svm is used to train a support vector machine. It can be used to carry out general regression and classification (of nu and epsilon-type), as well as density-estimation. A formula interface is provided.
## S3 method for class 'formula': svm((formula, data = NULL, ..., subset, na.action = na.omit, scale = TRUE)) ## S3 method for class 'default': svm((x, y = NULL, scale = TRUE, type = NULL, kernel = "radial", degree = 3, gamma = if (is.vector(x)) 1 else 1 / ncol(x), coef0 = 0, cost = 1, nu = 0.5, class.weights = NULL, cachesize = 40, tolerance = 0.001, epsilon = 0.1, shrinking = TRUE, cross = 0, probability = FALSE, fitted = TRUE, ..., subset, na.action = na.omit))
- a symbolic description of the model to be fit.
- an optional data frame containing the variables in the model. By default the variables are taken from the environment which ‘svm’ is called from.
- a data matrix, a vector, or a sparse matrix (object of class
Matrixprovided by the Matrix package, or of class
matrix.csrprovided by the SparseM package, or of class
simple_triplet_matrixprovided by the slam package).
- a response vector with one label for each row/component of
x. Can be either a factor (for classification tasks) or a numeric vector (for regression).
- A logical vector indicating the variables to be scaled. If
scaleis of length 1, the value is recycled as many times as needed. Per default, data are scaled internally (both
yvariables) to zero mean and unit variance. The center and scale values are returned and used for later predictions.
svmcan be used as a classification machine, as a regression machine, or for novelty detection. Depending of whether
yis a factor or not, the default setting for
eps-regression, respectively, but may be overwritten by setting an explicit value.
Valid options are:
one-classification(for novelty detection)
- the kernel used in training and predicting. You might consider changing some of the following parameters, depending on the kernel type.
- (gamma*u'*v + coef0)^degree
- radial basis:
- tanh(gamma*u'*v + coef0)
- parameter needed for kernel of type
- parameter needed for all kernels except
linear(default: 1/(data dimension))
- parameter needed for kernels of type
- cost of constraints violation (default: 1)---it is the ‘C’-constant of the regularization term in the Lagrange formulation.
- parameter needed for
- a named vector of weights for the different classes, used for asymmetric class sizes. Not all factor levels have to be supplied (default weight: 1). All components have to be named.
- cache memory in MB (default 40)
- tolerance of termination criterion (default: 0.001)
- epsilon in the insensitive-loss function (default: 0.1)
- option whether to use the shrinking-heuristics (default:
- if a integer value k>0 is specified, a k-fold cross validation on the training data is performed to assess the quality of the model: the accuracy rate for classification and the Mean Squared Error for regression
- logical indicating whether the fitted values should be computed and included in the model or not (default:
- logical indicating whether the model should allow for probability predictions.
- additional parameters for the low level fitting function
- An index vector specifying the cases to be used in the training sample. (NOTE: If given, this argument must be named.)
- A function to specify the action to be taken if
NAs are found. The default action is
na.omit, which leads to rejection of cases with missing values on any required variable. An alternative is
na.fail, which causes an error if
NAcases are found. (NOTE: If given, this argument must be named.)
For multiclass-classification with k levels, k>2,
libsvm uses the ‘one-against-one’-approach, in which k(k-1)/2 binary classifiers are trained; the appropriate class is found by a voting scheme.
libsvm internally uses a sparse data representation, which is also high-level supported by the package SparseM. If the predictor variables include factors, the formula interface must be used to get a correct model matrix.
plot.svm allows a simple graphical visualization of classification models.
The probability model for classification fits a logistic distribution using maximum likelihood to the decision values of all binary classifiers, and computes the a-posteriori class probabilities for the multi-class problem using quadratic optimization. The probabilistic regression model assumes (zero-mean) laplace-distributed errors for the predictions, and estimates the scale parameter using maximum likelihood.
An object of class
"svm" containing the fitted model, including:
- The resulting support vectors (possibly scaled).
- The index of the resulting support vectors in the data matrix. Note that this index refers to the preprocessed data (after the possible effect of
- The corresponding coefficients times the training labels.
- The negative intercept.
- In case of a probabilistic regression model, the scale parameter of the hypothesized (zero-mean) laplace distribution estimated by maximum likelihood.
- probA, probB
- numeric vectors of length k(k-1)/2, k number of classes, containing the parameters of the logistic distributions fitted to the decision values of the binary classifiers (1 / (1 + exp(a x + b))).
- Chang, Chih-Chung and Lin, Chih-Jen:
LIBSVM: a library for Support Vector Machines
- Exact formulations of models, algorithms, etc. can be found in the document:
Chang, Chih-Chung and Lin, Chih-Jen:
LIBSVM: a library for Support Vector Machines
- More implementation details and speed benchmarks can be found on: Rong-En Fan and Pai-Hsune Chen and Chih-Jen Lin:
Working Set Selection Using the Second Order Information for Training SVM
Data are scaled internally, usually yielding better results.
Parameters of SVM-models usually must be tuned to yield sensible results!
matrix.csr (in package SparseM)
data(iris) attach(iris) ## classification mode # default with factor response: model <- svm(Species ~ ., data = iris) # alternatively the traditional interface: x <- subset(iris, select = -Species) y <- Species model <- svm(x, y) print(model) summary(model) # test with train data pred <- predict(model, x) # (same as:) pred <- fitted(model) # Check accuracy: table(pred, y) # compute decision values and probabilities: pred <- predict(model, x, decision.values = TRUE) attr(pred, "decision.values")[1:4,] # visualize (classes by color, SV by crosses): plot(cmdscale(dist(iris[,-5])), col = as.integer(iris[,5]), pch = c("o","+")[1:150 %in% model$index + 1]) ## try regression mode on two dimensions # create data x <- seq(0.1, 5, by = 0.05) y <- log(x) + rnorm(x, sd = 0.2) # estimate model and predict input values m <- svm(x, y) new <- predict(m, x) # visualize plot(x, y) points(x, log(x), col = 2) points(x, new, col = 4) ## density-estimation # create 2-dim. normal with rho=0: X <- data.frame(a = rnorm(1000), b = rnorm(1000)) attach(X) # traditional way: m <- svm(X, gamma = 0.1) # formula interface: m <- svm(~., data = X, gamma = 0.1) # or: m <- svm(~ a + b, gamma = 0.1) # test: newdata <- data.frame(a = c(0, 4), b = c(0, 4)) predict (m, newdata) # visualize: plot(X, col = 1:1000 %in% m$index + 1, xlim = c(-5,5), ylim=c(-5,5)) points(newdata, pch = "+", col = 2, cex = 5) # weights: (example not particularly sensible) i2 <- iris levels(i2$Species) <- "versicolor" summary(i2$Species) wts <- 100 / table(i2$Species) wts m <- svm(Species ~ ., data = i2, class.weights = wts)
Documentation reproduced from package e1071, version 1.6-7. License: GPL-2