I am trying to understand cost complexity pruning in classification trees.
I found that DecisionTree in sklearn has a function called cost_complexity_pruning_path, which gives the effective alphas of subtrees during pruning.
What does effective alpas means? I though alpha, that ranges between 0 And 1, is the parameter in an optimization problem. So I thought that all possible alpha values should be tested to find the min value of the objective function. Why in python specific effective alpha values are produced and not all possible values are not tested?
Thanks,
Andreas
