prudant
/

es_intent_classification

Text Classification

sklearn and sentence transformers

Not-For-All-Audiences

Model card Files Files and versions Community

prudant commited on Jul 2

Commit

19b7d59

•

1 Parent(s): 9cf7923

Update README.md

Files changed (1) hide show

README.md +1 -0

README.md CHANGED Viewed

@@ -17,6 +17,7 @@ This repository contain the implementation of an efficient and fast user intent
 ## Limitations
 This model was train on spanish text corpus and only for text that represents question or requests, it wont perform well with multi turn chathistory or multiline long texts, it's recommended for 128< len token texts. Also the dataset is not big enought to generalize well, but the intention of this repository is to lay the foundations for an assembly of models that, with a larger dataset, should generalize very well for a real use case. DO NOT USE ON PRODUCTIVE ENVIRONMENTS.
 `If you build a larger dataset with longer text secuences or multi turn conversations and train the model, it should be work pretty well, the Jina embedding model support up to 8k tokens =)`
 ## Performance Note

 ## Limitations
 This model was train on spanish text corpus and only for text that represents question or requests, it wont perform well with multi turn chathistory or multiline long texts, it's recommended for 128< len token texts. Also the dataset is not big enought to generalize well, but the intention of this repository is to lay the foundations for an assembly of models that, with a larger dataset, should generalize very well for a real use case. DO NOT USE ON PRODUCTIVE ENVIRONMENTS.
 `If you build a larger dataset with longer text secuences or multi turn conversations and train the model, it should be work pretty well, the Jina embedding model support up to 8k tokens =)`
 ## Performance Note