1. import pandas as pd 
2. data=pd.read_excel("./data/实验3data.xlsx")# 加载数据
3. data.shape# 查看数据大小

data.head()# 查看数据形式

1. data.额外通话时长[data.额外通话时长<=0]=0
2. data.额外流量[data.额外流量<=0]=0
3. print(data["额外流量"])
4. print(data["额外通话时长"])
5. data

1. data.drop('ID',axis=1,inplace=True)
2. data

data.dtypes# 查看数据类型

data.info()         # 打印摘要

data.describe()      # 描述性统计信息

data.values          # 数据 <ndarray>

data.to_numpy()       # 数据 <ndarray> (推荐)

data.shape           # 形状 (行数, 列数)

data.columns         # 列标签 <Index>

1. data.columns.values  # 列标签 <ndarray>
2.

data.index           # 行标签 <Index>

data.index.values    # 行标签 <ndarray>

data.head(5)         # 前n行

data.tail(5)       # 尾n行

data.memory_usage()                # 占用内存(字节B)

1. User_info=data.groupby(by="流失用户")["流失用户"].count()
2. User_info=pd.DataFrame(User_info)
3. User_info

1. import pyecharts.options as opts
2. from pyecharts.charts import Pie
3. 
4. x_data = [ "流失", "未流失"]
5. y_data = [3146,882]
6. 
7. c=(
8.     Pie(init_opts=opts.InitOpts(width="160px", height="100px"))
9.     .add(
10.         series_name="访问来源",
11.         data_pair=[list(z) for z in zip(x_data, y_data)],
12.         radius=["50%", "70%"],
13.         label_opts=opts.LabelOpts(is_show=False, position="center"),
14.     )
15.     .set_global_opts(legend_opts=opts.LegendOpts(pos_left="legft", orient="vertical"))
16.     .set_series_opts(
17.         tooltip_opts=opts.TooltipOpts(
18.             trigger="item", formatter="{a} <br/>{b}: {c} ({d}%)"
19.         ),
20. # label_opts=opts.LabelOpts(formatter="{b}: {c}")
21.     )
22. #     .render("doughnut_chart.html")
23. 
24. )
25. c.render_notebook()

1. Tc_info=data.groupby(by=["套餐金额","流失用户"])["流失用户"].count()
2. Tc_info=pd.DataFrame(Tc_info)
3. Tc_info

1. Change_info=data.groupby(by=["改变行为","流失用户"])["流失用户"].count()
2. Change_info=pd.DataFrame(Change_info)
3. Change_info

1. Server_info=data.groupby(by=["服务合约","流失用户"])["流失用户"].count()
2. Server_info=pd.DataFrame(Server_info)
3. Server_info

1. Relevance_info=data.groupby(by=["关联购买","流失用户"])["流失用户"].count()
2. Relevance_info=pd.DataFrame(Relevance_info)
3. Relevance_info

1. 
2. import numpy as np 
3. r=np.array(Relevance_info)
4. list_t=r[::2]
5. list_f=r[1::2]
6. list_t.tolist()

1. Group_info=data.groupby(by=["集团用户","流失用户"])["流失用户"].count()
2. Group_info=pd.DataFrame(Group_info)
3. Group_info
4. # Group_info["流失用户"]

1. from pyecharts.charts import Bar
2. from pyecharts.faker import Faker
3. from pyecharts.globals import ThemeType
4. 
5. c = (
6.     Bar({"套餐金额与客户流失条形图": ThemeType.MACARONS})
7.     .add_xaxis(['套餐金额：1','套餐金额：2','套餐金额：3'])
8.     .add_yaxis("流失", [3065,76,5])
9.     .add_yaxis("未流失",[756,106,20])
10.     .set_global_opts(
11.         title_opts={"text": "套餐金额与客户流失条形图", "subtext": "各套餐中流失与未流失"}
12.     )
13. #      .render("套餐金额与客户流失条形图.html") 
14. )
15. c.render_notebook()