二手房房价预测案例 -- 代码实现

 '''

    
 dist-所在区
    
 roomnum-室的数量
    
 halls-厅的数量
    
 AREA-房屋面积
    
 floor-楼层
    
 subway-是否临近地铁
    
 school-是否学区房
    
 price-平米单价
    
 '''

 import pandas as pd

    
 import numpy as np
    
 import math
    
 import matplotlib.pyplot as plt
    
 import matplotlib
    
 import seaborn as sns
    
 import statsmodels.api as sm
    
 from numpy import corrcoef,array
    
 from statsmodels.formula.api import ols

 datall=pd.read_csv(r"D:\EVAN\Python\practice\sndHsPr.csv")  '读入清洗过后的数据'

    
 print("%d",datall.shape[0])

 dat0=datall

    
 dat0.describe(include="all").T

 dat0.price=dat0.price/10000

    
 dict1 = {
    
     u'chaoyang' : "朝阳",
    
     u'dongcheng' : "东城",
    
     u'fengtai' :  "丰台",
    
     u'haidian' : "海淀",
    
     u'shijingshan' : "石景山",
    
     u'xicheng': "西城"
    
     }  
    
 dat0.dist = dat0.dist.apply(lambda x : dict1[x])
    
 dat0.head() '查看前五项数据'

 matplotlib.rcParams['axes.unicode_minus']=False "解决保存图像时负号'-'显示为方块的问题"

    
 plt.rcParams['font.sans-serif'] = ['SimHei']"指定默认字体"
    
 dat0.price.hist(bins=20)'因变量直方图'
    
 plt.xlabel("单位面积房价（万元/平方米）")
    
 plt.ylabel("频数")

 '查看price的均值、中位数和标准差等特征'

    
 print(dat0.price.agg(['mean','median','std']))  
    
 print(dat0.price.quantile([0.25,0.5,0.75]))

 '查看房价最高和最低的两条观测'

    
 pd.concat([(dat0[dat0.price==min(dat0.price)]),(dat0[dat0.price==max(dat0.price)])])

 for i in range(7):

    
     if i != 3:
    
         print(dat0.columns.values[i],":")
    
         print(dat0[dat0.columns.values[i]].agg(['value_counts']).T)
    
         print("=======================================================================")
    
     else:
    
         continue
    
 print('AREA:')
    
 print(dat0.AREA.agg(['min','mean','median','max','std']).T)

 def get_sample(df, sampling="simple_random", k=1, stratified_col=None):

    
     """
    
     对输入的 dataframe 进行抽样的函数
    
   5.     参数:
    
     - df: 输入的数据框 pandas.dataframe 对象
    
   8.         - sampling:抽样方法 str
    
         可选值有 ["simple_random", "stratified", "systematic"]
    
         按顺序分别为: 简单随机抽样、分层抽样、系统抽样
    
   12.         - k: 抽样个数或抽样比例 int or float
    
         (int, 则必须大于0; float, 则必须在区间(0,1)中)
    
         如果 0 < k < 1 , 则 k 表示抽样对于总体的比例
    
         如果 k >= 1 , 则 k 表示抽样的个数；当为分层抽样时，代表每层的样本量
    
   17.         - stratified_col: 需要分层的列名的列表 list
    
         只有在分层抽样时才生效
    
   20.     返回值:
    
     pandas.dataframe 对象, 抽样结果
    
     """
    
     import random
    
     import pandas as pd
    
     from functools import reduce
    
     import numpy as np
    
     import math
    
     
    
     len_df = len(df)
    
     if k <= 0:
    
     raise AssertionError("k不能为负数")
    
     elif k >= 1:
    
     assert isinstance(k, int), "选择抽样个数时, k必须为正整数"
    
     sample_by_n=True
    
     if sampling is "stratified":
    
         alln=k*df.groupby(by=stratified_col)[stratified_col[0]].count().count() # 有问题的
    
         #alln=k*df[stratified_col].value_counts().count() 
    
         if alln >= len_df:
    
             raise AssertionError("请确认k乘以层数不能超过总样本量")
    
     else:
    
     sample_by_n=False
    
     if sampling in ("simple_random", "systematic"):
    
         k = math.ceil(len_df * k)
    
     
    
     #print(k)
    
  
    
     if sampling is "simple_random":
    
     print("使用简单随机抽样")
    
     idx = random.sample(range(len_df), k)
    
     res_df = df.iloc[idx,:].copy()
    
     return res_df
    
  
    
     elif sampling is "systematic":
    
     print("使用系统抽样")
    
     step = len_df // k+1          #step=len_df//k-1
    
     start = 0                  #start=0
    
     idx = range(len_df)[start::step]  #idx=range(len_df+1)[start::step]
    
     res_df = df.iloc[idx,:].copy()
    
     #print("k=%d,step=%d,idx=%d"%(k,step,len(idx)))
    
     return res_df
    
  
    
     elif sampling is "stratified":
    
     assert stratified_col is not None, "请传入包含需要分层的列名的列表"
    
     assert all(np.in1d(stratified_col, df.columns)), "请检查输入的列名"
    
     
    
     grouped = df.groupby(by=stratified_col)[stratified_col[0]].count()
    
     if sample_by_n==True:
    
         group_k = grouped.map(lambda x:k)
    
     else:
    
         group_k = grouped.map(lambda x: math.ceil(x * k))
    
     
    
     res_df = df.head(0)
    
     for df_idx in group_k.index:
    
         df1=df
    
         if len(stratified_col)==1:
    
             df1=df1[df1[stratified_col[0]]==df_idx]
    
         else:
    
             for i in range(len(df_idx)):
    
                 df1=df1[df1[stratified_col[i]]==df_idx[i]]
    
         idx = random.sample(range(len(df1)), group_k[df_idx])
    
         group_df = df1.iloc[idx,:].copy()
    
         res_df = res_df.append(group_df)
    
     return res_df
    
  
    
     else:
    
     raise AssertionError("sampling is illegal")

 dat1=pd.concat([data,dat01[['school','subway','style_new','roomnum','AREA','price']]],axis=1)

    
 dat1.head()

 lm1 = ols("price ~ dist_丰台+dist_朝阳+dist_东城+dist_海淀+dist_西城+school+subway+

    
 floor_middle+floor_low+style_new+roomnum+AREA", data=dat1).fit()
    
 lm1_summary = lm1.summary()
    
 lm1_summary

 schools=['丰台','朝阳','东城','海淀','西城']

    
 print('石景山非学区房\t',round(dat0[(dat0['dist']=='石景山')&(dat0['school']==0)]['price'].mean(),2),'万元/平方米\t',
    
      '石景山学区房\t',round(dat0[(dat0['dist']=='石景山')&(dat0['school']==1)]['price'].mean(),2),'万元/平方米')
    
 print('-------------------------------------------------------------------------')
    
 for i in schools:
    
     print(i+'非学区房\t',round(dat1[(dat1['dist_'+i]==1)&(dat1['school']==0)]['price'].mean(),2),'万元/平方米\t',i+'学区房\t',round(dat1[(dat1['dist_'+i]==1)&(dat1['school']==1)]['price'].mean(),2),'万元/平方米')

 x_new1=dat1.head(1)

    
 x_new1
    
 x_new1['dist_朝阳']=0
    
 x_new1['dist_东城']=1
    
 x_new1['roomnum']=2
    
 x_new1['halls']=1
    
 x_new1['AREA_ln']=np.log(70)
    
 x_new1['subway']=1
    
 x_new1['school']=1
    
 x_new1['style_new']="有厅"

 print("单位面积房价：",round(math.exp(lm3.predict(x_new1)),2),"万元/平方米")

    
 print("总价：",round(math.exp(lm3.predict(x_new1))*70,2),"万元")