Hausa BertForMaskedLM Base Cased model (from Davlan)

Description

Pretrained BertForMaskedLM model, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP. bert-base-multilingual-cased-finetuned-swahili is a Hausa model originally trained by Davlan.

Download Copy S3 URI

How to use

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

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

bert_loaded = BertEmbeddings.pretrained("bert_embeddings_base_multilingual_cased_finetuned_swahili","ha") \
    .setInputCols(["document", "token"]) \
    .setOutputCol("embeddings") \
    .setCaseSensitive(True)

pipeline = Pipeline(stages=[documentAssembler, tokenizer, bert_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 bert_loaded = BertEmbeddings.pretrained("bert_embeddings_base_multilingual_cased_finetuned_swahili","ha")
    .setInputCols(Array("document", "token"))
    .setOutputCol("embeddings")
    .setCaseSensitive(True)

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

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

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

import nlu
nlu.load("ha.embed.bert.cased_multilingual_base_finetuned.by_davlan").predict("""I love Spark NLP""")

Model Information

Model Name:	bert_embeddings_base_multilingual_cased_finetuned_swahili
Compatibility:	Spark NLP 4.2.4+
License:	Open Source
Edition:	Official
Input Labels:	[sentence, token]
Output Labels:	[bert]
Language:	ha
Size:	666.9 MB
Case sensitive:	true

References

https://huggingface.co/Davlan/bert-base-multilingual-cased-finetuned-swahili
http://data.statmt.org/cc-100/
https://github.com/masakhane-io/masakhane-ner

PREVIOUSHausa BertForMaskedLM Base Cased model (from Davlan)

NEXTYoruba BertForMaskedLM Base Cased model (from Davlan)