import pandas as pd
import numpy as np
data = pd.read_csv("PelicanStores.csv")
data = data.iloc[:,:8]#过滤数据选择前8列
data.head()#看一下前五行

#Items（购买商品的总件数）百分数频数分布
Items_percent_frequency_distribution = (pd.value_counts(data["Items"])/np.size(data["Items"]))*100
Items_percent_frequency_distribution

#Net Sales(净销售额)百分数频数分布
bins = [0, 50, 100, 150, 200, 250, 300]
data["Net Sales Group"] = pd.cut(data["Net Sales"], bins, right=False)#对数据进行分组
Net_Sales_percent_frequency_distribution = (pd.value_counts(data["Net Sales Group"])/np.size(data["Net Sales Group"]))*100
Net_Sales_percent_frequency_distribution

#Age(年龄)百分数频数分布
bins = [20, 30, 40, 50, 60, 70, 80]
data["Age_Group"] = pd.cut(data["Age"], bins, right=False)
Age_percent_frequency_distribution = (pd.value_counts(data["Age_Group"])/np.size(data["Age_Group"]))*100
Age_percent_frequency_distribution

#购物数量有两个维度，购买商品的总件数和销售额
M_P_Items_Net_Sales = data.groupby(by=["Method of Payment"])["Items","Net Sales"].agg({"Items":np.sum,"Net Sales":np.sum})
M_P_Items_Net_Sales = M_P_Items_Net_Sales.sort_values(by="Net Sales",ascending=False)
M_P_Items_Net_Sales

#条形图
from pyecharts import Bar, Grid
x_line = [i for i in M_P_Items_Net_Sales.index.format()]
y_line_Items = [i[0] for i in M_P_Items_Net_Sales.values]
y_line_NS = [i[1] for i in M_P_Items_Net_Sales.values]
bar1 = Bar("Items",title_text_size=26, title_top=10, title_pos=10, width=1000, height=400, background_color="rgb(255,255,255)")
bar1.add("", x_line, y_line_Items ,xaxis_label_textsize=12,bar_category_gap='20%')
bar2 = Bar("Net Sales",title_text_size=26, title_top=300, title_pos=10, width=1000, height=400, background_color="rgb(255,255,255)")
bar2.add("", x_line, y_line_NS ,xaxis_label_textsize=12,bar_category_gap='20%')
grid = Grid(height=600, width=1000)
grid.add(bar1, grid_bottom="60%")
grid.add(bar2, grid_top="60%")
grid

#饼图
from pyecharts import Pie
pie1 = Pie("Items", title_text_size=26, title_top=10, title_pos=10)
pie1.add("", x_line, y_line_Items, is_label_show=True, is_legend_show=False, label_text_size=10,center=[30,50], radius=[0,45])
pie2 = Pie("Net Sales", title_text_size=26, title_top=10, title_pos=500)
pie2.add("", x_line, y_line_NS, is_label_show=True, is_legend_show=False, label_text_size=10,legend_orient='vertical',center=[70,50], radius=[0,45])
grid = Grid(height=600, width=1000)
grid.add(pie1, grid_left="60%")
grid.add(pie2, grid_right="60%")
grid

data_CN_table = data.pivot_table(values=["Net Sales"], index=["Age_Group"], columns=["Type of Customer"], aggfunc=[np.sum], fill_value=0)
data_CN_table

from pyecharts import Scatter
data_NS_Age = data[["Net Sales", "Age"]]
x_line_age = [i[1] for i in data_NS_Age.values]
y_line_NS = [i[0] for i in data_NS_Age.values]
scatter = Scatter()
scatter.add("", x_line_age, y_line_NS)

data["Net Sales"].describe()#净销售额的描述统计量

data.groupby(["Type of Customer"])["Net Sales"].describe() #分组进行统计描述

#相关性分析
data_NS_Age.corr()