036 pandas常用

1.pandas基础,Series,DataFrame

1
2
import pandas as pd
import numpy as np    #导包

1
2
s1 = pd.Series([4,7,-5,3])#创建一个series，索引为默认值
print(s1)

执行结果:

1
2
3
4
5
0    4
1    7
2   -5
3    3
dtype: int64

1
s1.values #series的值

执行结果:

1
array([ 4,  7, -5,  3], dtype=int64)

1
s1.index#series的索引

执行结果:

1
RangeIndex(start=0, stop=4, step=1)

1
2
s2 = pd.Series([4.0,6.5,-0.5,4.2],index=['d','b','a','c'])  #带索引创建一个series
print(s2)

执行结果:

1
2
3
4
5
d    4.0
b    6.5
a   -0.5
c    4.2
dtype: float64

1
s2['a']#根据索引取值,跟列表相似

执行结果:

1
-0.5

1
s2[['a','b','c']]#根据索引取多行值,记得是双括号,单个值是单括号

执行结果:

1
2
3
4
a   -0.5
b    6.5
c    4.2
dtype: float64

1
'b' in s2   #判断索引在不在series中

执行结果:

1
True

1
'e' in s2

执行结果:

1
False

1
2
3
4
#Series可以看成是一个定长的有序字典,使用字典创建series
dic1 = {'apple':5,'pen':3,'applepen':10}
s3 = pd.Series(dic1)
print(s3)

执行结果:

1
2
3
4
apple        5
pen          3
applepen    10
dtype: int64

1
2
3
4
5
6
7
#DataFrame  ,使用字典创建dataframe
data = {'year':[2014,2015,2016,2017],
        'income':[10000,30000,50000,80000],
        'pay':[5000,20000,30000,30000]
}
df1 = pd.DataFrame(data)
df1

执行结果:

1
2
3
4
5
    year    income  pay
0   2014    10000   5000
1   2015    30000   20000
2   2016    50000   30000
3   2017    80000   30000

1
2
df2 = pd.DataFrame(np.arange(12).reshape((3,4)))
df2

执行结果:

1
2
3
4
    0   1   2   3
0   0   1   2   3
1   4   5   6   7
2   8   9   10  11

1
2
df3 = pd.DataFrame(np.arange(12).reshape((3,4)),index=['a','c','b'],columns=[2,33,44,5])   #创建dataframe时指定索引和列名
df3

执行结果:

1
2
3
4
    2   33  44  5
a   0   1   2   3
c   4   5   6   7
b   8   9   10  11

1
df1.columns #列,取出列名

执行结果:

1
Index(['year', 'income', 'pay'], dtype='object')

1
df1.index #行,取出行索引

执行结果:

1
RangeIndex(start=0, stop=4, step=1)

1
df1.values   #打印dataframe的所有值,是一个数组的形式

执行结果:

1
2
3
4
array([[ 2014, 10000,  5000],
       [ 2015, 30000, 20000],
       [ 2016, 50000, 30000],
       [ 2017, 80000, 30000]], dtype=int64)

1
df1.describe()    #describe描述表,对数据做了一些分析,计数,平均值,标准差,最小值,最大值等等

执行结果:

1
2
3
4
5
6
7
8
9
    year    income  pay
count   4.000000    4.000000    4.000000
mean    2015.500000 42500.000000    21250.000000
std 1.290994    29860.788112    11814.539066
min 2014.000000 10000.000000    5000.000000
25% 2014.750000 25000.000000    16250.000000
50% 2015.500000 40000.000000    25000.000000
75% 2016.250000 57500.000000    30000.000000
max 2017.000000 80000.000000    30000.000000

1
df1.T   #对dataframe进行转置

执行结果:

1
2
3
4
    0   1   2   3
year    2014    2015    2016    2017
income  10000   30000   50000   80000
pay 5000    20000   30000   30000

1
df3

执行结果:

1
2
3
4
    2   33  44  5
a   0   1   2   3
c   4   5   6   7
b   8   9   10  11

1
df3.sort_index(axis=1)#列排序,根据列索引名排序

执行结果:

1
2
3
4
    2   5   33  44
a   0   3   1   2
c   4   7   5   6
b   8   11  9   10

1
df3.sort_index(axis=0)#行排序,根据行索引名排序

执行结果:

1
2
3
4
2   33  44  5
a   0   1   2   3
b   8   9   10  11
c   4   5   6   7

1
df3.sort_values(by=44)   #根据列名对值进行排序,全部会变

执行结果:

1
2
3
4
2   33  44  5
a   0   1   2   3
c   4   5   6   7
b   8   9   10  11

2 pandas选择数据

1
2
import pandas as pd
import numpy as np

1
2
3
4
5
#创建一个dataframe
dates = pd.date_range('20170101',periods=6)   #periods是周期的意思,初始值:20170101
print(dates)
df1 = pd.DataFrame(np.arange(24).reshape((6,4)),index=dates,columns=['A','B','C','D'])
df1

执行结果:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
DatetimeIndex(['2017-01-01', '2017-01-02', '2017-01-03', '2017-01-04',
               '2017-01-05', '2017-01-06'],
              dtype='datetime64[ns]', freq='D')
A   B   C   D
2017-01-01  0   1   2   3
2017-01-02  4   5   6   7
2017-01-03  8   9   10  11
2017-01-04  12  13  14  15
2017-01-05  16  17  18  19
2017-01-06  20  21  22  23

1
df1['A']#将DataFrame的列获取为一个Series

执行结果:

1
2
3
4
5
6
7
2017-01-01     0
2017-01-02     4
2017-01-03     8
2017-01-04    12
2017-01-05    16
2017-01-06    20
Freq: D, Name: A, dtype: int32

1
df1.A    ##将DataFrame的列获取为一个Series,跟上面的方法是一样的

执行结果:

1
2
3
4
5
6
7
2017-01-01     0
2017-01-02     4
2017-01-03     8
2017-01-04    12
2017-01-05    16
2017-01-06    20
Freq: D, Name: A, dtype: int32

1
df1[0:2]#取0-1行的dataframe

执行结果:

1
2
3
    A   B   C   D
2017-01-01  0   1   2   3
2017-01-02  4   5   6   7

1
df1['20170102':'20170104']   #通过索引取多行

执行结果:

1
2
3
4
    A   B   C   D
2017-01-02  4   5   6   7
2017-01-03  8   9   10  11
2017-01-04  12  13  14  15

1
2
#通过行标签选择数据
df1.loc['20170102']

执行结果:

1
2
3
4
5
A    4
B    5
C    6
D    7
Name: 2017-01-02 00:00:00, dtype: int32

1
2
#通过行标签及列名取数据,先取出行,再取列
df1.loc['20170101',['A','C']]

执行结果:

1
2
3
A    0
C    2
Name: 2017-01-01 00:00:00, dtype: int32

1
df1.loc[:,['A','B']]   #取出所有行的某几列

执行结果:

1
2
3
4
5
6
7
    A   B
2017-01-01  0   1
2017-01-02  4   5
2017-01-03  8   9
2017-01-04  12  13
2017-01-05  16  17
2017-01-06  20  21

1
2
#通过位置选择数据
df1.iloc[2] #第二行

执行结果:

1
2
3
4
5
A     8
B     9
C    10
D    11
Name: 2017-01-03 00:00:00, dtype: int32

1
df1.iloc[1:3,2:4]   #取得是2-3行,3-4列

执行结果:

1
2
3
C   D
2017-01-02  6   7
2017-01-03  10  11

1
df1.iloc[[1,2,4],[1,3]]   #取2,3,5行的2,4列

执行结果:

1
2
3
4
    B   D
2017-01-02  5   7
2017-01-03  9   11
2017-01-05  17  19

1
2
#混合标签位置选择(标签加位置)
df1.ix[2:4,['A','C']]

执行结果:

1
2
3
    A   C
2017-01-03  8   10
2017-01-04  12  14

1
df1.ix['20170102':'20170104',2:4]

执行结果:

1
2
3
4
    C   D
2017-01-02  6   7
2017-01-03  10  11
2017-01-04  14  15

1
df1.A  #按列名取出该列

执行结果:

1
2
3
4
5
6
7
2017-01-01     0
2017-01-02     4
2017-01-03     8
2017-01-04    12
2017-01-05    16
2017-01-06    20
Freq: D, Name: A, dtype: int32

1
df1.A > 6   #提取某一列进行判断

执行结果

1
2
3
4
5
6
7
2017-01-01    False
2017-01-02    False
2017-01-03     True
2017-01-04     True
2017-01-05     True
2017-01-06     True
Freq: D, Name: A, dtype: bool

1
df1[df1.A>6]    #将符合判断条件的数据再取出来,这一列数据中大于6的数据

执行结果:

1
2
3
4
5
A   B   C   D
2017-01-03  8   9   10  11
2017-01-04  12  13  14  15
2017-01-05  16  17  18  19
2017-01-06  20  21  22  23

3.pandas赋值及操作

1
2
import pandas as pd
import numpy as np

1
2
3
dates = np.arange(20170101,20170107)
df1 = pd.DataFrame(np.arange(24).reshape((6,4)),index=dates,columns=['A','B','C','D'])
df1

执行结果:

1
2
3
4
5
6
7
    A   B   C   D
20170101    0   1   2   3
20170102    4   5   6   7
20170103    8   9   10  11
20170104    12  13  14  15
20170105    16  17  18  19
20170106    20  21  22  23

1
df1.iloc[2,2]    #按照位置取出3行3列的数

执行结果:

1
10

1
2
df1.iloc[2,2] = 100   #将3行3列位置的数进行重新赋值
df1

执行结果:

1
2
3
4
5
6
7
    A   B   C   D
20170101    0   1   2   3
20170102    4   5   6   7
20170103    8   9   100 11
20170104    12  13  14  15
20170105    16  17  18  19
20170106    20  21  22  23

1
2
df1.loc[20170102,'B'] = 200   #使用标签取出对应位置的数据进行赋值
df1

执行结果:

1
2
3
4
5
6
7
    A   B   C   D
20170101    0   1   2   3
20170102    4   200 6   7
20170103    8   9   100 11
20170104    12  13  14  15
20170105    16  17  18  19
20170106    20  21  22  23

1
2
df1[df1.A>10] = 0    #使用条件判断对满足条件的数据进行赋值
df1

执行结果:

1
2
3
4
5
6
7
A   B   C   D
20170101    0   1   2   3
20170102    4   200 6   7
20170103    8   9   100 11
20170104    0   0   0   0
20170105    0   0   0   0
20170106    0   0   0   0

1
2
df1.A[df1.A==0] = 1
df1

执行结果:

1
2
3
4
5
6
7
    A   B   C   D
20170101    1   1   2   3
20170102    4   200 6   7
20170103    8   9   100 11
20170104    1   0   0   0
20170105    1   0   0   0
20170106    1   0   0   0

1
2
df1['E'] = 10 #添加一列,这一列的所有数据都是这个数
df1

执行结果:

1
2
3
4
5
6
7
    A   B   C   D   E
20170101    1   1   2   3   10
20170102    4   200 6   7   10
20170103    8   9   100 11  10
20170104    1   0   0   0   10
20170105    1   0   0   0   10
20170106    1   0   0   0   10

1
df1['F'] = pd.Series([1,2,3,4,5,6],index=dates)#添加一列,使用series添加

执行结果:

1
2
3
4
5
6
7
    A   B   C   D   E   F
20170101    1   1   2   3   10  1
20170102    4   200 6   7   10  2
20170103    8   9   100 11  10  3
20170104    1   0   0   0   10  4
20170105    1   0   0   0   10  5
20170106    1   0   0   0   10  6

1
2
df1.loc[20170107,['A','B','C']] = [1,2,3]   #添加一行中哪些列(按照标签)
df1

执行结果:没有赋值的列就是nan

1
2
3
4
5
6
7
8
A   B   C   D   E   F
20170101    1.0 1.0 2.0 3.0 10.0    1.0
20170102    4.0 200.0   6.0 7.0 10.0    2.0
20170103    8.0 9.0 100.0   11.0    10.0    3.0
20170104    1.0 0.0 0.0 0.0 10.0    4.0
20170105    1.0 0.0 0.0 0.0 10.0    5.0
20170106    1.0 0.0 0.0 0.0 10.0    6.0
20170107    1.0 2.0 3.0 NaN NaN NaN

1
2
3
4
s1 = pd.Series([1,2,3,4,5,6],index=['A','B','C','D','E','F'])
s1.name = 'S1'
df2 = df1.append(s1)  #使用append添加一行(series)
df2

执行结果:

1
2
3
4
5
6
7
8
9
A   B   C   D   E   F
20170101    1.0 1.0 2.0 3.0 10.0    1.0
20170102    4.0 200.0   6.0 7.0 10.0    2.0
20170103    8.0 9.0 100.0   11.0    10.0    3.0
20170104    1.0 0.0 0.0 0.0 10.0    4.0
20170105    1.0 0.0 0.0 0.0 10.0    5.0
20170106    1.0 0.0 0.0 0.0 10.0    6.0
20170107    1.0 2.0 3.0 NaN NaN NaN
S1  1.0 2.0 3.0 4.0 5.0 6.0

1
2
df1.insert(1,'G',df2['E'])#在第一列插入索引为G的df2中的E列
df1

执行结果:

1
2
3
4
5
6
7
8
                      A   G  B  C   D   E   F
20170101    1.0 10.0    1.0 2.0 3.0 10.0    1.0
20170102    4.0 10.0    200.0 6.0   7.0 10.0    2.0
20170103    8.0 10.0    9.0 100.0   11.0    10.0    3.0
20170104    1.0 10.0    0.0 0.0 0.0 10.0    4.0
20170105    1.0 10.0    0.0 0.0 0.0 10.0    5.0
20170106    1.0 10.0    0.0 0.0 0.0 10.0    6.0
20170107    1.0 NaN 2.0 3.0 NaN NaN NaN

1
2
3
g = df1.pop('G')#弹出G列
df1.insert(6,'G',g)#在最后插入
df1

执行结果:

1
2
3
4
5
6
7
8
    A   B   C   D   E   F   G
20170101    1.0 1.0 2.0 3.0 10.0    1.0 10.0
20170102    4.0 200.0   6.0 7.0 10.0    2.0 10.0
20170103    8.0 9.0 100.0   11.0    10.0    3.0 10.0
20170104    1.0 0.0 0.0 0.0 10.0    4.0 10.0
20170105    1.0 0.0 0.0 0.0 10.0    5.0 10.0
20170106    1.0 0.0 0.0 0.0 10.0    6.0 10.0
20170107    1.0 2.0 3.0 NaN NaN NaN NaN

1
2
del df1['G']#删除G列
df1

执行结果:

1
2
3
4
5
6
7
8
    A   B   C   D   E   F
20170101    1.0 1.0 2.0 3.0 10.0    1.0
20170102    4.0 200.0   6.0 7.0 10.0    2.0
20170103    8.0 9.0 100.0   11.0    10.0    3.0
20170104    1.0 0.0 0.0 0.0 10.0    4.0
20170105    1.0 0.0 0.0 0.0 10.0    5.0
20170106    1.0 0.0 0.0 0.0 10.0    6.0
20170107    1.0 2.0 3.0 NaN NaN NaN

1
2
df2 = df1.drop(['A','B'],axis=1)#删除AB列
df2

执行结果:

1
2
3
4
5
6
7
8
C   D   E   F
20170101    2.0 3.0 10.0    1.0
20170102    6.0 7.0 10.0    2.0
20170103    100.0   11.0    10.0    3.0
20170104    0.0 0.0 10.0    4.0
20170105    0.0 0.0 10.0    5.0
20170106    0.0 0.0 10.0    6.0
20170107    3.0 NaN NaN NaN

1
2
df2 = df1.drop([20170101,20170102],axis=0)#删除20170101,20170102行,axis=0代表删除的是行
df1

执行结果:

1
2
3
4
5
6
7
8
    A   B   C   D   E   F
20170101    1.0 1.0 2.0 3.0 10.0    1.0
20170102    4.0 200.0   6.0 7.0 10.0    2.0
20170103    8.0 9.0 100.0   11.0    10.0    3.0
20170104    1.0 0.0 0.0 0.0 10.0    4.0
20170105    1.0 0.0 0.0 0.0 10.0    5.0
20170106    1.0 0.0 0.0 0.0 10.0    6.0
20170107    1.0 2.0 3.0 NaN NaN NaN

1
df2

执行结果:

1
2
3
4
5
6
    A   B   C   D   E   F
20170103    8.0 9.0 100.0   11.0    10.0    3.0
20170104    1.0 0.0 0.0 0.0 10.0    4.0
20170105    1.0 0.0 0.0 0.0 10.0    5.0
20170106    1.0 0.0 0.0 0.0 10.0    6.0
20170107    1.0 2.0 3.0 NaN NaN NaN

4.pandas处理丢失数据

1
2
3
4
5
import pandas as pd
import numpy as np
dates = np.arange(20170101,20170105)
df1 = pd.DataFrame(np.arange(12).reshape((4,3)),index=dates,columns=['A','B','C'])
df1

执行结果: image.png

1
2
df2 = pd.DataFrame(df1,index=dates,columns=['A','B','C','D','E'])   #创建包含nan的dataframe
df2

执行: image.png

1
2
3
4
5
s1 = pd.Series([3,4,6],index=dates[:3])
s2 = pd.Series([32,5,2],index=dates[1:])
df2['D'] = s1
df2['E'] = s2
df2

image.png

1
df2.dropna(axis=0,how='any') #axis=[0,1] 0代表行，1代表列。how=['any','all'] any任意一个或多个 all全部,删除含有空值的行

image.png

1
df2.dropna(axis=1,how='any') #axis=[0,1] 0代表行，1代表列。how=['any','all'] any任意一个或多个 all全部

image.png

1
df2.fillna(value=0)#把空值赋值为0,fill填充的意思

image.png

1
df2.isnull()#查看空值

image.png

1
np.any(df2.isnull())#只要有一个或多个空值就会返回true

执行结果:

1
True

1
np.all(df2.isnull())#所有为空值才返回true

1
False

5.pandas读取及写入文件

1
import pandas as pd

1
2
file = pd.read_csv('people.csv',encoding='gbk')
file

image.png

1
2
file.iloc[2,0] = '深圳'  #赋值操作
file

image.png

1
file.to_csv('people2.csv')  #写入文件

6.pandas合并,concat

1
2
import pandas as pd
import numpy as np

1
2
3
4
5
6
df1 = pd.DataFrame(np.arange(12).reshape((3,4)),columns=['a','b','c','d'])
df2 = pd.DataFrame(np.arange(12,24).reshape((3,4)),columns=['a','b','c','d'])
df3 = pd.DataFrame(np.arange(24,36).reshape((3,4)),columns=['a','b','c','d'])
print(df1)
print(df2)
print(df3)

image.png

1
2
df4 = pd.concat([df1,df2,df3],axis=0)#纵向合并,concat连接
df4

image.png

1
2
df4 = pd.concat([df1,df2,df3],axis=0,ignore_index=True)#纵向合并，不考虑原来的index
df4

image.png

1
2
df5 = pd.concat([df1,df2,df3],axis=1)#横向合并
df5

image.png

1
2
3
4
df1 = pd.DataFrame(np.arange(12).reshape((3,4)),columns=['a','b','c','f'])
df2 = pd.DataFrame(np.arange(12,24).reshape((3,4)),columns=['a','c','d','e'])
print(df1)
print(df2)

image.png

1
2
df6 = pd.concat([df1,df2],join='outer',ignore_index=True)#合并两个表，缺少的部分填充NaN
df6

image.png

1
2
df7 = pd.concat([df1,df2],join='inner',ignore_index=True)#合并两个表，缺少的部分去掉
df7

image.png

1
2
3
4
df1 = pd.DataFrame(np.arange(12).reshape((3,4)),columns=['a','b','c','f'])
df2 = pd.DataFrame(np.arange(12,24).reshape((4,3)),columns=['a','c','d'])
print(df1)
print(df2)

image.png

1
2
df8 = pd.concat([df1,df2],axis=1,join_axes=[df1.index])#横向合并，index使用df1的index
df8

image.png

1
2
df8 = pd.concat([df1,df2],axis=1)#横向合并
df8

image.png

7.pandas合并merge

1
import pandas as pd

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
left = pd.DataFrame({'key':['K0','K1','K2','K3'],
                     'A':['A0','A1','A2','A3'],
                     'B':['B0','B1','B2','B3']})

right = pd.DataFrame({'key':['K0','K1','K2','K3'],
                     'C':['C0','C1','C2','C3'],
                     'D':['D0','D1','D2','D3']})

print(left)
print(right)

image.png

1
2
res = pd.merge(left,right,on='key')   #根据哪一列进行合并
res

image.png

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
left = pd.DataFrame({'key1':['K0','K0','K1','K2'],
                     'key2':['K0','K1','K0','K1'],
                     'A':['A0','A1','A2','A3'],
                     'B':['B0','B1','B2','B3']})

right = pd.DataFrame({'key1':['K0','K1','K1','K3'],
                      'key2':['K0','K0','K0','K0'],
                     'C':['C0','C1','C2','C3'],
                     'D':['D0','D1','D2','D3']})

print(left)
print(right)

image.png

1
2
3
#how = ['left','right','inner','outer']
res = pd.merge(left,right,on=['key1','key2'],how='outer')#how默认inner,带缺失值
res

image.png

1
2
3
#how = ['left','right','inner','outer']
res = pd.merge(left,right,on=['key1','key2'],how='inner')#how默认inner,不带缺失值
res

image.png

1
2
3
#how = ['left','right','inner','outer']
res = pd.merge(left,right,on=['key1','key2'],how='left')#how默认inner,使用left,按照左边表的索引来合并,由于2表中有两个k1,k0行.因此需要两行
res

image.png

1
2
3
#how = ['left','right','inner','outer']
res = pd.merge(left,right,on=['key1','key2'],how='outer',indicator=True)#显示merge信息
res

image.png

1
2
3
#how = ['left','right','inner','outer']
res = pd.merge(left,right,on=['key1','key2'],how='outer',indicator='indicator_column')#显示merge信息
res

image.png

1
2
3
4
5
6
7
8
left = pd.DataFrame({'A':['A0','A1','A2'],
                     'B':['B0','B1','B2']},
                     index = ['K0','K1','K2'])
right = pd.DataFrame({'C':['C0','C2','C3'],
                      'D':['D0','D2','D3']},
                      index=['K0','K2','K3'])
print(left)
print(right)

image.png

1
2
3
4
5
6
boys = pd.DataFrame({'k':['K0','K1','K2'],'age':[1,2,3]})

girls = pd.DataFrame({'k':['K0','K0','K3'],'age':[4,5,6]})

print(boys)
print(girls)

image.png

1
2
res = pd.merge(boys,girls,on='k',suffixes=['_boy','_girl'],how='outer')   #根据k进行合并,suffixs为了区别合并后的信息
res

image.png

8.pandas plot

1
2
3
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

1
2
3
4
5
6
data = pd.Series(np.random.randn(1000),index=np.arange(1000))
# print(data)
data = data.cumsum()  #累加操作
# print(data)    #索引加数据
data.plot()    #调用画图,索引为x,值为y
plt.show()

image.png

1
2
3
data = pd.DataFrame(np.random.randn(1000,4),index=np.arange(1000),columns=['A','B','C','D'])
data = data.cumsum()
print(data.head())

image.png

1
2
data.plot()
plt.show()

image.png

1
2
3
ax = data.plot.scatter(x='A',y='B',color='Blue',label='class 1')
data.plot.scatter(x='A',y='C',color='Green',label='class 2',ax=ax)
plt.show()

image.png

9 导出csv文件参数的设置

to_csv 方法和主要参数

to_csv 方法可以将Series和DataFrame对象输出成逗号分隔的csv文件

1
df.to_csv(path_or_buf, sep, na_rep, float_format, columns, header, index, index_label, mode, encoding, line_terminator, quoting, quotechar, doublequote, escapechar, chunksize, tupleize_cols, date_format)

参数说明 * path_or_buf ：要保存的路径及文件名

1
df.to_csv("./output.csv")    #保存在当前文件夹
  • sep ：分隔符，默认是","
1
df.to_csv("./output.csv", sep="\t")    #用Tab做分隔符
  • na_rep ：指定空值的输出方式，默认是空字符串 float_format：浮点数的输出格式，要用双引号括起来
1
df.to_csv("./output.csv", float_format="%.2f")      #浮点数格式表示方法
  • columns ：指定要输出的列，用列名列表表示，默认是None 注意文档有些地方写的参数名是cols，是不对的
1
df.to_csv("./output.csv", cols=["month","fruit"])
  • header ：是否输出列名，默认是True
1
df.to_csv("./output.csv",header=False)  #不输出列名
  • index ：是否输出索引，默认是True
1
df.to_csv("./output.csv",index=False)    #不输出索引
  • index_label ：索引列的列名，默认是None
1
df.to_csv("./output.csv",index_label="id")    #索引列的列名为id
  • encoding ：编码方式，Python2下默认“ascii”，Python3下默认“utf-8”
1
df.to_csv("./output.csv",encoding="utf-8")
  • line_terminator ：换行符，默认是'\n'
1
df.to_csv("./output.csv",line_terminator="\r\n")   #用dos下的换行符输出
  • quoting ：输出是否用引号，默认参数值为0，表示不加双引号，参数值为1，则每个字段都会加上引号，数值也会被当作字符串看待
1
df.to_csv("./output.csv",quoting=1)    #给输出的每个字段加上双引号
  • quotechar ：引用字符，当quoting=1可以指定引号字符为双引号"\""或单引号"\'"
1
df.to_csv("./output.csv",quoting=1,quotechar="\'")
  • chunksize ：一次写入csv文件的行数，当df表内容特别大时需要一点一点写入csv文件
1
df.to_csv("./output.csv",chunksize=100)
  • date_format ：日期输出格式

10pandas重置DataFrame或Series的索引index(重置索引)

当我们在清洗数据时往往会将带有空值的行删除，不论是DataFrame还是Series的index都将不再是连续的索引了，那么这个时候我们可以使用reset_index()方法来重置它们的索引，以便后续的操作。

具体例子1：

1
2
3
4
import pandas as pd
import numpy as np
df = pd.DataFrame(np.arange(20).reshape(5,4),index=[1,3,4,6,8])
print(df)

执行结果:

1
2
3
4
5
6
    0   1   2   3
1   0   1   2   3
3   4   5   6   7
4   8   9  10  11
6  12  13  14  15
8  16  17  18  19
  • 我们使用reset_index()来处理：

python print(df.reset_index())

执行结果:

python index 0 1 2 3 0 1 0 1 2 3 1 3 4 5 6 7 2 4 8 9 10 11 3 6 12 13 14 15 4 8 16 17 18 19

  • 可以看到此时获得了新的index列，而原来的index变成了我们的数据列，保留了下来。如果我们不想保留原来的index，直接使用重置后的索引，那么可以使用参数drop=True，默认值是False

python print(df.reset_index(drop=True))

执行结果:

1
2
3
4
5
6
    0   1   2   3
0   0   1   2   3
1   4   5   6   7
2   8   9  10  11
3  12  13  14  15
4  16  17  18  19

df.reset_index(drop=True)仅仅只是去除了原本的索引，但是并没有替换，df的索引会有部分缺失。若要直接替换，应该是这样df.reset_index(drop=True,inplace=True)

1
2
3
4
5
6
7
8
df
Out[4]: 
    0   1   2   3
1   0   1   2   3
3   4   5   6   7
4   8   9  10  11
6  12  13  14  15
8  16  17  18  19

在原来的列表上上重置索引

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
df.reset_index(drop=True,inplace=True)

df
Out[6]: 
    0   1   2   3
0   0   1   2   3
1   4   5   6   7
2   8   9  10  11
3  12  13  14  15
4  16  17  18  19