LoslegenKostenlos loslegen

A first CTR model

In this exercise, you will build a first CTR model on the Avazu dataset using a decision tree and evaluate the accuracy of the model using accuracy_score() from sklearn. Additionally, you will use train_test_split() from sklearn to split training and testing data instead of manually defining a split point as before.

In your workspace, sample data in DataFrame form is loaded as df along with sklearn and pandas as pd.

We will do a basic training and testing split and evaluate our results using accuracy.

Diese Übung ist Teil des Kurses

Predicting CTR with Machine Learning in Python

Kurs anzeigen

Anleitung zur Übung

  • Define both X and y to be the features and target respectively based on the click column.
  • Split the data into training and testing sets using train_test_split(X, y).
  • Create a decision tree classifier.
  • Create predictions using the classifier and evaluate the accuracy of its predictions.

Interaktive Übung

Versuche dich an dieser Übung, indem du diesen Beispielcode vervollständigst.

# Define X and y 
X = df.____[:, ~df.columns.____(['click'])]
y = df.click

# Define training and testing
X_train, X_test, y_train, y_test = \
	____(____, _____, test_size = .2, random_state = 0)

# Create decision tree classifier
clf = ____()

# Train classifier - predict label and evaluate accuracy
y_pred = clf.fit(____, _____).____(X_test) 
print(____(y_test, y_pred))
Code bearbeiten und ausführen