python - How to tell scikit-learn vectorizer use specific features? -


I have a set of features - these are not all the same words above the hand; Some of them are elderly and some others are tragrams. I want to prepare my own texts - which are clearly made available in the form of raw texts based on these characteristics, how can I do this in scalin? In this way I have defined my vector so far. Import from sklearn.feature_extraction.text CountVectorizer vectorizer = CountVectorizer (ngram_range = (1, 3)) back vector size

countvatiser and TfIdfVectorizer allows you to specify the terminology to use For them, give them as keyword logic in the form of glossary glossary : mapping or alternative, optional

either Mapping (e.g., A Word) where the key words and values ​​are not given if a notch, according to the feature indices matrix, or conditions, is defined a vocabulary of input documents.

Comments

Popular posts from this blog

python - how we can use ajax() in views.py in django? -

matlab - Using loops to get multiple values into a cell -

python - Sequence Pattern recognition with Pybrain -