Question

I'm using the next code to train a ClassificationTree with one node (Decision stump), using the CutPoint property i was able to get the value of the predicate split, but how can i get the predicate used ? (in this example feature1 or feature2)

x1 = 10 * rand(10, 1);
x2 = 10 * rand(10, 1);

x = [x1 x2];
y = randi(2, 10, 1);

t = ClassificationTree.fit(x, y, 'MinParent', 10, 'MergeLeaves', 'off');

view(t)

t.CutPoint(1,:)

Output

Decision tree for classification
1  if x1<3.60576 then node 2 elseif x1>=3.60576 then node 3 else 1
2  class = 1
3  class = 2


ans =

    3.6058

so i was able to get the value 3.6058 but i need a way to get that x1 was the feature used.

Was it helpful?

Solution

I think you want:

t.CutVar{1}

ans =
x1
Licensed under: CC-BY-SA with attribution
Not affiliated with StackOverflow
scroll top