NB.py

train, test = train_test_split(data, test_size = 0.3)

cols = train.columns[:-1]

gnb = MultinomialNB()
gnb.fit(train[cols], train['sentiment'])
y_pred = gnb.predict(test[cols])

print("Number of mislabeled points out of a total {} points : {}, performance {:05.2f}%"
      .format(
          test.shape[0],
          (test["sentiment"] != y_pred).sum(),
          100*(1-(test["sentiment"] != y_pred).sum()/test.shape[0])
))