Python实现决策树 Desision Tree & 可视化

最新推荐文章于 2024-08-01 08:51:58 发布

hxxjxw

最新推荐文章于 2024-08-01 08:51:58 发布

阅读量982

点赞数 1

文章标签：决策树 Python

本文链接：https://blog.csdn.net/hxxjxw/article/details/108877589

版权

diabetes_data_upload.csv

使用了sklearn库中决策树tree.DecisionTreeClassifier()函数

默认是使用gini index作为impurity measures

import pandas as pd
from sklearn import tree

df = pd.read_csv("diabetes_data_upload.csv")
X = pd.get_dummies(df.drop(columns="class")) #Convert categorical attributes to binary attributes using get_dummies()
y = df["class"]

dtc = tree.DecisionTreeClassifier().fit(X, y)

print(tree.export_text(dtc, feature_names=X.columns.tolist()))

可视化决策树

import pandas as pd
from sklearn import tree
import graphviz

df = pd.read_csv("diabetes_data_upload.csv")
X = pd.get_dummies(df.drop(columns="class"))
y = df["class"]

dtc = tree.DecisionTreeClassifier().fit(X, y)

dot_data = tree.export_graphviz(dtc, out_file=None)
graph = graphviz.Source(dot_data)
graph.render("diabetes") #Generate diabetes.pdf