When discussing the evaluation of model performance, we introduced the method of dividing data into training...

Transcribed Image Text:

When discussing the evaluation of model performance, we introduced the method of dividing data into training vs. test set. In the illustration of the Caret package, we further demonstrate that the summary statistics of training and test are similar. The question. for the discussion is, in machine learning, Why should we divide the data into two sets of similar characteristics, and how do we do it? When discussing the evaluation of model performance, we introduced the method of dividing data into training vs. test set. In the illustration of the Caret package, we further demonstrate that the summary statistics of training and test are similar. The question. for the discussion is, in machine learning, Why should we divide the data into two sets of similar characteristics, and how do we do it?