from pyecharts import Bar,Pie
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import timedf=pd.read_excel("all_data_meituan.xlsx")df.drop(‘comment‘,axis=1).head(2)
df[‘avgPrice‘].value_counts()
# 同一家店的均价应该为同一个数值，所以这列数据没多大的意义73 17400
Name: avgPrice, dtype: int64df[‘anonymous‘].value_counts()
# 匿名评价与实名评价的比例大致在5:1左右False 14402
True 2998
Name: anonymous, dtype: int64def convertTime(x):
y=time.localtime(x/1000)
z=time.strftime("%Y-%m-%d %H:%M:%S",y)
return z
df["commentTime"]=df["commentTime"].apply(convertTime)
df["commentTime"].head()0 2018-05-09 22:21:48
1 2018-06-01 19:41:31
2 2018-04-04 11:52:23
3 2018-05-01 17:12:22
4 2018-05-17 16:48:04
Name: commentTime, dtype: object# 在excel可以用筛选器直接看到这列中的数据含有缺失值，或者在拿到数据的时候，使用df.info() 查看每列的数据信息情况
df[‘dealEndtime‘].isna().value_counts()
# 这列数据中含有177个缺失值，其余完整False 17223
True 177
Name: dealEndtime, dtype: int64df[‘commentTime‘]=pd.to_datetime(df[‘commentTime‘])
df1 = df.set_index(‘commentTime‘)
df1.resample(‘D‘).size().sort_values(ascending=False).head(100)
df2=df1.resample(‘M‘).size().to_period()
df2=df2.reset_index()
# df2.columns
# from pyecharts import Bar
bar =Bar("按月统计",width=1000,height=800)
bar.add("月统计表",df2[‘commentTime‘],df2[0],is_label_show=True, is_datazoom_show=True,is_toolbox_show=True,is_more_utils=True)
bar
df[‘commentTime‘]=pd.to_datetime(df[‘commentTime‘])
df[‘hour‘] = df[‘commentTime‘].dt.hour
df2= df.groupby([‘hour‘]).size()
df2
from pyecharts import Bar
bar =Bar("分时统计",width=1000,height=600)
bar.add("分时计表",[‘{} h‘.format(i) for i in df2.index],df2.values,is_label_show=True, is_datazoom_show=True,is_toolbox_show=True,is_more_utils=True,is_random=True)
bar
df[‘commentTime‘]=pd.to_datetime(df[‘commentTime‘])
df[‘weekday‘] = df[‘commentTime‘].dt.weekday
df2= df.groupby([‘weekday‘]).size()
# 周末吃外卖的还是教平时多了一些
from pyecharts import Bar
bar =Bar("周总计",width=750,height=400)
weekday=["一","二","三","四","五","六","日"]
bar.add("周总计",[‘周{}‘.format(i) for i in weekday],df2.values,is_label_show=True, is_datazoom_show=False,is_toolbox_show=True,is_more_utils=True,is_random=True)
bar
# 处理数据前需要先处理缺失值
# 订单结束时间清洗
df[‘dealEndtime‘].fillna(method=‘ffill‘).apply(lambda x:time.strftime("%Y-%m-%d %H:%M:%S",time.localtime(x))).head()0 2018-06-30 14:00:00
1 2018-06-30 14:00:00
2 2018-06-30 14:00:00
3 2018-06-30 14:00:00
4 2018-06-30 14:00:00
Name: dealEndtime, dtype: objectdf[‘menu‘].dropna().astype(‘category‘).value_counts()2人午晚餐 7640
单人午晚餐 3920
学生专享午晚自助 2638
4人午/晚自助 1581
单人下午自助烤肉 639
6人午/晚自助 507
周一至周五自助烤肉/周六日及节假日自助烤肉2选1 209
单人午/晚自助 67
周一至周五自助烤肉，免费WiFi 22
Name: menu, dtype: int64
df[‘readCnt‘].corr(df[‘star‘])
# 评论阅读书与客户评价分数高低的相关性0.05909293203205019df_most=df[(df["menu"]=="2人午晚餐")][‘star‘].value_counts().reindex(range(10,60,10))10 329
20 533
30 2002
40 2704
50 2072
Name: star, dtype: int64
df[(df["menu"]=="单人午晚餐")][‘star‘].value_counts()30 1215
40 1208
50 1093
20 298
10 106
Name: star, dtype: int64# 学生专享午晚自助
df[(df["menu"]=="学生专享午晚自助")][‘star‘].value_counts()40 954
50 863
30 529
20 191
10 101
Name: star, dtype: int64df[(df["menu"]=="4人午/晚自助")][‘star‘].value_counts()50 536
30 432
40 414
10 131
20 68
Name: star, dtype: int64df[(df["menu"]=="单人下午自助烤肉")][‘star‘].value_counts()30 208
50 169
40 144
10 98
20 20
Name: star, dtype: int64df[(df["menu"]=="6人午/晚自助")][‘star‘].value_counts()50 245
40 142
30 112
10 8
Name: star, dtype: int64#周一至周五自助烤肉/周六日及节假日自助烤肉2选1
df[(df["menu"]=="周一至周五自助烤肉/周六日及节假日自助烤肉2选1")][‘star‘].value_counts()50 87
40 66
30 46
20 10
Name: star, dtype: int64df[(df["menu"]=="单人午/晚自助")][‘star‘].value_counts()50 30
40 27
30 10
Name: star, dtype: int64df[(df["menu"]=="周一至周五自助烤肉，免费WiFi")][‘star‘].value_counts().reindex(range(10,51,10)).fillna(0)10 0.0
20 0.0
30 0.0
40 0.0
50 22.0
Name: star, dtype: float64# df.groupby([‘menu‘,‘star‘]).size().to_excel("all_menu_star.xls")
df.groupby([‘menu‘,‘star‘]).size()menu star
2人午晚餐 10 329
20 533
30 2002
40 2704
50 2072
4人午/晚自助 10 131
20 68
30 432
40 414
50 536
6人午/晚自助 10 8
30 112
40 142
50 245
单人下午自助烤肉 10 98
20 20
30 208
40 144
50 169
单人午/晚自助 30 10
40 27
50 30
单人午晚餐 10 106
20 298
30 1215
40 1208
50 1093
周一至周五自助烤肉/周六日及节假日自助烤肉2选1 20 10
30 46
40 66
50 87
周一至周五自助烤肉，免费WiFi 50 22
学生专享午晚自助 10 101
20 191
30 529
40 954
50 863
dtype: int64df.groupby([‘star‘,‘menu‘,]).size()star menu
10 2人午晚餐 329
4人午/晚自助 131
6人午/晚自助 8
单人下午自助烤肉 98
单人午晚餐 106
学生专享午晚自助 101
20 2人午晚餐 533
4人午/晚自助 68
单人下午自助烤肉 20
单人午晚餐 298
周一至周五自助烤肉/周六日及节假日自助烤肉2选1 10
学生专享午晚自助 191
30 2人午晚餐 2002
4人午/晚自助 432
6人午/晚自助 112
单人下午自助烤肉 208
单人午/晚自助 10
单人午晚餐 1215
周一至周五自助烤肉/周六日及节假日自助烤肉2选1 46
学生专享午晚自助 529
40 2人午晚餐 2704
4人午/晚自助 414
6人午/晚自助 142
单人下午自助烤肉 144
单人午/晚自助 27
单人午晚餐 1208
周一至周五自助烤肉/周六日及节假日自助烤肉2选1 66
学生专享午晚自助 954
50 2人午晚餐 2072
4人午/晚自助 536
6人午/晚自助 245
单人下午自助烤肉 169
单人午/晚自助 30
单人午晚餐 1093
周一至周五自助烤肉/周六日及节假日自助烤肉2选1 87
周一至周五自助烤肉，免费WiFi 22
学生专享午晚自助 863
dtype: int64df.groupby([‘star‘,‘menu‘,]).size()[50]menu
2人午晚餐 2072
4人午/晚自助 536
6人午/晚自助 245
单人下午自助烤肉 169
单人午/晚自助 30
单人午晚餐 1093
周一至周五自助烤肉/周六日及节假日自助烤肉2选1 87
周一至周五自助烤肉，免费WiFi 22
学生专享午晚自助 863
dtype: int64
# userId
# 这家店铺有好多回头客，万万没想到
df[df[‘userId‘]!=0][‘userId‘].value_counts().head(40)266045270 64
152775497 60
80372612 60
129840082 60
336387962 60
34216474 60
617772217 60
82682689 54
287219504 49
884729389 45
868838851 40
409054441 40
86939815 40
776086712 40
48597225 40
111808598 40
240199490 40
83068123 40
298504911 40
1042639014 40
912472277 40
98198819 40
1494880345 40
152930400 40
139581136 40
404183587 40
714781743 40
292809386 40
18111538 40
1097689674 40
300905323 40
232697160 40
141718492 40
879430090 40
696143486 40
13257519 40
983797146 40
911947863 40
993057629 40
494215297 40
Name: userId, dtype: int64df[df[‘userName‘]!="匿名用户"][‘userName‘].value_counts().head(40)xuruiss1026 64
黑发飘呀飘 60
么么哒我是你聪叔 60
jIx325233926 60
siisgood 60
vTF610712604 60
始于初见的你 60
yumengkou 54
Daaaav 49
梁子7543 45
oev575457132 40
oUI806055883 40
joF498901567 40
liE32679330 40
张齐齐123 40
VPA342570392 40
kingd123 40
Nqr695642404 40
Mvo148723747 40
ree177064067 40
大游 40
_qq3sh1369887220 40
bQl271583480 40
凯蒂宝 40
安然～轩 40
FQe845913598 40
清晨cxh98 40
cBj31240225 40
天蛟Wing 40
oMz861346972 40
热带鱼7697 40
Mqg827794346 40
nXu534267448 40
aYH197128794 40
榴莲馅月饼 40
leeman666888 40
迅行天下 40
滨海之恋33 40
pHO437742850 40
SzX539077433 40
Name: userName, dtype: int64df.groupby([‘star‘,‘userLevel‘,]).size()star userLevel
10 0 187
1 139
2 164
3 193
4 80
5 10
20 0 223
1 88
2 304
3 294
4 207
5 21
30 0 1147
1 405
2 1057
3 1230
4 570
5 165
6 20
40 0 870
1 432
2 1360
3 1751
4 1026
5 261
6 25
50 0 698
1 386
2 1167
3 1670
4 802
5 318
6 130
dtype: int64df_level_star = df.groupby([‘userLevel‘,‘star‘]).size()
attr = np.arange(10,60,10)
from pyecharts import Bar
bar = Bar("用户等级与评分",title_pos="center")
df_0 = df_level_star[0].values
df_1 = df_level_star[1].values
df_2 = df_level_star[2].values
df_3 = df_level_star[3].values
df_4 = df_level_star[4].values
df_5 = df_level_star[5].values
# df_6 = df_level_star[6].values
df_6 = df_level_star[6].reindex(attr).fillna(0).values
bar.add("level 0",attr,df_0,is_label_show=True,legend_pos=‘right‘,legend_orient=‘vertical‘,label_text_size=12)
bar.add("level 1",attr,df_1,is_label_show=True,legend_pos=‘right‘,legend_orient=‘vertical‘,label_text_size=12)
bar.add("level 2",attr,df_2,is_label_show=True,legend_pos=‘right‘,legend_orient=‘vertical‘,label_text_size=12)
bar.add("level 3",attr,df_3,mark_line=["average"],mark_point=[‘max‘,‘min‘],is_label_show=True,legend_pos=‘right‘,legend_orient=‘vertical‘,label_text_size=12)
bar.add("level 4",attr,df_4,is_label_show=True,legend_pos=‘right‘,legend_orient=‘vertical‘,label_text_size=12)
bar.add("level 5",attr,df_5,is_label_show=True,legend_pos=‘right‘,legend_orient=‘vertical‘,label_text_size=12)
bar.add("level 6",attr,df_6,is_label_show=True,legend_pos=‘right‘,legend_orient=‘vertical‘,label_text_size=12)
bar<div id="5fcf9a4be1814dae9a66e63db26848a9" style="width:800px;height:400px;"></div>bar = Bar("用户等级与评分",title_pos="center",title_color="red")
attr = np.arange(10,60,10)
df_0 = df_level_star[0].values
df_1 = df_level_star[1].values
df_2 = df_level_star[2].values
df_3 = df_level_star[3].values
df_4 = df_level_star[4].values
df_5 = df_level_star[5].values
# df_6 = df_level_star[6].values
df_6 = df_level_star[6].reindex(attr).fillna(0).values
bar.add("level 0",attr,df_0,legend_pos=‘right‘,legend_orient=‘vertical‘,label_text_size=12)
bar.add("level 1",attr,df_1,legend_pos=‘right‘,legend_orient=‘vertical‘,label_text_size=12)
bar.add("level 2",attr,df_2,legend_pos=‘right‘,legend_orient=‘vertical‘,label_text_size=12)
bar.add("level 3",attr,df_3,is_stack=True,legend_pos=‘right‘,legend_orient=‘vertical‘,label_text_size=12)
bar.add("level 4",attr,df_4,is_stack=True,legend_pos=‘right‘,legend_orient=‘vertical‘,label_text_size=12)
bar.add("level 5",attr,df_5,is_stack=True,legend_pos=‘right‘,legend_orient=‘vertical‘,label_text_size=12)
bar.add("level 6",attr,df_6,is_stack=True,legend_pos=‘right‘,legend_orient=‘vertical‘,label_text_size=12)
bar
df[‘star‘].corr(df[‘userLevel‘])0.14389808871897794df_zan=df[‘zanCnt‘].value_counts()
from pyecharts import Bar
bar=Bar("点赞统计")
bar.add("点赞分布",df_zan.index[1:],df_zan.values[1:],is_label_show=True)
bar<div id="3123fe244a684d7e97c8c3d9f47aa715" style="width:800px;height:400px;"></div>df.describe().dataframe tbody tr th:only-of-type { vertical-align: middle }
.dataframe tbody tr th { vertical-align: top }
.dataframe thead th { text-align: right }
| avgPrice | dealEndtime | did | readCnt | replyCnt | reviewId | star | userId | userLevel | zanCnt | hour | weekday | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| count | 17400.0 | 1.722300e+04 | 1.740000e+04 | 17400.000000 | 17400.000000 | 1.740000e+04 | 17400.000000 | 1.740000e+04 | 17400.000000 | 17400.000000 | 17400.000000 | 17400.000000 |
| mean | 73.0 | 1.529633e+09 | 4.376319e+07 | 1622.936149 | 0.032759 | 1.443980e+09 | 37.691954 | 3.224900e+08 | 2.335230 | 0.096264 | 14.955460 | 3.152356 |
| std | 0.0 | 5.730086e+06 | 5.749815e+06 | 4981.816447 | 0.260349 | 2.208396e+08 | 10.813002 | 3.914649e+08 | 1.470979 | 0.511591 | 5.046872 | 2.044944 |
| min | 73.0 | 1.483078e+09 | 1.330754e+06 | 20.000000 | 0.000000 | 1.093178e+09 | 10.000000 | 0.000000e+00 | 0.000000 | 0.000000 | 0.000000 | 0.000000 |
| 25% | 73.0 | 1.530338e+09 | 4.432824e+07 | 162.000000 | 0.000000 | 1.197515e+09 | 30.000000 | 4.527015e+07 | 1.000000 | 0.000000 | 11.000000 | 1.000000 |
| 50% | 73.0 | 1.530338e+09 | 4.432824e+07 | 304.000000 | 0.000000 | 1.606347e+09 | 40.000000 | 1.527755e+08 | 3.000000 | 0.000000 | 15.000000 | 3.000000 |
| 75% | 73.0 | 1.530338e+09 | 4.432853e+07 | 751.000000 | 0.000000 | 1.646467e+09 | 50.000000 | 4.859086e+08 | 3.000000 | 0.000000 | 19.000000 | 5.000000 |
| max | 73.0 | 1.530338e+09 | 4.597465e+07 | 77837.000000 | 4.000000 | 1.698204e+09 | 50.000000 | 1.771740e+09 | 6.000000 | 8.000000 | 23.000000 | 6.000000 |
df[‘userLevel‘].value_counts().reindex(range(7))0 3125
1 1450
2 4052
3 5138
4 2685
5 775
6 175
Name: userLevel, dtype: int64df_level=df[‘userLevel‘].value_counts().reindex(range(7))
from pyecharts import Pie
pie=Pie("用户等级分布",title_pos="center",width=900)
pie.add("levels distribution",["level "+str(i) for i in range(7)],df_level.values,is_random=True,radidus=[30,45],legend_pos=‘left‘,rosetype=‘area‘,legend_orient=‘vertical‘,is_label_show=True,label_text_size=20)
pie
原文地址：https://www.cnblogs.com/onemorepoint/p/9461628.html
时间： 2024-11-06 07:25:30