二手房数据的分析与挖掘（三）-- 描述性分析2D
实验环境
Windows7
Anaconda3
pyecharts
pandas
numpy
实验内容
在Jupyter notebook中利用numpy、pandas、pyecharts进行数据的可视化，通过数据分析完成一份基于Python的郑州市二手房数据分析报告。
实验步骤
一、数据来源
数据下载：https://download.****.net/download/qq_35809147/11176627
二、代码解析
import numpy as np
import pandas as pd
from pyecharts import online
online()
%matplotlib inline
data = pd.read_excel(u'C://Users/liulei/Desktop/正常值数据.xlsx')
data.head()
data.info()
data.loc[:,['UnitPrice', 'TotalPrice']].describe()
from pyecharts import Bar
floor_pivot = data.pivot_table(index='Floor',values='UnitPrice', aggfunc=np.mean)
floor_pivot = floor_pivot.UnitPrice.apply(lambda x : round(x, 2))
attr = floor_pivot.index.tolist()
v1 = floor_pivot.values
bar = Bar("楼层分析","朝向与单价的关系")
bar.add("楼层", attr, v1, is_more_utils=True, is_label_show=True)
bar
from pyecharts import Bar
direction_pivot = data.pivot_table(index='Direction',values='UnitPrice', aggfunc=np.mean)
direction_pivot = direction_pivot.UnitPrice.apply(lambda x : round(x, 2))
attr = direction_pivot.index.tolist()
v1 = direction_pivot.values
bar = Bar("房屋朝向分析","朝向与单价的关系")
bar.add("房屋朝向", attr, v1, mark_point=['min', 'max'], is_more_utils=True, is_label_show=True)
bar
from pyecharts import Bar
decoration_pivot = data.pivot_table(index='Decoration',values='UnitPrice', aggfunc=np.mean)
decoration_pivot = decoration_pivot.sort_values('UnitPrice', ascending=False)
decoration_pivot = decoration_pivot.UnitPrice.apply(lambda x : round(x, 2))
attr = decoration_pivot.index.tolist()
v1 = decoration_pivot.values.tolist()
bar = Bar("装修程度分析","装修程度与单价的关系")
bar.add("装修程度", attr, v1, mark_point=['min', 'max'], is_more_utils=True, is_label_show=True)
bar
from pyecharts import Bar
housetype_pivot = data.pivot_table(index='HouseType',values='UnitPrice', aggfunc=np.mean)
housetype_pivot = housetype_pivot.UnitPrice.apply(lambda x : round(x, 2))
attr = housetype_pivot.index.tolist()
v1 = housetype_pivot.values.tolist()
bar = Bar("房屋类型", "房屋类型与单价的关系")
bar.add("房屋类型", attr, v1, is_datazoom_show=True, mark_point=['min', 'max'], is_more_utils=True)
bar
from pyecharts import Bar
region_pivot = data.pivot_table(index='XiaoQu',values='UnitPrice', aggfunc=np.median)
region_pivot = region_pivot.sort_values('UnitPrice', ascending=False)
attr = region_pivot.index.values.tolist()
v1 = region_pivot.UnitPrice.values
bar = Bar("小区", "小区与单价的关系")
bar.add("小区", attr, v1, is_datazoom_show=True, mark_point=['min', 'max'], is_more_utils=True)
bar
region_data = data['QuYu'].dropna()
region_A = lambda x : x.strip().split(' ')[0]
data['QuYu_A'] = region_data.apply(region_A)
data.head()
from pyecharts import Pie
quyu_a_counts = data.groupby('QuYu_A')['QuYu_A'].value_counts()
attr = quyu_a_counts.index.tolist()
v1 = quyu_a_counts.values
pie = Pie("各区挂牌数量占比", title_pos='center')
pie.add("", attr, v1, radius=[40, 75], label_text_color=None,
is_label_show=True, legend_orient='vertical',
legend_pos='left')
pie
from pyecharts import Bar
year_pivot = data.pivot_table(index='BuildingYear', values='UnitPrice', aggfunc=np.median)
year_pivot = year_pivot.UnitPrice.apply(lambda x : round(x, 2))
attr = year_pivot.index.tolist()
v1 = year_pivot.values.tolist()
bar = Bar("房屋建造年份", "建造年份与单价的关系")
bar.add("建造年份", attr, v1, is_datazoom_show=True, mark_point=['min', 'max'], is_more_utils=True)
bar