机器学习——决策树的可视化配置graphviz +windows+pygraphcizd

机器学习——决策树的可视化配置

                                                                    ——Windows 下 graphviz 以及 pygraphcizd 的使用

资料已打包如下，下载地址为：https://pan.baidu.com/s/1RfYrSQ3i9qfOlOY2cjfI_g，密码为：k4go

主要用于二叉树，决策树的学习，直观显示二叉树。

1、如何在python中安装graphviz？

官网：https://pypi.python.org/pypi/graphviz，下载graphviz 0.8.2版本，然后再用pip进行安装，具体步骤如下：

（1）下载

（2）安装：找到路径D:\python2.7\anaconda2.4\Scripts下的pip，详细步骤如下

（3）在pycharm中测试是否安装成功

2、graphviz-2.38.zip下载与安装

下载地址：http://download.csdn.net/download/u010953266/8591169

安装方式地址：http://blog.csdn.net/lanchunhui/article/details/49472949

由于安装后软件没有桌面图标，如何运行？

可以发送到桌面（快捷方式）

打开后的界面为

3、如何在python中安装pygraphviz？

下载地址：https://www.lfd.uci.edu/~gohlke/pythonlibs/#pygraphviz

具体的安装方式与安装graphviz一样，只是不用配置环境变量。

4、决策树的应用说明

机器学习之决策树练习网址：http://scikit-learn.org/stable/modules/tree.html

代码如下：

from sklearn.datasets import load_iris
from sklearn import tree
iris = load_iris()
clf = tree.DecisionTreeClassifier()
clf = clf.fit(iris.data, iris.target)

import graphviz
dot_data = tree.export_graphviz(clf, out_file=None)
graph = graphviz.Source(dot_data)
graph.render("iris")

# 添加颜色
dot_data = tree.export_graphviz(clf, out_file=None,
                         feature_names=iris.feature_names,
                         class_names=iris.target_names,
                         filled=True, rounded=True,
                         special_characters=True)
graph = graphviz.Source(dot_data)
graph

如何运行并生出最终的决策树图片？断点调试：shift+F9

点击后得到的结构序列为：

digraph Tree {
node [shape=box, style="filled, rounded", color="black", fontname=helvetica] ;
edge [fontname=helvetica] ;
0 [label=gini = 0.6667
samples = 150
value = [50, 50, 50]
class = setosa>, fillcolor="#e5813900"] ;
1 [label=samples = 50
value = [50, 0, 0]
class = setosa>, fillcolor="#e58139ff"] ;
0 -> 1 [labeldistance=2.5, labelangle=45, headlabel="True"] ;
2 [label=gini = 0.5
samples = 100
value = [0, 50, 50]
class = versicolor>, fillcolor="#39e58100"] ;
0 -> 2 [labeldistance=2.5, labelangle=-45, headlabel="False"] ;
3 [label=gini = 0.168
samples = 54
value = [0, 49, 5]
class = versicolor>, fillcolor="#39e581e5"] ;
2 -> 3 ;
4 [label=gini = 0.0408
samples = 48
value = [0, 47, 1]
class = versicolor>, fillcolor="#39e581fa"] ;
3 -> 4 ;
5 [label=samples = 47
value = [0, 47, 0]
class = versicolor>, fillcolor="#39e581ff"] ;
4 -> 5 ;
6 [label=samples = 1
value = [0, 0, 1]
class = virginica>, fillcolor="#8139e5ff"] ;
4 -> 6 ;
7 [label=gini = 0.4444
samples = 6
value = [0, 2, 4]
class = virginica>, fillcolor="#8139e57f"] ;
3 -> 7 ;
8 [label=samples = 3
value = [0, 0, 3]
class = virginica>, fillcolor="#8139e5ff"] ;
7 -> 8 ;
9 [label=gini = 0.4444
samples = 3
value = [0, 2, 1]
class = versicolor>, fillcolor="#39e5817f"] ;
7 -> 9 ;
10 [label=samples = 2
value = [0, 2, 0]
class = versicolor>, fillcolor="#39e581ff"] ;
9 -> 10 ;
11 [label=samples = 1
value = [0, 0, 1]
class = virginica>, fillcolor="#8139e5ff"] ;
9 -> 11 ;
12 [label=gini = 0.0425
samples = 46
value = [0, 1, 45]
class = virginica>, fillcolor="#8139e5f9"] ;
2 -> 12 ;
13 [label=gini = 0.4444
samples = 3
value = [0, 1, 2]
class = virginica>, fillcolor="#8139e57f"] ;
12 -> 13 ;
14 [label=samples = 1
value = [0, 1, 0]
class = versicolor>, fillcolor="#39e581ff"] ;
13 -> 14 ;
15 [label=samples = 2
value = [0, 0, 2]
class = virginica>, fillcolor="#8139e5ff"] ;
13 -> 15 ;
16 [label=samples = 43
value = [0, 0, 43]
class = virginica>, fillcolor="#8139e5ff"] ;
12 -> 16 ;
}

最后将方框的内容复制到graphviz中，得到下图：

参考网址：

1、pygraphviz的使用与安装

https://www.cnblogs.com/youngzii/archive/2013/05/23/Python.html

2、windows下graphviz与pygraphviz的安装及入门教程

http://blog.csdn.net/chirebingxue/article/details/50393755

2、安装pygraphviz

http://blog.csdn.net/fencingworld/article/details/53047901