Btukfyl

I have been working on this all day (struggled rather). Having read through the documentation, many other tutorials and due to my inexperience, I can't figure out how to use my own data with a MultinomialNB classifier?

Here is the code from the main tutorial:

from sklearn.datasets import fetch_20newsgroups

from sklearn.naive_bayes import MultinomialNB



categories = ['alt.atheism', 'soc.religion.christian',

              'comp.graphics', 'sci.med']



text_clf = Pipeline([('vect', CountVectorizer()),

                     ('tfidf', TfidfTransformer()),

                     ('clf', MultinomialNB()),

])

twenty_train = fetch_20newsgroups(subset='train',

    categories=categories, shuffle=True, random_state=42)

text_clf.fit(twenty_train.data, twenty_train.target)  



docs_test = ['Graphics is love', 'the brain is part of the body']



predicted = text_clf.predict(docs_test)



for doc, category in zip(docs_test, predicted):

    print('%r => %s' % (doc, twenty_train.target_names[category]))

Obviously, it works. But how can I replace fetch_20newsgroups with my own data (Stored in a python dictionary or the like)? And each item in the training data below is classified as one of the categories, how is this achieved?

I appreciate this is not a great question, but in this time of need, I just want to gain an understanding of how this works. Thanks

Almost all sklearn fit methods take a list of training data, and list of labels as input. In your case, the list of training data would be a list of strings (texts on which you have to train your model). Like ['this is my first training sample', 'this is second string', 'and this is third', ...], and another list of labels like ['label1', 'label2', 'label1', ...].

And you'll pass these lists to the fit method:

text_clf.fit(list_of_training_datas, list_of_labels)

predict method would remain the same, as it would also take a list of samples you want to test, and will return a list containing the predicted label for each of the test samples.