我正在尝试将weka多层感知器输出转换为matlab代码，以便我可以使用此代码对新输入进行分类。 weka输出如下所示。我知道关于MLP算法的理论。但我不知道如何将weka输出转换为matlab代码。什么是sigmoid节点0的等式。 out put如下所示。我试着这样做。
node0= node4(weight)+node5(weight)+node6(weight);
node0 = 1 / (1 + exp(-node0));
if(node0 < threshold)
node0 = 0;
end
我为所有节点做了但输出不正确。
Sigmoid Node 0
Inputs Weights
Threshold 0.9406969488131344
Node 4 -2.1903200701922616
Node 5 -15.897044602974933
Node 6 0.3041300920465129
Sigmoid Node 1
Inputs Weights
Threshold -0.47458154982115386
Node 4 -2.404682382816731
Node 5 -1.0175108003621611
Node 6 -1.248161956451173
Sigmoid Node 2
Inputs Weights
Threshold -2.494927166085809
Node 4 -2.5916900088001404
Node 5 2.6569952694128207
Node 6 3.6695765419956974
Sigmoid Node 3
Inputs Weights
Threshold -2.8679148239112555
Node 4 3.548745253783415
Node 5 1.8555144089470257
Node 6 -3.5768482414822405
Sigmoid Node 4
Inputs Weights
Threshold 2.5705895442897604
Attrib length 19.329302599023382
Attrib width 0.46083507698821957
Sigmoid Node 5
Inputs Weights
Threshold 23.728757673706202
Attrib length 26.251966218668777
Attrib width 3.4888088475120576
Sigmoid Node 6
Inputs Weights
Threshold -3.6494152797182733
Attrib length 10.163410415907084
Attrib width -3.6153885660597807
Class pede
Input
Node 0
Class bike
Input
Node 1
Class auto
Input
Node 2
Class lkw
Input
Node 3