ML之DT：構建一個二元DT(sk的DTR)來進行評分預測+Graphviz視覺化

輸出結果

dot檔案內容

digraph Tree {
node [shape=box] ;
1 [label="X[10] <= 10.525\nmse = 0.652\nsamples = 1599\nvalue = 5.636"] ;
2 [label="X[9] <= 0.575\nmse = 0.431\nsamples = 983\nvalue = 5.366"] ;
1 -> 1 [labeldistance=2.5, labelangle=45, headlabel="True"] ;
2 [label="X[1] <= 0.748\nmse = 0.328\nsamples = 391\nvalue = 5.151"] ;
1 -> 2 ;
3 [label="mse = 0.289\nsamples = 328\nvalue = 5.201"] ;
2 -> 3 ;
4 [label="mse = 0.448\nsamples = 63\nvalue = 4.889"] ;
2 -> 4 ;
5 [label="X[1] <= 0.405\nmse = 0.449\nsamples = 592\nvalue = 5.508"] ;
1 -> 5 ;
6 [label="mse = 0.486\nsamples = 144\nvalue = 5.833"] ;
5 -> 6 ;
7 [label="mse = 0.393\nsamples = 448\nvalue = 5.404"] ;
5 -> 7 ;
8 [label="X[9] <= 0.645\nmse = 0.702\nsamples = 616\nvalue = 6.067"] ;
0 -> 8 [labeldistance=2.5, labelangle=-45, headlabel="False"] ;
9 [label="X[1] <= 1.015\nmse = 0.705\nsamples = 272\nvalue = 5.728"] ;
8 -> 9 ;
10 [label="mse = 0.591\nsamples = 262\nvalue = 5.794"] ;
9 -> 10 ;
11 [label="mse = 0.6\nsamples = 10\nvalue = 4.0"] ;
9 -> 11 ;
12 [label="X[10] <= 11.55\nmse = 0.536\nsamples = 344\nvalue = 6.334"] ;
8 -> 12 ;
13 [label="mse = 0.495\nsamples = 206\nvalue = 6.121"] ;
12 -> 13 ;
14 [label="mse = 0.43\nsamples = 138\nvalue = 6.652"] ;
12 -> 14 ;
}
 

Graphviz視覺化

評分預測的決策樹

核心程式碼

with open("wineTree.dot", 'w') as f:
    f = tree.export_graphviz(wineTree, out_file=f)