predict.RemoraModel {frbf} | R Documentation |
The predict.RemoraModel funcion overloads the predict function to support the trained
model contained in the RemoraModel
class.
This function recieves the model, a data matrix to classify, and classifies, i.e. predicts, to which class each of the given data points belong to.
predict.RemoraModel(object, data_matrix, ...)
object |
the model , obtained from the learning procedure (see frbf ) |
data_matrix |
the data to classify |
... |
additional arguments affecting the predictions produced |
The data_matrix
can be a matrix
or data.frame
.
It can have the class column if it has the same name, or index, as the training matrix used in frbf
).
In such case it will be automatically ignored, otherwise the class column cannot be present in the data set.
The result is a prediction list containing the class name of each data point. The position of the data point in result is the same as the position in the matrix given for classification.
Fernando Martins and Andre Falcao
Andre O. Falcao, Thibault Langlois and Andreas Wichert (2006) Flexible kernels for RBF networks. Jornal of Neurocomputing, volume 69, pp 2356-2359. Elsevier.
# infert data is composed by 248 points and will be splitted data(infert) # the training matrix will be use the first 100 points training_matrix <- infert[c(1:100) ,] # the matrix to classify will use all the other points classification_matrix <- infert[c(101:248) ,] # create the model model <- frbf(training_matrix, class_name = "education", number_clusters = 10, scale_variance = FALSE) # predict classification <- predict(model, classification_matrix) # the classification points for the last print(classification)