Practical Predictive Analytics

Output from the str() function

• The str() function tells us that there are 768 observations and 9 variables. All of the variables have been loaded as numeric, except for the target variable diabetes, which has been coded as a factor with two levels.
• neg indicates the diabetic condition that is not present.
• pos indicates the diabetic condition that is present.

Again, the str() is a great way to quickly get insights. Always scan for 0's and NA's in the first few rows and start asking yourself if it is reasonable to expect those values.

Refer to the following figure: