Another stop criteria - only for regression trees. As soon as the estimated node value differs from the node training samples responses by less than the parameter value, the node is not split further.

public float RegressionAccuracy { get; set; }

Public Property RegressionAccuracy As Single
	Get
	Set

public:
property float RegressionAccuracy {
	float get ();
	void set (float value);
}

member RegressionAccuracy : float32 with get, set