i am following the starter pack but i change the language on spicy to spanish for doing some tests but now in spanish never show a correct intent, i installed spicy and the language package and i modified the intents with spanish language i need to do something more?
Hi! I use a example to spanish
the spacy model isnāt amazing for spanish, consider using the tensorflw_embedding pipeline
Does somebody has been created an spanish chatbot? im new and im interested in create an spanish chatbot
I am trying to create one for the moment is on development process, but Tensorflow works fine and i recommend you to use some generator for the training data.
Hola Ramon, tambiƩn estoy construyendo un bot en espaƱol, ya logre entrenar el interprete con NLU y el motor de diƔlogos, pero tengo problemas al momento de invocar el interpreter desde el gestor de diƔlogos. has tenido un problema similar?
Hey buenas. Yo tambiƩn ando implementando un bot en espaƱol. Si les apetece podemos formar un grupo de facebook o whatsapp y ayudarnos unos a los otros!
2 Likes
Good Idea, i am taking on the initiativeā¦
Hello akelad,
I am using tensorflow but I am having issues with spanish accent. Do you know how do I take care of spanish accent too? Quick pointers would be helpful.
Hi @mpbalshetwar you can try using the strip_accents: True setting for the CountVectorFeaturizer.
Hey @akelad,
thanks for your reply. We are getting following warning.
rasa.nlu.featurizers.sparse_featurizer.count_vectors_featurizer - Unable to train CountVectorizer for message attribute text. Leaving an untrained CountVectorizer for it.
Somewhere I was reading, this is just a warning and can be ignored. But the thread was in different context and so I want to confirm with you.
Hmm, not sure thatās safe to ignore. Can you post the full training logs? And also point me to the post which says it can be ignored?
Spanish group on telegram: Telegram: Contact @rasaEsp