main.py

def main():

    X, Y = read_corpus(args.input, args.binary)

    Xtrain, Xtest, Ytrain, Ytest = shuffle_split(X, Y, 0.8)

    prior_prob = prior_probabilities(Y)
    info('Prior probabilities per class: {0}'.format(prior_prob))

    classifier = feature_union(count=True, tfidf=True, textstats=True)
    classifier.fit(Xtrain, Ytrain)  # fit the classifier on the training set
    Yguess = classifier.predict(Xtest)  # predict the labels on the test set
    posterior_prob = classifier.predict_proba(Xtest)  # calculate posterior probabilities

    df = tabular_results(Xtest, Ytest, Yguess, prior_prob, posterior_prob)
    with pd.option_context('display.max_rows', 10, 'display.max_columns', None):
        debug(df)

    baseline = baseline_classifier(Xtest, Ytest)
    class_report("Baseline classifier", Ytest, baseline, show_matrix=False)

    class_report("Naive Bayes classifier", Ytest, Yguess, show_matrix=True)