Ajuster un SVM linéaire

Dans cet exercice, vous allez étudier l’influence du paramètre de coût sur le nombre de vecteurs de support pour des SVM linéaires. Pour cela, vous allez entraîner deux SVM : l’un avec cost = 1 et l’autre avec cost = 100, puis relever le nombre de vecteurs de support. Un jeu de données d’entraînement est disponible dans le dataframe trainset.

Cet exercice fait partie du cours

<cours>Machines à vecteurs de support en R</cours>

Exercice interactif pratique

Essayez cet exercice en complétant ce code d’exemple.

#build svm model, cost = 1
svm_model_1 <- svm(y ~ .,
                   data = ___,
                   type = "C-classification",
                   cost = ___,
                   kernel = "___",
                   scale = FALSE)

#print model details
___

Modifier et exécuter le code

Cet exercice fait partie du cours

<cours>Machines à vecteurs de support en R</cours>

IntermédiaireNiveau de compétence

4.8+

Commencer le cours gratuitement

This chapter introduces some key concepts of support vector machines through a simple 1-dimensional example. Students are also walked through the creation of a linearly separable dataset that is used in the subsequent chapter.

Exercise 1: Sugar content of soft drinks Exercise 2: Visualizing a sugar content dataset Exercise 3: Identifying decision boundaries Exercise 4: Find the maximal margin separator Exercise 5: Visualize the maximal margin separator Exercise 6: Generating a linearly separable dataset Exercise 7: Generate a 2d uniformly distributed dataset.Exercise 8: Create a decision boundary Exercise 9: Introduce a margin in the dataset

Introduces students to the basic concepts of support vector machines by applying the svm algorithm to a dataset that is linearly separable. Key concepts are illustrated through ggplot visualisations that are built from the outputs of the algorithm and the role of the cost parameter is highlighted via a simple example. The chapter closes with a section on how the algorithm deals with multiclass problems.

Exercise 1: Machines à vecteurs de support linéaires Exercise 2: Créer des jeux d’entraînement et de test Exercise 3: Construire un classifieur SVM linéaire Exercise 4: Explorer le modèle et calculer la précision Exercise 5: Visualiser les SVM linéaires Exercise 6: Visualiser les vecteurs de support avec ggplot Exercise 7: Visualiser les frontières de décision et de marge avec `ggplot2`Exercise 8: Visualiser les frontières de décision et de marge avec `plot()`Exercise 9: Ajuster les SVM linéaires Exercise 10: Ajuster un SVM linéaire

Exercice actuel

Exercise 11: Visualiser les frontières de décision et les marges Exercise 12: Dans quels cas les classificateurs à marge souple sont-ils utiles ?Exercise 13: Problèmes multiclasse Exercise 14: Un problème de classification multiclasse Exercise 15: Iris, version 2 : une précision plus robuste.

Provides an introduction to polynomial kernels via a dataset that is radially separable (i.e. has a circular decision boundary). After demonstrating the inadequacy of linear kernels for this dataset, students will see how a simple transformation renders the problem linearly separable thus motivating an intuitive discussion of the kernel trick. Students will then apply the polynomial kernel to the dataset and tune the resulting classifier.

Exercise 1: Generating a radially separable dataset Exercise 2: Generating a 2d radially separable dataset Exercise 3: Visualizing the dataset Exercise 4: Linear SVMs on radially separable data Exercise 5: Linear SVM for a radially separable dataset Exercise 6: Average accuracy for linear SVM Exercise 7: The kernel trick Exercise 8: Visualizing transformed radially separable data Exercise 9: SVM with polynomial kernel Exercise 10: Tuning SVMs Exercise 11: Using `tune.svm()`Exercise 12: Building and visualizing the tuned model

Builds on the previous three chapters by introducing the highly flexible Radial Basis Function (RBF) kernel. Students will create a "complex" dataset that shows up the limitations of polynomial kernels. Then, following an intuitive motivation for the RBF kernel, students see how it addresses the shortcomings of the other kernels discussed in this course.

Exercise 1: Generating a complex dataset Exercise 2: Generating a complex dataset - part 1 Exercise 3: Generating a complex dataset - part 2 Exercise 4: Visualizing the dataset Exercise 5: Motivating the RBF kernel Exercise 6: Linear SVM for complex dataset Exercise 7: Quadratic SVM for complex dataset Exercise 8: The RBF Kernel Exercise 9: Polynomial SVM on a complex dataset Exercise 10: RBF SVM on a complex dataset Exercise 11: Tuning an RBF kernel SVM