =
val documentAssembler = new DocumentAssembler() 
val tokenizer = new Tokenizer() 

val embeddings = BertEmbeddings.pretrained("bert_embeddings_marathi_bert","mr") 
    .setInputCols(Array("document", "token")) 

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

val data = Seq("मला स्पार्क एनएलपी आवडते").toDF("text")

val result =
import nlu
nlu.load("mr.embed.marathi_bert").predict("""मला स्पार्क एनएलपी आवडते""")

Model Information

Model Name: bert_embeddings_marathi_bert
Compatibility: Spark NLP 3.4.2+
License: Open Source
Edition: Official
Input Labels: [sentence, token]
Output Labels: [bert]
Language: mr
Size: 668.0 MB
Case sensitive: true