nok/sklearn-porter

Port CountVectorizer

nicolalandro opened this issue · 3 comments

nicolalandro commented

For text mining it's important to fit also a CountVectorizer (or a TFIDFTransformer), so should be possible to export it in the targhet lenguage

Opdoop commented

Currently, is there any alternative way?

nicolalandro commented

@Opdoop the way is to extract the dictionary from CountVectorizer (get_feature_names()) and reimplement the logic manually.
You can do the same thing on TFIDF.

  • 👍1
nok commented

Hello @nicolalandro ,

can you provide some snippets? Withit I will have a better start.

Kind regards,
Darius