Session Ready
Exercise

Defining feature columns

Defining feature columns is an important first step when creating a model with Estimators. It is here that you'll define which columns are predictors in your dataset.

In the exercises for this chapter, you're going to add on to the dataset where you attempted to predict high school grades, Finalpercent, based on the number of minutes students spent studying each day, minstudytime. You'll add three more variables to see if you can increase the predictive power of the model:

  1. If a student is from a rural (1) or urban area (0): Rural
  2. The number of absences a student had in the school year: absences
  3. The number of past classes a student has failed: failures
Instructions
100 XP
  • Define a proper feature columns list using all four of the predictor columns.
  • Make sure you specify the correct types of columns each are, including numeric_column and categorical_column_with_identity.
Loading