1. Aprende
    2. /
  2. Cursos
    3. /
  3. Introduction to Predictive Analytics in Python
  • 1

    Building Logistic Regression Models

    Gratuito

    In this Chapter, you'll learn the basics of logistic regression: how can you predict a binary target with continuous variables and, how should you interpret this model and use it to make predictions for new examples?

  • 2

    Forward stepwise variable selection for logistic regression

    In this chapter you'll learn why variable selection is crucial for building a useful model. You'll also learn how to implement forward stepwise variable selection for logistic regression and how to decide on the number of variables to include in your final model.

  • 3

    Explaining model performance to business

    Now that you know how to build a good model, you should convince stakeholders to use it by creating appropriate graphs. You will learn how to construct and interpret the cumulative gains curve and lift graph.

  • 4

    Interpreting and explaining models

    In a business context, it is often important to explain the intuition behind the model you built. Indeed, if the model and its variables do not make sense, the model might not be used. In this chapter you'll learn how to explain the relationship between the variables in the model and the target by means of predictor insight graphs.

Initializing

Ejercicio

Evaluating a model on test and train

The function auc_train_test calculates the AUC of model that is built on a train set and evaluated on a test set: 

auc_train, auc_test = auc_train_test(variables, target, train, test)

with variables a list of the names of the variables that is used in the model.

In this exercise, you will apply this function, and check whether the train and test AUC are similar.

Instrucciones

100 XP
  • The basetable is loaded. Partition the basetable such that the train set contains 70% of the data, and make sure that train and test set have equal target incidence.
  • Calculate the train and test AUC of the model using "age" and "gender_F" as predictors using the auc_train_test function.