Edit on GitHub
  • John Snow Labs
  • Nov 15, 2022

Spanish Skipgram Legal Fast Text Embeddings (Uncased, D50)

  • fasttext
  • es
  • spanish
  • embeddings
  • open_source

Description

Word Embeddings lookup annotator that maps tokens to vectors. In the Skip-gram model, the distributed representation of the input word is used to predict the context.

Download Copy S3 URI

How to use

 
model = WordEmbeddingsModel.pretrained("word2vec_skipgram_legal_d50_uncased","es")\
	            .setInputCols(["document","token"])\
	            .setOutputCol("word_embeddings")



val model = WordEmbeddingsModel.pretrained("word2vec_skipgram_legal_d50_uncased","es")
	                .setInputCols("document","token")
	                .setOutputCol("word_embeddings")


import nlu
nlu.load("es.embed.legal.skipgram.uncased_d50").predict("""Put your text here.""")

Model Information

Model Name: word2vec_skipgram_legal_d50_uncased
Type: embeddings
Compatibility: Spark NLP 4.2.1+
License: Open Source
Edition: Official
Input Labels: [document, token]
Output Labels: [embeddings]
Language: es
Size: 172.2 MB
Case sensitive: false
Dimension: 100

References

https://zenodo.org/record/5036147#.Y3Op0XZBxD-