machine learning - How to use different dataset for scikit and NLTK? -

I am trying to implement the inkled inexpensive Bias classifier for raw data with skikit and nltk. I have different rows set from the Data tab, each of which has some labels, paragraphs and some other features. I am interested in classifying the paragraphs. I need to convert this data into the format for an inbuilt classifier of Consultant / NLTK. I want to implement Gaussian, Barnoli and Multilateral Navy Bays for all the paragraphs. Question 1: For science, given import iris data I checked the iris data, in this there are exact values ​​from the data set. How do I convert my data into such a format and call the Gaussian function directly? Is there any standard way to do this?
Question 2: For NLTK, what should be the input for the NaiveBayesClassifier.classify function? Is it with boolean values? How can it be made of multi-golaal or gausi?

@ Q2:

nltk .NaiveBayesClassifier.classify is a so-called 'featurecats' 'Is expected to have a feature with the name of a feature pack, and values ​​such as attribute values, e.g. {'word1': True, 'word2': true, 'word3': wrong} . The inexperienced Bias Classifier of NLTX can not be used as a multi-purpose approach. However, you can learn skimmeter and use the nltk.classify.scikitlearn wrapper module to deploy skillet's multi-purpose classifier.