欢迎您访问程序员文章站本站旨在为大家提供分享程序员计算机编程知识!
您现在的位置是: 首页  >  IT编程

基于id3算法根据房价数据进行画图预测python

程序员文章站 2022-03-03 21:22:25
根据已给的波士顿房价数据,对波斯顿房价进行预测。即,实现给出若干条件(如房间数、社区的低收入阶层的比率和镇上学生与教师数量比例的部分数据),要能说出给出的条件是否能够有效进行预测,如可以做有效预测,则给出预测的结果。上面的是数据下面是ID3的算法#coding:utf-8__author__ = 'liukai'from math import logclass DecisonTree: trainData = [] trainLabel = [] feature...

根据已给的波士顿房价数据,对波斯顿房价进行预测。即,实现给出若干条件(如房间数、社区的低收入阶层的比率和镇上学生与教师数量比例的部分数据),要能说出给出的条件是否能够有效进行预测,如可以做有效预测,则给出预测的结果。
基于id3算法根据房价数据进行画图预测python
上面的是数据
下面是ID3的算法

#coding:utf-8

__author__ = 'liukai'

from math import log

class DecisonTree:
    trainData = []
    trainLabel = []
    featureValus = {} #每个特征所有可能的取值

    def __init__(self, trainData, trainLabel, threshold):
        self.loadData(trainData, trainLabel)
        self.threshold = threshold
        self.tree = self.createTree(range(0,len(trainLabel)), range(0,len(trainData[0])))


    #加载数据
    def loadData(self, trainData, trainLabel):
        if len(trainData) != len(trainLabel):
            raise ValueError('input error')
        self.trainData = trainData
        self.trainLabel = trainLabel

        #计算 featureValus
        for data in trainData:
            for index, value in enumerate(data):
                if not index in self.featureValus.keys():
                    self.featureValus[index] = [value]
                if not value in self.featureValus[index]:
                    self.featureValus[index].append(value)

    #计算信息熵
    def caculateEntropy(self, dataset):
        labelCount = self.labelCount(dataset)
        size = len(dataset)
        result = 0
        for i in labelCount.values():
            pi = i / float(size)
            result -= pi * (log(pi) /log(2))
        return result

    #计算信息增益
    def caculateGain(self, dataset, feature):
        values = self.featureValus[feature] #特征feature 所有可能的取值
        result = 0
        for v in values:
            subDataset = self.splitDataset(dataset=dataset, feature=feature, value=v)
            result += len(subDataset) / float(len(dataset)) * self.caculateEntropy(subDataset)
        return self.caculateEntropy(dataset=dataset) - result

    #计算数据集中,每个标签出现的次数
    def labelCount(self, dataset):
        labelCount = {}
        for i in dataset:
            if trainLabel[i] in labelCount.keys():
                labelCount[trainLabel[i]] += 1
            else:
                labelCount[trainLabel[i]] = 1

        return labelCount

    '''
    dataset:数据集
    features:特征集
    '''
    def createTree(self, dataset, features):

        labelCount = self.labelCount(dataset)
        #如果特征集为空,则该树为单节点树
        #计算数据集中出现次数最多的标签
        if not features:
            return max(list(labelCount.items()),key = lambda x:x[1])[0]

        #如果数据集中,只包同一种标签,则该树为单节点树
        if len(labelCount) == 1:
            # return labelCount.keys()[0]
            return labelCount.keys()

        #计算特征集中每个特征的信息增益
        l = map(lambda x : [x, self.caculateGain(dataset=dataset, feature=x)], features)

        #选取信息增益最大的特征
        feature, gain = max(l, key = lambda x: x[1])

        #如果最大信息增益小于阈值,则该树为单节点树
        #
        if self.threshold > gain:
            return max(list(labelCount.items()),key = lambda x:x[1])[0]

        tree = {}
        #选取特征子集
        subFeatures = filter(lambda x : x != feature, features)
        tree['feature'] = feature
        #构建子树
        for value in self.featureValus[feature]:
            subDataset = self.splitDataset(dataset=dataset, feature=feature, value=value)

            #保证子数据集非空
            if not subDataset:
                continue
            tree[value] = self.createTree(dataset=subDataset, features=subFeatures)
        return tree

    def splitDataset(self, dataset, feature, value):
        reslut = []
        for index in dataset:
            if self.trainData[index][feature] == value:
                reslut.append(index)
        return reslut

    def classify(self, data):
        def f(tree, data):
            if type(tree) != dict:
                return tree
            else:
                return f(tree[data[tree['feature']]], data)
        return f(self.tree, data)


if __name__ == '__main__':

    trainData = [
        [0, 0, 0, 0],
        [0, 0, 0, 1],
        [0, 1, 0, 1],
        [0, 1, 1, 0],
        [0, 0, 0, 0],
        [1, 0, 0, 0],
        [1, 0, 0, 1],
        [1, 1, 1, 1],
        [1, 0, 1, 2],
        [1, 0, 1, 2],
        [2, 0, 1, 2],
        [2, 0, 1, 1],
        [2, 1, 0, 1],
        [2, 1, 0, 2],
        [2, 0, 0, 0],
    ]

    trainLabel = [0, 0, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 0]

    tree = DecisonTree(trainData=trainData, trainLabel=trainLabel, threshold=0)
    print (tree.tree)

# {'feature': 2,
#  0: {'feature': 1, 0: dict_keys([0]),1: dict_keys([1])},
#  1: dict_keys([1])}

接下来就是画图的实现

#### -*- coding: utf-8 -*-
import matplotlib.pyplot as plt
import time
import math
from math import sin
import numpy as np

# plt.rcParams['font.sans-serif']=['SimHei'] #用来正常显示中文标签
# plt.rcParams['axes.unicode_minus']=False #用来正常显示负号

# plt.xlabel(“x轴标签”)
# plt.ylabel("y轴标签")
# plt.title("图像标题")
# plt.xlim(0,5)     在画好的图形中选取x范围内的图形片段。
# plt.ylim(0,5)     y片段
# plt.plot(x,y,linewidth=4)    设置线的宽度
# plt.plot(x,y,"g字符")     g代表绿色 后面的字符表示线的种类。如虚线,点线等

# -*- coding: utf-8 -*-
import numpy as np
import matplotlib.mlab as mlab
import matplotlib.pyplot as plt

#D.柱状图bar
import matplotlib.pyplot as plt
x = [1, 2, 3, 4, 5]
y = [2.3, 3.4, 1.2, 6.6, 7.0]

plt.figure()
plt.bar(x, y)
plt.title("bar")
plt.show()
exit()
##饼图###
labels = ['China', 'Swiss', 'USA', 'UK', 'Laos', 'Spain']
X = [222, 42, 455, 664, 454, 334]
fig = plt.figure()
plt.pie(X, labels=labels, autopct='%1.2f%%')  # 画饼图(数据,数据对应的标签,百分数保留两位小数点)
plt.title("Pie chart")

plt.show()
plt.savefig("PieChart.jpg")
exit()

x=np.arange(0,2*np.pi,0.01)
y=np.sin(x)
plt.xlabel('角度')
plt.ylabel("SIN")
# plt.ylim(-1,1)   #片段选择
plt.plot(x,y)
plt.show()


exit()

x = [1, 2, 3, 4, 5]
y = [2.3, 3.4, 1.2, 6.6, 7.0]
fig = plt.figure(figsize=(12, 6))
plt.subplot(121)
plt.plot(x, y, color='r', linestyle='-')
plt.subplot(122)
plt.title("正弦图片")
plt.plot(x, y, color='r', linestyle='--')
plt.show()
exit()
x = [1, 2, 3, 4, 5]
y = [2.3, 3.4, 1.2, 6.6, 7.0]
plt.scatter(x, y, color='r', marker='+')
plt.show()
exit()
plt.figure(figsize=(6,6))
plt.subplot(231)
plt.subplot(232)
plt.subplot(233)
plt.subplot(234)
plt.subplot(235)
plt.subplot(236)
plt.show()

exit()
x_data = [1, 2, 3, 4, 5]
y_data = [2.3, 3.4, 1.2, 6.6, 7.0]
fig = plt.figure()
ax = fig.add_subplot(1,1,1)
ax.scatter(x_data, y_data)
# plt.ion()#本次运行请注释,全局运行不要注释
plt.show()
time.sleep(20)

本文地址:https://blog.csdn.net/RONNIE_Zz/article/details/107664183

相关标签: python