This function creates a data frame that contains a grid of complexity
parameters specific methods.
Usage
createGrid(method, len = 3, data = NULL, pp = NULL)
Arguments
method
a string specifying which classification model to use. See train for a full list.
len
an integer specifying the number of points on the grid for each tuning parameter.
data
the training data (only needed in the case where the method is cforest, earth, bagEarth, fda, bagFDA, rpart, svmRadial, pam, lars2
pp
an optional vector of pre-processing options.
Value
A data frame where the rows are combinations of tuning parameters and columns correspond to the parameters. The column names should be the parameter names preceded by a dot (e.g. .mtry)
Details
A grid is created with rows corresponding to complexity parameter combinations. If the model does not use tuning parameters (like a linear model), values of NA are returned. Columns are named the same as the parameter name, but preceded by a period.
For some models (see list above), the data should be passed to the function via the data argument. In these cases, the outcome should be included in a column named .outcome.
createGrid("rda", 4)
createGrid("lm")
createGrid("nnet")
## data needed for SVM with RBF:tmp <- iris
names(tmp)[5] <- ".outcome"head(tmp)
createGrid("svmRadial", data = tmp, len = 4)