网站建设都分几个阶段,网站建设与网页设计,怎么用电脑做网站主机,做网站开发要学什么软件文章目录 第5章 决策树—python 实践书上题目5.1利用ID3算法生成决策树#xff0c;例5.3scikit-learn实例 《统计学习方法#xff1a;李航》笔记 从原理到实现#xff08;基于python#xff09;-- 第5章 决策树 第5章 决策树—python 实践
import numpy as np
import pand… 文章目录 第5章 决策树—python 实践书上题目5.1利用ID3算法生成决策树例5.3scikit-learn实例 《统计学习方法李航》笔记 从原理到实现基于python-- 第5章 决策树 第5章 决策树—python 实践
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
%matplotlib inlinefrom sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from collections import Counter
import math
from math import log
import pprint书上题目5.1 def create_data():datasets [[青年, 否, 否, 一般, 否],[青年, 否, 否, 好, 否],[青年, 是, 否, 好, 是],[青年, 是, 是, 一般, 是],[青年, 否, 否, 一般, 否],[中年, 否, 否, 一般, 否],[中年, 否, 否, 好, 否],[中年, 是, 是, 好, 是],[中年, 否, 是, 非常好, 是],[中年, 否, 是, 非常好, 是],[老年, 否, 是, 非常好, 是],[老年, 否, 是, 好, 是],[老年, 是, 否, 好, 是],[老年, 是, 否, 非常好, 是],[老年, 否, 否, 一般, 否],]labels [u年龄, u有工作, u有自己的房子, u信贷情况, u类别]# 返回数据集和每个维度的名称return datasets, labelsdatasets, labels create_data()
train_data pd.DataFrame(datasets, columnslabels)# 熵
def calc_ent(datasets):data_length len(datasets)label_count {}for i in range(data_length):label datasets[i][-1]if label not in label_count:label_count[label] 0label_count[label] 1ent -sum([(p / data_length) * log(p / data_length, 2)for p in label_count.values()])return ent# 经验条件熵
def cond_ent(datasets, axis0):data_length len(datasets)feature_sets {}for i in range(data_length):feature datasets[i][axis]if feature not in feature_sets:feature_sets[feature] []feature_sets[feature].append(datasets[i])cond_ent sum([(len(p) / data_length) * calc_ent(p) for p in feature_sets.values()])return cond_ent# 信息增益熵-经验条件熵
def info_gain(ent, cond_ent):return ent - cond_entdef info_gain_train(datasets):count len(datasets[0]) - 1ent calc_ent(datasets)best_feature []for c in range(count):c_info_gain info_gain(ent, cond_ent(datasets, axisc))best_feature.append((c, c_info_gain))print(特征({}) - info_gain - {:.3f}.format(labels[c], c_info_gain))# 比较大小best_ max(best_feature, keylambda x: x[-1])return 特征({})的信息增益最大选择为根节点特征.format(labels[best_[0]])info_gain_train(np.array(datasets))特征(年龄) - info_gain - 0.083
特征(有工作) - info_gain - 0.324
特征(有自己的房子) - info_gain - 0.420
特征(信贷情况) - info_gain - 0.363
特征(有自己的房子)的信息增益最大选择为根节点特征利用ID3算法生成决策树例5.3
# 定义节点类 二叉树
class Node:def __init__(self, rootTrue, labelNone, feature_nameNone, featureNone):self.root rootself.label labelself.feature_name feature_nameself.feature featureself.tree {}self.result {label:: self.label,feature: self.feature,tree: self.tree}def __repr__(self):return {}.format(self.result)def add_node(self, val, node):self.tree[val] nodedef predict(self, features):if self.root is True:return self.labelreturn self.tree[features[self.feature]].predict(features)class DTree:def __init__(self, epsilon0.1):self.epsilon epsilonself._tree {}# 熵staticmethoddef calc_ent(datasets):data_length len(datasets)label_count {}for i in range(data_length):label datasets[i][-1]if label not in label_count:label_count[label] 0label_count[label] 1ent -sum([(p / data_length) * log(p / data_length, 2)for p in label_count.values()])return ent# 经验条件熵def cond_ent(self, datasets, axis0):data_length len(datasets)feature_sets {}for i in range(data_length):feature datasets[i][axis]if feature not in feature_sets:feature_sets[feature] []feature_sets[feature].append(datasets[i])cond_ent sum([(len(p) / data_length) * self.calc_ent(p)for p in feature_sets.values()])return cond_ent# 信息增益staticmethoddef info_gain(ent, cond_ent):return ent - cond_entdef info_gain_train(self, datasets):count len(datasets[0]) - 1ent self.calc_ent(datasets)best_feature []for c in range(count):c_info_gain self.info_gain(ent, self.cond_ent(datasets, axisc))best_feature.append((c, c_info_gain))# 比较大小best_ max(best_feature, keylambda x: x[-1])return best_def train(self, train_data):input:数据集D(DataFrame格式)特征集A阈值etaoutput:决策树T_, y_train, features train_data.iloc[:, :-1], train_data.iloc[:,-1], train_data.columns[:-1]# 1,若D中实例属于同一类Ck则T为单节点树并将类Ck作为结点的类标记返回Tif len(y_train.value_counts()) 1:return Node(rootTrue, labely_train.iloc[0])# 2, 若A为空则T为单节点树将D中实例树最大的类Ck作为该节点的类标记返回Tif len(features) 0:return Node(rootTrue,labely_train.value_counts().sort_values(ascendingFalse).index[0])# 3,计算最大信息增益 同5.1,Ag为信息增益最大的特征max_feature, max_info_gain self.info_gain_train(np.array(train_data))max_feature_name features[max_feature]# 4,Ag的信息增益小于阈值eta,则置T为单节点树并将D中是实例数最大的类Ck作为该节点的类标记返回Tif max_info_gain self.epsilon:return Node(rootTrue,labely_train.value_counts().sort_values(ascendingFalse).index[0])# 5,构建Ag子集node_tree Node(rootFalse, feature_namemax_feature_name, featuremax_feature)feature_list train_data[max_feature_name].value_counts().indexfor f in feature_list:sub_train_df train_data.loc[train_data[max_feature_name] f].drop([max_feature_name], axis1)# 6, 递归生成树sub_tree self.train(sub_train_df)node_tree.add_node(f, sub_tree)# pprint.pprint(node_tree.tree)return node_treedef fit(self, train_data):self._tree self.train(train_data)return self._treedef predict(self, X_test):return self._tree.predict(X_test)datasets, labels create_data()
data_df pd.DataFrame(datasets, columnslabels)
dt DTree()
tree dt.fit(data_df)tree{label:: None, feature: 2, tree: {否: {label:: None, feature: 1, tree: {否: {label:: 否, feature: None, tree: {}}, 是: {label:: 是, feature: None, tree: {}}}}, 是: {label:: 是, feature: None, tree: {}}}}dt.predict([老年, 否, 否, 一般])否scikit-learn实例
# data
def create_data():iris load_iris()df pd.DataFrame(iris.data, columnsiris.feature_names)df[label] iris.targetdf.columns [sepal length, sepal width, petal length, petal width, label]data np.array(df.iloc[:100, [0, 1, -1]])# print(data)return data[:, :2], data[:, -1]X, y create_data()
X_train, X_test, y_train, y_test train_test_split(X, y, test_size0.3)from sklearn.tree import DecisionTreeClassifier
from sklearn.tree import export_graphviz
import graphvizclf DecisionTreeClassifier()
clf.fit(X_train, y_train,)DecisionTreeClassifier(class_weightNone, criteriongini, max_depthNone,max_featuresNone, max_leaf_nodesNone,min_impurity_decrease0.0, min_impurity_splitNone,min_samples_leaf1, min_samples_split2,min_weight_fraction_leaf0.0, presortFalse, random_stateNone,splitterbest)clf.score(X_test, y_test)0.9666666666666667tree_pic export_graphviz(clf, out_filemytree.pdf)
with open(mytree.pdf) as f:dot_graph f.read()
graphviz.Source(dot_graph)
graphviz.files.Source at 0x1f159bc2780
