def learn(self, model, inf=None): model.regularization = 0.01 model.inter_node_pos_same_label_default = 0.1 p = mmln.estimate_p_values_inter_node(self.n) for label1 in p: for label2 in p[label1]: if p[label1][label2] < 0.1: model.inter_node_pos[(label1, label2)] = 2 elif p[label1][label2] < 0.2: model.inter_node_pos[(label1, label2)] = 1 elif p[label1][label2] < 0.5: model.inter_node_pos[(label1, label2)] = 0.5
def test_estimate_p_values_inter_node(self): net = self._get_network() p = mmln.estimate_p_values_inter_node(net, 100) self.assertEqual(len(p), 2)