Python实战项目——O2O_优惠券使用情况分析（五）

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns  #绘图模块，基于matplotlib的可视化python包,不能完全替代matplotlib,只是对matplotlib进行升级
plt.rcParams['font.sans-serif'] = ['SimHei']  # 用来正常显示中文标签
plt.rcParams['axes.unicode_minus'] = False  # 用来正常显示负号
%matplotlib inline

#pip install seaborn

offline = pd.read_csv('ccf_offline_stage1_train.csv',parse_dates=['Date_received','Date'])
offline.info()  #175+万条数据

offline.head(10)  

offline.isnull().sum()

offline['Discount_rate'] = offline['Discount_rate'].fillna('null')
offline.head()

def discount_rate_opt(s): #s代表每一个元素
    if ':' in s:
        split = s.split(':')
        discount_rate = (int(split[0]) - int(split[1]))/int(split[0])
        return round(discount_rate,2)  #折扣率保留两位小数
    elif s == 'null':
        return np.NaN
    else:
        return float(s)

offline['Discount_rate']  = offline['Discount_rate'].map(discount_rate_opt)
offline.head()

np.all([True,False,True])

cpon_no_consume = offline[(offline['Date'].isnull() & offline['Coupon_id'].notnull())]
no_cpon_no_consume = offline[(offline['Date'].isnull() & offline['Coupon_id'].isnull())]
no_cpon_consume = offline[(offline['Date'].notnull() & offline['Coupon_id'].isnull())]
cpon_consume = offline[(offline['Date'].notnull() & offline['Coupon_id'].notnull())]

print('有券未消费:{}'.format(len(cpon_no_consume)))
print('无券未消费:{}'.format(len(no_cpon_no_consume)))  #无意义，不需分析
print('无券消费:{}'.format(len(no_cpon_consume)))
print('有券消费:{}'.format(len(cpon_consume)))

#绘制饼图占比
consume_status_dict = {'cpon_no_consume':len(cpon_no_consume),'no_cpon_consume':len(no_cpon_consume),'cpon_consume':len(cpon_consume)}
consume_status = pd.Series(consume_status_dict)
consume_status

#消费方式构成的饼图（figure：看作是一张画布，axes：代表画布内的多个坐标系）
fig,ax=plt.subplots(1,1,figsize=(8,10))
consume_status.plot.pie(ax = ax,
                       autopct='%1.1f%%',
                       shadow=True,
                        explode=[0.02,0.05,0.2],
                        textprops={'fontsize':15,'color':'blue'},
                        wedgeprops={'linewidth':1,'edgecolor':'black'},
                        labels=['有券未消费 \n ({})'.format(len(cpon_no_consume)),
                                '无券消费 \n ({})'.format(len(no_cpon_consume)),
                               '用券消费 \n ({})'.format(len(cpon_consume))
                               ]
                       )
ax.set_ylabel('')  #去除ylable
ax.set_title('消费占比情况')
plt.legend(labels=['有券未消费','无券消费','用券消费'])

Merchant_distance = cpon_consume.groupby('Merchant_id')['Distance'].mean()
Merchant_distance[Merchant_distance==0]

Merchant_discount_rate = cpon_consume.groupby('Merchant_id')['Discount_rate'].mean()
Merchant_discount_rate.sort_values()
Merchant_discount_rate.hist()
Merchant_discount_rate.mean()  #所有商家平均折扣的平均值：0.88
Merchant_discount_rate

popular_merchant = cpon_consume.groupby('Merchant_id')['User_id'].apply(lambda x:len(x.unique())).sort_values(ascending=False)
#找出持券消费人数>500的商家id
popular_merchant500 = popular_merchant[popular_merchant>500]
popular_merchant500.name = 'customer_count'  #指定列名为消费者数量（持券消费者）
print(len(popular_merchant500))
print(popular_merchant500)

merchant_pop_dis = pd.merge(left=popular_merchant500,right=Merchant_distance,on='Merchant_id',how='inner')
merchant_pop_dis_rate = pd.merge(left=merchant_pop_dis,right=Merchant_discount_rate,on='Merchant_id',how='inner')
merchant_pop_dis_rate

merchant_pop_dis_rate.corr()

sns.heatmap(data=merchant_pop_dis_rate.corr(),annot=True,cmap='Accent',vmax=1,vmin=-1)

consume_num_everday =  cpon_consume[['User_id','Date_received']]
consume_num_everday = consume_num_everday.groupby('Date_received').count()
consume_num_everday = consume_num_everday.rename(columns={'User_id':'count'})
consume_num_everday

coupon_sendout_everyday = offline[offline['Date_received'].notnull()][['Date_received','User_id']]
coupon_sendout_everyday = coupon_sendout_everyday.groupby('Date_received').count()
coupon_sendout_everyday = coupon_sendout_everyday.rename(columns={'User_id':'count'})
coupon_sendout_everyday

plt.figure(figsize=(18,6))
# plt.bar(x=date_receive_sort,height=coupon_sendout_everyday['count'],label='每天发券量')
plt.bar(x=date_sort,height=consume_num_everday['count'],label='每天用券量')
plt.yscale('log')  #对y轴进行对数缩放
plt.legend()

plt.figure(figsize=(18,6))
plt.bar(x=date_receive_sort,height=consume_num_everday['count']/coupon_sendout_everyday['count'],
       label='百分比')
plt.legend()