bClassif(train, predClass, predProb, ...)train.
This function should return a vector with the predicted classes.
For usage information see Details.train.
This function should return a matrix with the predicted
probabilities. The matrix, should have:
the number of rows equal to the size of prediction set and
the number of columns equal to the number of possible classes.
For usage information see Details.train.bClassif.
train function is used as follows:model <- train(m, y, ...)
The training data is provided in the two first arguments. Here, m
is a distance matrix between the training instances and y is a vector
with the classes of those instances. Additional arguments of the classifier can
be passed to train through the additional arguments (...)
specified in bClassif.
The predClass and predProb functions are used as follows:
cls <- predClass(model, pm)
cls <- predProb(model, pm)
In both cases, the prediction data is supplied using pm.
pm is a distance matrix with the following dimensions:
the number of rows equal to the size of the prediction set and the number
of columns equal to the size of the training set. predClass returns
a vector with the predicted classes and predProb returns a matrix
that contains the class probabilities. In this matrix, there is a column
for each class and a row for each instance.