1,表分析:分析两个变量的联合分布情况
- crosstab():
pandas.crosstab(file.subway,file.school).plot(kind = 'bar')
plt.show()
- 换成百分比:(堆叠图)
subway_school = pandas.crosstab(file.subway,file.school).plot(kind = 'bar')
subway_school = subway_school.div(subway_school.sum(1),axis=0)
subway_school.plot(kind='bar',stacked=True)
plt.show()
2,分类汇总:
使用groupby()之后返回的是一个DataFrame对象
file.price.groupby(file.dist).mean().sort_values(ascending=False).plot(kind='bar')
plt.show
- 聚合函数agg():归并各个函数的结果
ret = file.price.groupby(file.dist).agg(['mean','max','min','var'])
print(ret)
- 箱形图:
import seaborn
seaborn.boxplot(x='dist',y='price',data=file)
plt.show()
- 汇总表:
file.pivot_table(values='price',index='dist',columns='school').plot(kind='bar')
plt.show()
- 散点图,scatter()
#散点图
file.plot.scatter(x='AREA',y='price')
plt.show()
- 双轴图
gdp = pandas.read_csv('C:/Users/76189/Desktop/routine项目/python学习/gdp_gdpcr.csv')
print(gdp)
x = list(gdp.year)
gdp_list = list(gdp.GDP)
gdpcr = list(gdp.GDPCR)
#绘制画布
fig = plt.figure()
ax1 = fig.add_subplot(111)
ax1.bar(x,gdp_list)
ax1.set_xlim(2000,2018)
#再第一幅的基础上画
ax2 = ax1.twinx()
ax2.plot(x,gdpcr)
plt.show()