French Bert Embeddings (from Geotrend)


Pretrained Bert Embeddings model, uploaded to Hugging Face, adapted and imported into Spark NLP. bert-base-fr-cased is a French model orginally trained by Geotrend.


How to use

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

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

data = spark.createDataFrame([["J'adore Spark Nlp"]]).toDF("text")

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

val embeddings = BertEmbeddings.pretrained("bert_embeddings_bert_base_fr_cased","fr") 
    .setInputCols(Array("document", "token")) 

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

val data = Seq("J'adore Spark Nlp").toDF("text")

val result =
import nlu
nlu.load("fr.embed.bert_base_fr_cased").predict("""J'adore Spark Nlp""")

Model Information

Model Name: bert_embeddings_bert_base_fr_cased
Compatibility: Spark NLP 3.4.2+
License: Open Source
Edition: Official
Input Labels: [sentence, token]
Output Labels: [bert]
Language: fr
Size: 393.7 MB
Case sensitive: true