生成决策树的时候为什么有个黑块？怎么解决？只靠.replace(‘\n‘, ‘‘)就可以了吗？

老规矩，先上代码：

代码1，这是有黑块的，dot_data存到本地：

import pydotplus
import os
from sklearn.datasets import load_iris
from sklearn.tree import DecisionTreeClassifier
from sklearn.tree import export_graphviz
from PIL import Image


Iris = load_iris()
X = Iris.data[:, 2:]  # petal length and width
y = Iris.target

tree_clf = DecisionTreeClassifier(max_depth=2, criterion='entropy')
tree_clf.fit(X, y)

os.makedirs("output/决策树/", exist_ok=True)  # 创建目录，如果不存在则创建
export_graphviz(
    tree_clf,
    out_file="output/决策树/Iris_tree.dot",
    feature_names=Iris.feature_names[2:],
    class_names=Iris.target_names,
    rounded=True,
    filled=True
)
graph = pydotplus.graph_from_dot_file("output/决策树/Iris_tree.dot")
dot_data = graph.to_string()
# dot_data = dot_data.replace('\n', '<br/>')
print(dot_data)
graph = pydotplus.graph_from_dot_data(dot_data)
graph.write_png("output/决策树/Iris_tree.png")
# image = Image.open("output/决策树/Iris_tree.png")
# image.show()

代码2，这是没有黑块的，用的IO流：

from sklearn import datasets
from sklearn import tree
from sklearn.metrics import classification_report
from sklearn.model_selection import train_test_split
from sklearn.tree import export_graphviz
from sklearn.tree import DecisionTreeClassifier
from sklearn.datasets import load_iris
from six import StringIO
import pydotplus

Iris = load_iris()
X = Iris.data[:, 2:]  # petal length and width
y = Iris.target

tree_clf = DecisionTreeClassifier(max_depth=2, criterion='entropy')
tree_clf.fit(X, y)

# 生成决策树可视化的DOT文件
dot_data = StringIO()
export_graphviz(
    tree_clf,
    out_file=dot_data,
    feature_names=Iris.feature_names[2:],
    class_names=Iris.target_names,
    rounded=True,
    filled=True
)

dot_data = dot_data.getvalue()
print(type(dot_data))
print(dot_data)
dot_data = dot_data.replace('\n', '')  # 去除右上角黑块
print(type(dot_data))
print(dot_data)

graph = pydotplus.graph_from_dot_data(dot_data)
graph.write_png("output/决策树/iris_decision_tree.png")

生成的图像，对比如下，可以看到右图（代码1）有个黑块：

因为之前解决过黑块问题（就是代码2），所以我在代码1里也尝试

.replace('\n', '')

发现不行。为什么？

先告诉解决方法：

dot_data = dot_data.replace('"\\r\\n";', '')
dot_data = dot_data.replace('\n', '')

再告诉为什么：

请对比二者的dot_data（上图是代码1的第一次print(dot_data)，下图是代码2的dot_data）：

可以发现，存在本地的

dot_data

是包含了

"\r\n";

这个奇奇怪怪的东西的。

因此，需要先去除"\r\n";，也就是代码：

dot_data = dot_data.replace('"\\r\\n";', '')

然后跟IO流的一样，去除\n：

dot_data = dot_data.replace('\n', '')

也就是这样的一个转换：

dot_data1
 digraph Tree {
node [color="black", fontname="helvetica", shape=box, style="filled, rounded"];
edge [fontname="helvetica"];
0 [fillcolor="#ffffff", label="petal width (cm) <= 0.8\nentropy = 1.585\nsamples = 150\nvalue = [50, 50, 50]\nclass = setosa"];
1 [fillcolor="#e58139", label="entropy = 0.0\nsamples = 50\nvalue = [50, 0, 0]\nclass = setosa"];
0 -> 1  [headlabel="True", labelangle=45, labeldistance="2.5"];
2 [fillcolor="#ffffff", label="petal width (cm) <= 1.75\nentropy = 1.0\nsamples = 100\nvalue = [0, 50, 50]\nclass = versicolor"];
0 -> 2  [headlabel="False", labelangle="-45", labeldistance="2.5"];
3 [fillcolor="#4de88e", label="entropy = 0.445\nsamples = 54\nvalue = [0, 49, 5]\nclass = versicolor"];
2 -> 3;
4 [fillcolor="#843de6", label="entropy = 0.151\nsamples = 46\nvalue = [0, 1, 45]\nclass = virginica"];
2 -> 4;
"\r\n";
}

dot_data2
 digraph Tree {
node [color="black", fontname="helvetica", shape=box, style="filled, rounded"];
edge [fontname="helvetica"];
0 [fillcolor="#ffffff", label="petal width (cm) <= 0.8\nentropy = 1.585\nsamples = 150\nvalue = [50, 50, 50]\nclass = setosa"];
1 [fillcolor="#e58139", label="entropy = 0.0\nsamples = 50\nvalue = [50, 0, 0]\nclass = setosa"];
0 -> 1  [headlabel="True", labelangle=45, labeldistance="2.5"];
2 [fillcolor="#ffffff", label="petal width (cm) <= 1.75\nentropy = 1.0\nsamples = 100\nvalue = [0, 50, 50]\nclass = versicolor"];
0 -> 2  [headlabel="False", labelangle="-45", labeldistance="2.5"];
3 [fillcolor="#4de88e", label="entropy = 0.445\nsamples = 54\nvalue = [0, 49, 5]\nclass = versicolor"];
2 -> 3;
4 [fillcolor="#843de6", label="entropy = 0.151\nsamples = 46\nvalue = [0, 1, 45]\nclass = virginica"];
2 -> 4;

}

dot_data3
 digraph Tree {node [color="black", fontname="helvetica", shape=box, style="filled, rounded"];edge [fontname="helvetica"];0 [fillcolor="#ffffff", label="petal width (cm) <= 0.8\nentropy = 1.585\nsamples = 150\nvalue = [50, 50, 50]\nclass = setosa"];1 [fillcolor="#e58139", label="entropy = 0.0\nsamples = 50\nvalue = [50, 0, 0]\nclass = setosa"];0 -> 1  [headlabel="True", labelangle=45, labeldistance="2.5"];2 [fillcolor="#ffffff", label="petal width (cm) <= 1.75\nentropy = 1.0\nsamples = 100\nvalue = [0, 50, 50]\nclass = versicolor"];0 -> 2  [headlabel="False", labelangle="-45", labeldistance="2.5"];3 [fillcolor="#4de88e", label="entropy = 0.445\nsamples = 54\nvalue = [0, 49, 5]\nclass = versicolor"];2 -> 3;4 [fillcolor="#843de6", label="entropy = 0.151\nsamples = 46\nvalue = [0, 1, 45]\nclass = virginica"];2 -> 4;}

需要无黑块的输出图片，就需要类似于dot_data3的输入。

IO流的不会有\r\n，因此只需要：

保存到本地的有\r\n，因此多一步：

大家可以自行print，看看与这个的不同，就知道如何避免黑块了。

因为这个的数据类型就是str，对其的操作其实是十分简单的，希望大家都能将黑块干掉！