Naive Bayes: How can we tell when to tune data? Today in class we were told to: Apply naiveBayes to your train and test set of exercise 2 and compute the auc. Do you obtain a good or bad
auc? But how do we know if we need to tune the data?
Answers and follow-up questions Answer or follow-up question 1
Can you elaborate on what you mean by `tune the data'?
If you mean to ask: `How do we know if we need to tune a model?', then the answer is: if the algorithm has parameters that require tuning.
These are all the parameters that we tune in class.
Naive Bayes does not require tuning.
Michel Ballings Answer or follow-up question 2
When do we know which parameters of the model need tuning?
Thanks! Answer or follow-up question 3
You need to tune all the parameters that have a large impact on the estimation process. These are all the parameters that we tune in the
Michel BallingsSign in to be able to add an answer or mark this question as resolved.