标签：source log sort proc ossh ext mod ips code
如今大数据行业十分火热，本人认为python是比较强大的分析工具，在网易云课堂上学习了python数据分析。做了案例，写下代码分析过程以及分析结论。
以下是电商打折套路的python数据分析项目。
# -*- coding: utf-8 -*-
"""
Created on Wed Jan 9 15:31:45 2019
@author: Administrator
"""
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import warnings
from datetime import datetime
from bokeh.transform import jitter
warnings.filterwarnings(‘ignore‘)
from bokeh.plotting import figure ,show,output_file
from bokeh.models import ColumnDataSource
#导入数据
import os
os.chdir(‘C:\\Users\\Administrator\\Desktop\\python项目\\2电商打折‘)
#工作路径
df=pd.read_excel(‘双十一数据.xlsx‘,sheetname=0)
df.fillna(0,inplace=True)
df.index=df[‘update-time‘]
df[‘date‘]=df.index.day
#双十一当天在售商品占比数
data1=df[["id","title","店名","date"]]
d1=data1[["id","date"]].groupby(by="id").agg(["max","min"])["date"]
#统计不同商品的销售开始和结束日期
id_11=data1[data1["date"]==11]["id"]
d2=pd.DataFrame({"id":id_11,"双十一是否售卖":True})
id_data=pd.merge(d1,d2,left_index=True,right_on="id",how="left")
id_data.fillna(False,inplace=True)
#双十一当天参与活动的商品个数与比例
m=len(d1)
m_11=len(id_11)
m_pre=m_11/m
print("双十一当天参与活动的商品个数是%i个，比例是%.2f%%"%(m_11,m_pre*100))
结论：双十一当天参与活动的商品个数是405个，比例是74.18%
#------------------------------------------------------------------------
#商品销售分类
id_data["type"]="待分类"
id_data["type"][(id_data["min"]<11)&(id_data["max"]>11)]="A"
id_data["type"][(id_data["min"]<11)&(id_data["max"]==11)]="B"
id_data["type"][(id_data["min"]==11)&(id_data["max"]>11)]="C"
id_data["type"][(id_data["min"]==11)&(id_data["max"]==11)]="D"
id_data["type"][(id_data["双十一是否售卖"]==False)]="F"
id_data["type"][(id_data["max"]<11)]="E"
id_data["type"][(id_data["min"]>11)]="G"
result1=id_data["type"].value_counts()
result1=result1.loc[["A","B","C","D","E","F","G"]]
#不同类别商品比例
from bokeh.palettes import brewer
colori=brewer["YlGn"][7]
plt.axis("equal")
plt.pie(result1,labels=result1.index,autopct="%.2f%%",colors=colori,
startangle=90,radius=1.5,counterclock=True)
#------------------------------------------------------------------------
#未参与双十一活动的商品去向如何
id_not11=id_data[id_data["双十一是否售卖"]==False]#暂时下架商品----id_con2
df_not11=id_not11[["id","type"]]
data_not11=pd.merge(df_not11,df,on="id",how="left")#分组字段不够用需要从原始总数据里借，所以要合并
#不合并就没法分组，没法分组，就没法统计
id_con1=id_data["id"][id_data["type"]=="F"].values
data_con2=data_not11[["id","title","date"]].groupby(by=["id","title"]).count()
title_count=data_con2.reset_index()["id"].value_counts()
id_con2=title_count[title_count>1].index
data_con3=data_not11[data_not11["title"].str.contains("预售")]
id_con3=data_con3["id"].value_counts().index
print("未参与双十一当天活动的商品里，%i个为暂时下架商品，%i个为重新上架商品，%i个为预售商品"%
(len(id_con1),len(id_con2),len(id_con3))
)
结论：未参与双十一当天活动的商品里，95个为暂时下架商品，155个为重新上架商品，69个为预售商品
#------------------------------------------------------------------------
#商品销售分类
id_data["type"]="待分类"
id_data["type"][(id_data["min"]<11)&(id_data["max"]>11)]="A"
id_data["type"][(id_data["min"]<11)&(id_data["max"]==11)]="B"
id_data["type"][(id_data["min"]==11)&(id_data["max"]>11)]="C"
id_data["type"][(id_data["min"]==11)&(id_data["max"]==11)]="D"
id_data["type"][(id_data["双十一是否售卖"]==False)]="F"
id_data["type"][(id_data["max"]<11)]="E"
id_data["type"][(id_data["min"]>11)]="G"
result1=id_data["type"].value_counts()
result1=result1.loc[["A","B","C","D","E","F","G"]]
#不同类别商品比例
from bokeh.palettes import brewer
colori=brewer["YlGn"][7]
plt.axis("equal")
plt.pie(result1,labels=result1.index,autopct="%.2f%%",colors=colori,
startangle=90,radius=1.5,counterclock=True)
#------------------------------------------------------------------------
#未参与双十一活动的商品去向如何
id_not11=id_data[id_data["双十一是否售卖"]==False]#暂时下架商品----id_con2
df_not11=id_not11[["id","type"]]
data_not11=pd.merge(df_not11,df,on="id",how="left")#分组字段不够用需要从原始总数据里借，所以要合并
#不合并就没法分组，没法分组，就没法统计
id_con1=id_data["id"][id_data["type"]=="F"].values
data_con2=data_not11[["id","title","date"]].groupby(by=["id","title"]).count()
title_count=data_con2.reset_index()["id"].value_counts()
id_con2=title_count[title_count>1].index
data_con3=data_not11[data_not11["title"].str.contains("预售")]
id_con3=data_con3["id"].value_counts().index
print("未参与双十一当天活动的商品里，%i个为暂时下架商品，%i个为重新上架商品，%i个为预售商品"%
(len(id_con1),len(id_con2),len(id_con3))
)
#------------------------------------------------------------------------
data_11sale=id_11
data_11sale_final=np.hstack((data_11sale,id_con3))
result2_i=pd.DataFrame({"id":data_11sale_final})
x1=pd.DataFrame({"id":id_11})
x1_df=pd.merge(x1,df,on="id",how="left")
brand_11sale=x1_df.groupby(by="店名")["id"].count()
x2=pd.DataFrame({"id":id_con3})
x2_df=pd.merge(x2,df,on="id",how="left")
brand_ys=x2_df.groupby(by="店名")["id"].count()
result2_data=pd.DataFrame({"当天参与活动的商品数量":brand_11sale,
"预售商品数量":brand_ys})
result2_data["总量"]=result2_data["当天参与活动的商品数量"]+result2_data["预售商品数量"]
result2_data.sort_values(by="总量",ascending=False)
from bokeh.models import HoverTool
from bokeh.core.properties import value
lst_brand=result2_data.index.tolist()
lst_type=result2_data.columns.tolist()[:2]#result2_data的列名columns.取前2个
color=["red","green"]
result2_data.index.name="brand"
result2_data.columns=["sale_on_11","presell","sum"]
source1=ColumnDataSource(result2_data)
hover=HoverTool(
tooltips=[
("品牌","@brand"),
("双十一当天参与活动商品数量","@sale_on_11"),
("预售商品数量","@presell"),
("商品总数","@sum")
])
output_file("project08.html")
p=figure(x_range=lst_brand,plot_width=900,plot_height=350,
title="各个品牌参与双十一活动的情况",
tools=[hover,"box_select,pan,reset,wheel_zoom,crosshair"]
)
p.vbar(top="sum",x="brand",source=source1,width=0.9,
#color=color,alpha=0.7,
#legend=[value(x) for x in lst_type],
muted_color="black", muted_alpha=0.2
)
show(p)
#不同品牌销售数量情况
#------------------------------------------------------------------------
标签：source log sort proc ossh ext mod ips code
原文地址：https://www.cnblogs.com/nigulasiximegn/p/14962943.html
