The function converts a given CoreModel model (decision or regression tree)
rpart.object prepared for visualization with
A tree model produced by
A data set which was used in learning of the
The conversion creates
rpart.object and copies CORElearn internal structures contained in memory controlled by
dynamic link library written in C++.
An alternative visualization is accessible via function
display, which outputs tree structure formatted
for screen or in dot format.
Function returns a
Initial version by John Adeyanju Alao, improvements by Marko Robnik-Sikonja.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
# plot a decision tree directly dataset <- CO2 md<-CoreModel(Plant ~ ., dataset, model="tree") plot(md, dataset) # or indirectly rpm <- getRpartModel(md, dataset) # set angle to tan(0.5)=45 (degrees) and length of branches at least 5 plot(rpm, branch=0.5, minbranch=5, compress=TRUE) # pretty=0 prints full names of attributes, # numbers to 3 decimals, try to make a dendrogram more compact text(rpm, pretty=0, digits=3) destroyModels(md) # clean up # an alternative is to use fancier rpart.plot package # rpart.plot(rpm) # rpart.plot has many parameters controlling the output # but it cannot plot models in tree leaves