naive_bayes

A ready-made Naive Bayes classification model for binary vectors. More info here

###Usage: A simple example usage is given below:

>>> from naive_bayes import naive_bayes
>>> data = [('customer', [1,1,1,0]), ('not_customer', [1,0,1,0]), ('potential_customer', [1,1,1,1])....('some_class', [some_vector])]
>>> features = ['employed', 'married', 'has_children', 'has_car']
>>> p = naive_bayes(data, features)
>>> p.process()
>>> p.predict(vector, key_only=True/False)
>>> ...

Here is the list of all the parameters for the parent naive_bayes method:

• data - a 2D dataset containing samples in this format: ('class', vector)
• features - a 1D array containing all the feature labels as 'strings' And the predictor method:
• vector - input vector to classify (provide either as a list or as a tuple)
• key_only - setting this to True returns only the predicted label, and to False returns probabilities for all the classes.