Supervised ML - Naive Bayes

Steps for Naive Bayes using sklearn(Python):

1. Import GaussianNB:
• from sklearn.naivebayes import GaussianNB
2.  Create classifier:
• clf = GaussianNB()
3. Fit the classifier:
• clf.fit(FeatureTrainingSet,LabelTrainingSet)
4. Prediction for testing data
• clf.predict(FeatureTestingSet)

Accuracy is one of the metrics to check fitting of model over available dataset.There are 3 ways to check accuracy

1. Manually calculating number of correctly predicted data set upon all dataset.
2. Using accuracy_score metrics:
• import statement should be: from sklearn_metrics import accuracy_score
• wherever you want to check accuracy : print accuracy_score(NumpyArrayToBePredicted, labelledArray)
• Advantage is that you can use other available metrics as well.
3.   Using Classifier Score:
• clf.score(FeatureDataTobePredicted,labelledArray)
• It will calculate predictions for provided data and then calculates accuracy for that data.
• Advantage is that you don't need to calculate predictions separately if you are interested in accuracy only.