MATLAB: How to determine which sample was assigned to which node in CLASSREGTREE in Statistics Toolbox 7.4 (R2010b)

Statistics and Machine Learning Toolbox

I would like to determine which sample was assigned to which node in a tree created using CLASSREGTREE.

Best Answer

You can use the EVAL method of the CLASSREGTREE class to determine which sample was assigned to which final node. For example:

load fisheriris;
t = classregtree(meas,species,'names',{'SL' 'SW' 'PL' 'PW'});
[fittedval, nodes] = eval(t,meas);    % determine mapping of training samples to nodes

This code stores the nodes for each sample in "meas" in the second output argument "nodes".

Related Solutions

MATLAB: Close classregtree window programmatically

I believe it creates a figure whose handle visibility is off. I do not know the name of the figure. Try

hiddenfigs = findall(0,'Type','figure', '-not', 'HandleVisibility', 'on');

MATLAB: How to retrieve the levels of a node, maximum level and maximum pruning level from a CLASSREGTREE in MATLAB 7.6 (R2008a)

There are at least two ways to obtain the level of a node in a CLASSREGTREE using MATLAB 7.6 (R2008a).

The first is to view the tree and read the number of levels from the GUI:

load fisheriris;
t = classregtree(meas,species,'names',{'SL' 'SW' 'PL' 'PW'});
view(t)

The second method allows you to programmatically obtain the number of levels. There is a file attached: 'levelOf.m' which uses recursion to obtain the level of any node in the tree, where the root node is level 1, the children of the root are level 2 and so forth. To obtain the number of levels, call LEVELOF with the following syntax:

nodeindex = t.numnodes;
nodelevel = levelOf(t,nodeindex)

To retrieve the maximum pruning level, use:

max(t.prunelist)

Best Answer

Related Solutions

MATLAB: Close classregtree window programmatically

MATLAB: How to retrieve the levels of a node, maximum level and maximum pruning level from a CLASSREGTREE in MATLAB 7.6 (R2008a)

Related Question