10分钟学习pandas

最新推荐文章于 2024-05-16 08:02:39 发布
最新推荐文章于 2024-05-16 08:02:39 发布
阅读量371 收藏 1
点赞数
分类专栏： Data Mining python 文章标签： pandas python 
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

s = pd.Series([1,3,5,np.nan,6,8])

s


0    1.0
1    3.0
2    5.0
3    NaN
4    6.0
5    8.0
dtype: float64

len(s)

6

s.describe()

count    5.000000
mean     4.600000
std      2.701851
min      1.000000
25%      3.000000
50%      5.000000
75%      6.000000
max      8.000000
dtype: float64

dates = pd.date_range('20190101',periods=6)

dates

DatetimeIndex(['2019-01-01', '2019-01-02', '2019-01-03', '2019-01-04',
               '2019-01-05', '2019-01-06'],
              dtype='datetime64[ns]', freq='D')

df = pd.DataFrame(np.random.randn(6,4),index=dates,columns=list('ABCD'))#randn返回一个标准正态分布

df

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
2019-01-010.3175750.3305490.055375-1.284453
2019-01-02-0.272201-0.7706180.2176581.349425
2019-01-030.842686-0.9733540.5961660.215889
2019-01-040.252868-0.778050-0.324255-1.378140
2019-01-050.2818761.143140-0.781625-1.245069
2019-01-06-0.468258-1.6823761.4940581.262588
df.describe()

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
count6.0000006.0000006.0000006.000000
mean0.159091-0.4551180.209563-0.179960
std0.4681601.0149140.7862121.293573
min-0.468258-1.682376-0.781625-1.378140
25%-0.140934-0.924528-0.229347-1.274607
50%0.267372-0.7743340.136517-0.514590
75%0.3086500.0552570.5015391.000913
max0.8426861.1431401.4940581.349425
df.shape

(6, 4)

df2 = pd.DataFrame({'A':1.,
                   'B':pd.Timestamp('20190101'),
                   'C':pd.Series(1,index=list(range(4)),dtype='float32'),
                   'D':np.array([3]*4,dtype='int32'),
                   'E':pd.Categorical(["test","train","test","train"]),
                   'F':'foo'})#foo被用作占位符的名字，用于实体举例，类似有foo,bar,baz

df2

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCDEF
01.02019-01-011.03testfoo
11.02019-01-011.03trainfoo
21.02019-01-011.03testfoo
31.02019-01-011.03trainfoo
df2.dtypes

A           float64
B    datetime64[ns]
C           float32
D             int32
E          category
F            object
dtype: object

df.head()

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
2019-01-010.3175750.3305490.055375-1.284453
2019-01-02-0.272201-0.7706180.2176581.349425
2019-01-030.842686-0.9733540.5961660.215889
2019-01-040.252868-0.778050-0.324255-1.378140
2019-01-050.2818761.143140-0.781625-1.245069
df.tail()

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
2019-01-02-0.272201-0.7706180.2176581.349425
2019-01-030.842686-0.9733540.5961660.215889
2019-01-040.252868-0.778050-0.324255-1.378140
2019-01-050.2818761.143140-0.781625-1.245069
2019-01-06-0.468258-1.6823761.4940581.262588
df.index

DatetimeIndex(['2019-01-01', '2019-01-02', '2019-01-03', '2019-01-04',
               '2019-01-05', '2019-01-06'],
              dtype='datetime64[ns]', freq='D')

df.columns

Index(['A', 'B', 'C', 'D'], dtype='object')


df.values

array([[ 0.31757454,  0.33054893,  0.05537508, -1.28445319],
       [-0.27220143, -0.77061807,  0.21765843,  1.34942538],
       [ 0.84268621, -0.97335385,  0.59616646,  0.21588867],
       [ 0.25286828, -0.77804969, -0.32425479, -1.37813964],
       [ 0.28187609,  1.14314031, -0.78162546, -1.24506887],
       [-0.4682577 , -1.68237556,  1.49405812,  1.26258772]])


df.describe()#以列统计

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
count6.0000006.0000006.0000006.000000
mean0.159091-0.4551180.209563-0.179960
std0.4681601.0149140.7862121.293573
min-0.468258-1.682376-0.781625-1.378140
25%-0.140934-0.924528-0.229347-1.274607
50%0.267372-0.7743340.136517-0.514590
75%0.3086500.0552570.5015391.000913
max0.8426861.1431401.4940581.349425
df2.describe(include='all')

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCDEF
count4.044.04.044
uniqueNaN1NaNNaN21
topNaN2019-01-01 00:00:00NaNNaNtrainfoo
freqNaN4NaNNaN24
firstNaN2019-01-01 00:00:00NaNNaNNaNNaN
lastNaN2019-01-01 00:00:00NaNNaNNaNNaN
mean1.0NaN1.03.0NaNNaN
std0.0NaN0.00.0NaNNaN
min1.0NaN1.03.0NaNNaN
25%1.0NaN1.03.0NaNNaN
50%1.0NaN1.03.0NaNNaN
75%1.0NaN1.03.0NaNNaN
max1.0NaN1.03.0NaNNaN
df.T

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
2019-01-01 00:00:002019-01-02 00:00:002019-01-03 00:00:002019-01-04 00:00:002019-01-05 00:00:002019-01-06 00:00:00
A0.317575-0.2722010.8426860.2528680.281876-0.468258
B0.330549-0.770618-0.973354-0.7780501.143140-1.682376
C0.0553750.2176580.596166-0.324255-0.7816251.494058
D-1.2844531.3494250.215889-1.378140-1.2450691.262588
df.sort_index(axis=1,ascending=True)#以行为序，降序

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
2019-01-010.3175750.3305490.055375-1.284453
2019-01-02-0.272201-0.7706180.2176581.349425
2019-01-030.842686-0.9733540.5961660.215889
2019-01-040.252868-0.778050-0.324255-1.378140
2019-01-050.2818761.143140-0.781625-1.245069
2019-01-06-0.468258-1.6823761.4940581.262588
df.sort_values(by='B')#以B的值升序排列，以列

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
2019-01-06-0.468258-1.6823761.4940581.262588
2019-01-030.842686-0.9733540.5961660.215889
2019-01-040.252868-0.778050-0.324255-1.378140
2019-01-02-0.272201-0.7706180.2176581.349425
2019-01-010.3175750.3305490.055375-1.284453
2019-01-050.2818761.143140-0.781625-1.245069
df['A']#df.A 选择列

2019-01-01    0.317575
2019-01-02   -0.272201
2019-01-03    0.842686
2019-01-04    0.252868
2019-01-05    0.281876
2019-01-06   -0.468258
Freq: D, Name: A, dtype: float64


df[0:3]#只显示0,1,2,此时切割行

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
2019-01-010.3175750.3305490.055375-1.284453
2019-01-02-0.272201-0.7706180.2176581.349425
2019-01-030.842686-0.9733540.5961660.215889
df['20190101':'20190103']#此时包含头尾

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
2019-01-010.3175750.3305490.055375-1.284453
2019-01-02-0.272201-0.7706180.2176581.349425
2019-01-030.842686-0.9733540.5961660.215889
df.loc[dates[0]]#行

A    0.317575
B    0.330549
C    0.055375
D   -1.284453
Name: 2019-01-01 00:00:00, dtype: float64


dates[0]

Timestamp('2019-01-01 00:00:00', freq='D')


df.loc[:,['A','B']]#列

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
AB
2019-01-010.3175750.330549
2019-01-02-0.272201-0.770618
2019-01-030.842686-0.973354
2019-01-040.252868-0.778050
2019-01-050.2818761.143140
2019-01-06-0.468258-1.682376
df.loc['20190104',['A','B']].shape#返回形状变小

(2,)


#loc,at也可用来获得标量值
df.loc[dates[0],'A']

0.3175745430028141


df.at[dates[0],'A']

0.3175745430028141


df.at[dates[0],'A']#其中at/iat通过便签获取某个数值的具体位置
df.iat[1,2]#iat可用行号

0.21765843113729494


df.iloc[3]

A    0.252868
B   -0.778050
C   -0.324255
D   -1.378140
Name: 2019-01-04 00:00:00, dtype: float64


df.iloc[3,1]

-0.7780496857268889


df.iloc[[1,2,4],[0,2]]

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
AC
2019-01-02-0.2722010.217658
2019-01-030.8426860.596166
2019-01-050.281876-0.781625
df.iloc[1:3,:]

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
2019-01-02-0.272201-0.7706180.2176581.349425
2019-01-030.842686-0.9733540.5961660.215889
df.iloc[1:3]

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
2019-01-02-0.272201-0.7706180.2176581.349425
2019-01-030.842686-0.9733540.5961660.215889
df[df.A>0]

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
2019-01-010.3175750.3305490.055375-1.284453
2019-01-030.842686-0.9733540.5961660.215889
2019-01-040.252868-0.778050-0.324255-1.378140
2019-01-050.2818761.143140-0.781625-1.245069
df[df>0]

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
2019-01-010.3175750.3305490.055375NaN
2019-01-02NaNNaN0.2176581.349425
2019-01-030.842686NaN0.5961660.215889
2019-01-040.252868NaNNaNNaN
2019-01-050.2818761.143140NaNNaN
2019-01-06NaNNaN1.4940581.262588
df3 = df.copy()

df3['E']=['one','two','three','four','three','one']

df3

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCDE
2019-01-010.3175750.3305490.055375-1.284453one
2019-01-02-0.272201-0.7706180.2176581.349425two
2019-01-030.842686-0.9733540.5961660.215889three
2019-01-040.252868-0.778050-0.324255-1.378140four
2019-01-050.2818761.143140-0.781625-1.245069three
2019-01-06-0.468258-1.6823761.4940581.262588one
df3[df3['E'].isin(['one','two'])]

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCDE
2019-01-010.3175750.3305490.055375-1.284453one
2019-01-02-0.272201-0.7706180.2176581.349425two
2019-01-06-0.468258-1.6823761.4940581.262588one
#设置新列时会自动按索引对齐
s1= pd.Series([1,2,3,4,5,6],index=pd.date_range('20190102',periods=6))

s1

2019-01-02    1
2019-01-03    2
2019-01-04    3
2019-01-05    4
2019-01-06    5
2019-01-07    6
Freq: D, dtype: int64


df['F']=s1

df

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCDF
2019-01-010.3175750.3305490.055375-1.284453NaN
2019-01-02-0.272201-0.7706180.2176581.3494251.0
2019-01-030.842686-0.9733540.5961660.2158892.0
2019-01-040.252868-0.778050-0.324255-1.3781403.0
2019-01-050.2818761.143140-0.781625-1.2450694.0
2019-01-06-0.468258-1.6823761.4940581.2625885.0
df.at[dates[0],'A']=0

df.iat[0,1]=0

df.loc[:,'D']=np.array([5]*len(df))

df

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCDF
2019-01-010.0000000.0000000.0553755NaN
2019-01-02-0.272201-0.7706180.21765851.0
2019-01-030.842686-0.9733540.59616652.0
2019-01-040.252868-0.778050-0.32425553.0
2019-01-050.2818761.143140-0.78162554.0
2019-01-06-0.468258-1.6823761.49405855.0
df4 = df.copy()

df4[df4>0]=-df4

df4

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCDF
2019-01-010.0000000.000000-0.055375-5NaN
2019-01-02-0.272201-0.770618-0.217658-5-1.0
2019-01-03-0.842686-0.973354-0.596166-5-2.0
2019-01-04-0.252868-0.778050-0.324255-5-3.0
2019-01-05-0.281876-1.143140-0.781625-5-4.0
2019-01-06-0.468258-1.682376-1.494058-5-5.0
df1 = df.reindex(index=dates[0:4],columns=list(df.columns)+['E'])

df1.loc[dates[0]:dates[1],'E']=1

df1

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCDFE
2019-01-010.0000000.0000000.0553755NaN1.0
2019-01-02-0.272201-0.7706180.21765851.01.0
2019-01-030.842686-0.9733540.59616652.0NaN
2019-01-040.252868-0.778050-0.32425553.0NaN
df1.dropna(how='any')

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCDFE
2019-01-02-0.272201-0.7706180.21765851.01.0
df1.fillna(value=5)

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCDFE
2019-01-010.0000000.0000000.05537555.01.0
2019-01-02-0.272201-0.7706180.21765851.01.0
2019-01-030.842686-0.9733540.59616652.05.0
2019-01-040.252868-0.778050-0.32425553.05.0
pd.isna(df1)

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCDFE
2019-01-01FalseFalseFalseFalseTrueFalse
2019-01-02FalseFalseFalseFalseFalseFalse
2019-01-03FalseFalseFalseFalseFalseTrue
2019-01-04FalseFalseFalseFalseFalseTrue
df.mean()#列

A    0.106162
B   -0.510209
C    0.209563
D    5.000000
F    3.000000
dtype: float64


df.mean(1)#行

2019-01-01    1.263844
2019-01-02    1.034968
2019-01-03    1.493100
2019-01-04    1.430113
2019-01-05    1.928678
2019-01-06    1.868685
Freq: D, dtype: float64


#操作具有不同维度且需要对齐的对象。此外，pandas会自动沿着指定的维度广播。
s = pd.Series([1,3,5,np.nan,6,8],index=dates).shift(2)

s

2019-01-01    NaN
2019-01-02    NaN
2019-01-03    1.0
2019-01-04    3.0
2019-01-05    5.0
2019-01-06    NaN
Freq: D, dtype: float64


df.sub(s,axis='index')#每行每列按照s值进行减法操作

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCDF
2019-01-01NaNNaNNaNNaNNaN
2019-01-02NaNNaNNaNNaNNaN
2019-01-03-0.157314-1.973354-0.4038344.01.0
2019-01-04-2.747132-3.778050-3.3242552.00.0
2019-01-05-4.718124-3.856860-5.7816250.0-1.0
2019-01-06NaNNaNNaNNaNNaN
df

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCDF
2019-01-010.0000000.0000000.0553755NaN
2019-01-02-0.272201-0.7706180.21765851.0
2019-01-030.842686-0.9733540.59616652.0
2019-01-040.252868-0.778050-0.32425553.0
2019-01-050.2818761.143140-0.78162554.0
2019-01-06-0.468258-1.6823761.49405855.0
df.apply(np.cumsum)#从第一行加到最后一行

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCDF
2019-01-010.0000000.0000000.0553755NaN
2019-01-02-0.272201-0.7706180.273034101.0
2019-01-030.570485-1.7439720.869200153.0
2019-01-040.823353-2.5220220.544945206.0
2019-01-051.105229-1.378881-0.2366802510.0
2019-01-060.636971-3.0612571.2573783015.0
df.apply(lambda x:x.max()-x.min())#以列

A    1.310944
B    2.825516
C    2.275684
D    0.000000
F    4.000000
dtype: float64


df['F'].value_counts()

5.0    1
4.0    1
3.0    1
2.0    1
1.0    1
Name: F, dtype: int64


s = pd.Series(np.random.randint(0,7,size=10))

s

0    1
1    1
2    1
3    2
4    3
5    2
6    6
7    0
8    4
9    6
dtype: int32


s.dtype

dtype('int32')


s.shape

(10,)


s.value_counts()

1    3
6    2
2    2
4    1
3    1
0    1
dtype: int64


s = pd.Series(['A', 'B', 'C', 'Aaba', 'Baca', np.nan, 'CABA', 'dog', 'cat'])

s.dtype

dtype('O')


s.str

<pandas.core.strings.StringMethods at 0x22f0313d160>


s.str.lower()

0       a
1       b
2       c
3    aaba
4    baca
5     NaN
6    caba
7     dog
8     cat
dtype: object


#数据合并
df = pd.DataFrame(np.random.randn(10,4))

df

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
0123
0-0.936960-0.2106501.8870690.128005
10.6576600.254801-0.092190-1.957160
2-0.920141-2.2599640.0839650.371865
3-0.7554030.6334260.090949-0.626964
40.1700521.164537-1.1937061.391785
50.084560-1.5138140.0690320.099851
6-0.6836630.031873-0.0629980.523253
7-0.9265940.125286-1.894089-0.449402
80.6107220.3291560.025149-2.673445
90.3366731.205792-1.3461790.214389
pieces = [df[:3],df[3:7],df[7:]]

pieces

[          0         1         2         3
 0 -0.936960 -0.210650  1.887069  0.128005
 1  0.657660  0.254801 -0.092190 -1.957160
 2 -0.920141 -2.259964  0.083965  0.371865,
           0         1         2         3
 3 -0.755403  0.633426  0.090949 -0.626964
 4  0.170052  1.164537 -1.193706  1.391785
 5  0.084560 -1.513814  0.069032  0.099851
 6 -0.683663  0.031873 -0.062998  0.523253,
           0         1         2         3
 7 -0.926594  0.125286 -1.894089 -0.449402
 8  0.610722  0.329156  0.025149 -2.673445
 9  0.336673  1.205792 -1.346179  0.214389]


pd.concat(pieces)

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
0123
0-0.936960-0.2106501.8870690.128005
10.6576600.254801-0.092190-1.957160
2-0.920141-2.2599640.0839650.371865
3-0.7554030.6334260.090949-0.626964
40.1700521.164537-1.1937061.391785
50.084560-1.5138140.0690320.099851
6-0.6836630.031873-0.0629980.523253
7-0.9265940.125286-1.894089-0.449402
80.6107220.3291560.025149-2.673445
90.3366731.205792-1.3461790.214389

concat

  1. axis=1列合并；axis=0行合并（default）
  2. join=‘inner’只返回合并的共同对象；join=‘outer’返回合并后的所有列
  3. ignore_index=True会自动连续生成索引值；ignore_index=False则不改变原有的索引值
  4. verify_integrity会检查当前合并是否重复索引
  5. keys会导致生成多维数组
  6. name会重新命名列名。
pd.concat(pieces,axis=1)

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
012301230123
0-0.936960-0.2106501.8870690.128005NaNNaNNaNNaNNaNNaNNaNNaN
10.6576600.254801-0.092190-1.957160NaNNaNNaNNaNNaNNaNNaNNaN
2-0.920141-2.2599640.0839650.371865NaNNaNNaNNaNNaNNaNNaNNaN
3NaNNaNNaNNaN-0.7554030.6334260.090949-0.626964NaNNaNNaNNaN
4NaNNaNNaNNaN0.1700521.164537-1.1937061.391785NaNNaNNaNNaN
5NaNNaNNaNNaN0.084560-1.5138140.0690320.099851NaNNaNNaNNaN
6NaNNaNNaNNaN-0.6836630.031873-0.0629980.523253NaNNaNNaNNaN
7NaNNaNNaNNaNNaNNaNNaNNaN-0.9265940.125286-1.894089-0.449402
8NaNNaNNaNNaNNaNNaNNaNNaN0.6107220.3291560.025149-2.673445
9NaNNaNNaNNaNNaNNaNNaNNaN0.3366731.205792-1.3461790.214389
pd.concat(pieces,join='outer',axis=1)

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
012301230123
0-0.936960-0.2106501.8870690.128005NaNNaNNaNNaNNaNNaNNaNNaN
10.6576600.254801-0.092190-1.957160NaNNaNNaNNaNNaNNaNNaNNaN
2-0.920141-2.2599640.0839650.371865NaNNaNNaNNaNNaNNaNNaNNaN
3NaNNaNNaNNaN-0.7554030.6334260.090949-0.626964NaNNaNNaNNaN
4NaNNaNNaNNaN0.1700521.164537-1.1937061.391785NaNNaNNaNNaN
5NaNNaNNaNNaN0.084560-1.5138140.0690320.099851NaNNaNNaNNaN
6NaNNaNNaNNaN-0.6836630.031873-0.0629980.523253NaNNaNNaNNaN
7NaNNaNNaNNaNNaNNaNNaNNaN-0.9265940.125286-1.894089-0.449402
8NaNNaNNaNNaNNaNNaNNaNNaN0.6107220.3291560.025149-2.673445
9NaNNaNNaNNaNNaNNaNNaNNaN0.3366731.205792-1.3461790.214389
#sql格式合并
left = pd.DataFrame({'key':['foo','bar'],'lval':[1,2]})
right = pd.DataFrame({'key':['foo','bar'],'rval':[4,5]})

left

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
keylval
0foo1
1bar2
right

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
keyrval
0foo4
1bar5
pd.merge(left,right,on='key')

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
keylvalrval
0foo14
1bar25
left = pd.DataFrame({'key':['foo','foo'],'lval':[1,2]})
right = pd.DataFrame({'key':['foo','foo'],'rval':[4,5]})

left

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
keylval
0foo1
1foo2
right

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
keyrval
0foo4
1foo5
pd.merge(left,right,on='key')#以某列为键值进行合并

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
keylvalrval
0foo14
1foo15
2foo24
3foo25
#将行append到dataframe中
df = pd.DataFrame(np.random.randn(8,4),columns=list('ABCD'))

df

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
00.862756-1.4366920.3673961.033803
1-0.732357-0.350199-0.116083-2.435210
20.3165820.4686160.433001-0.443120
3-0.189939-2.4371370.126893-2.273711
40.913514-0.752727-1.6511401.156839
5-0.3145811.2965850.579130-0.871556
60.3614730.687854-1.0446020.233138
70.0451992.176608-0.258569-1.018576
s = df.iloc[3]

df.append(s,ignore_index=False)

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
00.862756-1.4366920.3673961.033803
1-0.732357-0.350199-0.116083-2.435210
20.3165820.4686160.433001-0.443120
3-0.189939-2.4371370.126893-2.273711
40.913514-0.752727-1.6511401.156839
5-0.3145811.2965850.579130-0.871556
60.3614730.687854-1.0446020.233138
70.0451992.176608-0.258569-1.018576
3-0.189939-2.4371370.126893-2.273711

“分组依据”是指涉及以下一个或多个步骤的过程:
根据某些标准将数据分成组
独立地将函数应用于每个组
将结果组合成数据结构

df = pd.DataFrame({'A' : ['foo', 'bar', 'foo', 'bar',
   ....:                           'foo', 'bar', 'foo', 'foo'],
   ....:                    'B' : ['one', 'one', 'two', 'three',
   ....:                           'two', 'two', 'one', 'three'],
   ....:                    'C' : np.random.randn(8),
   ....:                    'D' : np.random.randn(8)})

df

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
0fooone0.982988-0.386029
1barone1.632482-0.327520
2footwo0.149619-0.138297
3barthree-1.4803971.105690
4footwo0.647044-1.097276
5bartwo-0.6755960.250176
6fooone0.4373091.031742
7foothree0.4346591.197695
df.groupby('A').sum()

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
CD
A
bar-0.5235121.028346
foo2.6516180.607834
df.groupby('A').max()

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
BCD
A
bartwo1.6324821.105690
footwo0.9829881.197695
df.groupby(['A','B']).sum()

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
CD
AB
barone1.632482-0.327520
three-1.4803971.105690
two-0.6755960.250176
fooone1.4202970.645713
three0.4346591.197695
two0.796663-1.235574
tuples = list(zip(*[['bar', 'bar', 'baz', 'baz',
   ....:                      'foo', 'foo', 'qux', 'qux'],
   ....:                     ['one', 'two', 'one', 'two',
   ....:                      'one', 'two', 'one', 'two']]))

tuples

[('bar', 'one'),
 ('bar', 'two'),
 ('baz', 'one'),
 ('baz', 'two'),
 ('foo', 'one'),
 ('foo', 'two'),
 ('qux', 'one'),
 ('qux', 'two')]

zip() 函数用于将可迭代的对象作为参数，将对象中对应的元素打包成一个个元组，然后返回由这些元组组成的列表,在python3中用于减少内存

index = pd.MultiIndex.from_tuples(tuples,names=['first','second'])

index

MultiIndex(levels=[['bar', 'baz', 'foo', 'qux'], ['one', 'two']],
           labels=[[0, 0, 1, 1, 2, 2, 3, 3], [0, 1, 0, 1, 0, 1, 0, 1]],
           names=['first', 'second'])


df = pd.DataFrame(np.random.randn(8,2),index=index,columns=['A','B'])

df

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
AB
firstsecond
barone-0.7523640.982241
two-0.6266901.358099
bazone0.342360-0.618870
two-0.399448-0.500175
fooone-0.746328-0.244752
two1.556458-1.340098
quxone-0.569393-0.340625
two0.5137140.151477
#stack压缩列
stacked = df.stack()

stacked

first  second   
bar    one     A   -0.752364
               B    0.982241
       two     A   -0.626690
               B    1.358099
baz    one     A    0.342360
               B   -0.618870
       two     A   -0.399448
               B   -0.500175
foo    one     A   -0.746328
               B   -0.244752
       two     A    1.556458
               B   -1.340098
qux    one     A   -0.569393
               B   -0.340625
       two     A    0.513714
               B    0.151477
dtype: float64


stacked.unstack()

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
AB
firstsecond
barone-0.7523640.982241
two-0.6266901.358099
bazone0.342360-0.618870
two-0.399448-0.500175
fooone-0.746328-0.244752
two1.556458-1.340098
quxone-0.569393-0.340625
two0.5137140.151477
stacked.unstack(1)

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
secondonetwo
first
barA-0.752364-0.626690
B0.9822411.358099
bazA0.342360-0.399448
B-0.618870-0.500175
fooA-0.7463281.556458
B-0.244752-1.340098
quxA-0.5693930.513714
B-0.3406250.151477
stacked.unstack(0)

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
firstbarbazfooqux
second
oneA-0.7523640.342360-0.746328-0.569393
B0.982241-0.618870-0.244752-0.340625
twoA-0.626690-0.3994481.5564580.513714
B1.358099-0.500175-1.3400980.151477
df = pd.DataFrame({'A' : ['one', 'one', 'two', 'three'] * 3,
   .....:                    'B' : ['A', 'B', 'C'] * 4,
   .....:                    'C' : ['foo', 'foo', 'foo', 'bar', 'bar', 'bar'] * 2,
   .....:                    'D' : np.random.randn(12),
   .....:                    'E' : np.random.randn(12)})

df

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCDE
0oneAfoo0.346892-0.615216
1oneBfoo0.8086810.161598
2twoCfoo-1.324783-0.088082
3threeAbar-0.227795-0.180022
4oneBbar1.130028-0.361439
5oneCbar0.510629-1.466063
6twoAfoo0.3795030.008279
7threeBfoo0.921087-0.148614
8oneCfoo0.332222-0.127428
9oneAbar-0.784876-0.736117
10twoBbar-0.7939571.705022
11threeCbar-0.8984851.038166
pd.pivot_table(df,values='D',index=['A','B'],columns=['C'])

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
Cbarfoo
AB
oneA-0.7848760.346892
B1.1300280.808681
C0.5106290.332222
threeA-0.227795NaN
BNaN0.921087
C-0.898485NaN
twoANaN0.379503
B-0.793957NaN
CNaN-1.324783
rng = pd.date_range('1/1/2012', periods=100, freq='S')

rng

DatetimeIndex(['2012-01-01 00:00:00', '2012-01-01 00:00:01',
               '2012-01-01 00:00:02', '2012-01-01 00:00:03',
               '2012-01-01 00:00:04', '2012-01-01 00:00:05',
               '2012-01-01 00:00:06', '2012-01-01 00:00:07',
               '2012-01-01 00:00:08', '2012-01-01 00:00:09',
               '2012-01-01 00:00:10', '2012-01-01 00:00:11',
               '2012-01-01 00:00:12', '2012-01-01 00:00:13',
               '2012-01-01 00:00:14', '2012-01-01 00:00:15',
               '2012-01-01 00:00:16', '2012-01-01 00:00:17',
               '2012-01-01 00:00:18', '2012-01-01 00:00:19',
               '2012-01-01 00:00:20', '2012-01-01 00:00:21',
               '2012-01-01 00:00:22', '2012-01-01 00:00:23',
               '2012-01-01 00:00:24', '2012-01-01 00:00:25',
               '2012-01-01 00:00:26', '2012-01-01 00:00:27',
               '2012-01-01 00:00:28', '2012-01-01 00:00:29',
               '2012-01-01 00:00:30', '2012-01-01 00:00:31',
               '2012-01-01 00:00:32', '2012-01-01 00:00:33',
               '2012-01-01 00:00:34', '2012-01-01 00:00:35',
               '2012-01-01 00:00:36', '2012-01-01 00:00:37',
               '2012-01-01 00:00:38', '2012-01-01 00:00:39',
               '2012-01-01 00:00:40', '2012-01-01 00:00:41',
               '2012-01-01 00:00:42', '2012-01-01 00:00:43',
               '2012-01-01 00:00:44', '2012-01-01 00:00:45',
               '2012-01-01 00:00:46', '2012-01-01 00:00:47',
               '2012-01-01 00:00:48', '2012-01-01 00:00:49',
               '2012-01-01 00:00:50', '2012-01-01 00:00:51',
               '2012-01-01 00:00:52', '2012-01-01 00:00:53',
               '2012-01-01 00:00:54', '2012-01-01 00:00:55',
               '2012-01-01 00:00:56', '2012-01-01 00:00:57',
               '2012-01-01 00:00:58', '2012-01-01 00:00:59',
               '2012-01-01 00:01:00', '2012-01-01 00:01:01',
               '2012-01-01 00:01:02', '2012-01-01 00:01:03',
               '2012-01-01 00:01:04', '2012-01-01 00:01:05',
               '2012-01-01 00:01:06', '2012-01-01 00:01:07',
               '2012-01-01 00:01:08', '2012-01-01 00:01:09',
               '2012-01-01 00:01:10', '2012-01-01 00:01:11',
               '2012-01-01 00:01:12', '2012-01-01 00:01:13',
               '2012-01-01 00:01:14', '2012-01-01 00:01:15',
               '2012-01-01 00:01:16', '2012-01-01 00:01:17',
               '2012-01-01 00:01:18', '2012-01-01 00:01:19',
               '2012-01-01 00:01:20', '2012-01-01 00:01:21',
               '2012-01-01 00:01:22', '2012-01-01 00:01:23',
               '2012-01-01 00:01:24', '2012-01-01 00:01:25',
               '2012-01-01 00:01:26', '2012-01-01 00:01:27',
               '2012-01-01 00:01:28', '2012-01-01 00:01:29',
               '2012-01-01 00:01:30', '2012-01-01 00:01:31',
               '2012-01-01 00:01:32', '2012-01-01 00:01:33',
               '2012-01-01 00:01:34', '2012-01-01 00:01:35',
               '2012-01-01 00:01:36', '2012-01-01 00:01:37',
               '2012-01-01 00:01:38', '2012-01-01 00:01:39'],
              dtype='datetime64[ns]', freq='S')


ts = pd.Series(np.random.randint(0, 500, len(rng)), index=rng)

ts

2012-01-01 00:00:00    225
2012-01-01 00:00:01    354
2012-01-01 00:00:02    438
2012-01-01 00:00:03    440
2012-01-01 00:00:04      9
2012-01-01 00:00:05    179
2012-01-01 00:00:06    396
2012-01-01 00:00:07    200
2012-01-01 00:00:08    413
2012-01-01 00:00:09    490
2012-01-01 00:00:10     37
2012-01-01 00:00:11     57
2012-01-01 00:00:12     33
2012-01-01 00:00:13    388
2012-01-01 00:00:14     44
2012-01-01 00:00:15     95
2012-01-01 00:00:16      8
2012-01-01 00:00:17      1
2012-01-01 00:00:18    307
2012-01-01 00:00:19    332
2012-01-01 00:00:20     20
2012-01-01 00:00:21     84
2012-01-01 00:00:22    309
2012-01-01 00:00:23    308
2012-01-01 00:00:24     67
2012-01-01 00:00:25    245
2012-01-01 00:00:26    180
2012-01-01 00:00:27      9
2012-01-01 00:00:28    126
2012-01-01 00:00:29    232
                      ... 
2012-01-01 00:01:10    409
2012-01-01 00:01:11    355
2012-01-01 00:01:12     70
2012-01-01 00:01:13    266
2012-01-01 00:01:14    118
2012-01-01 00:01:15    325
2012-01-01 00:01:16    214
2012-01-01 00:01:17      3
2012-01-01 00:01:18    143
2012-01-01 00:01:19     28
2012-01-01 00:01:20     56
2012-01-01 00:01:21    120
2012-01-01 00:01:22     99
2012-01-01 00:01:23    102
2012-01-01 00:01:24     71
2012-01-01 00:01:25    464
2012-01-01 00:01:26    489
2012-01-01 00:01:27    404
2012-01-01 00:01:28    356
2012-01-01 00:01:29    197
2012-01-01 00:01:30    390
2012-01-01 00:01:31    345
2012-01-01 00:01:32    115
2012-01-01 00:01:33    377
2012-01-01 00:01:34    388
2012-01-01 00:01:35     39
2012-01-01 00:01:36    406
2012-01-01 00:01:37    408
2012-01-01 00:01:38    410
2012-01-01 00:01:39    256
Freq: S, Length: 100, dtype: int32


rng = pd.date_range('3/6/2012 00:00', periods=5, freq='D')

rng

DatetimeIndex(['2012-03-06', '2012-03-07', '2012-03-08', '2012-03-09',
               '2012-03-10'],
              dtype='datetime64[ns]', freq='D')


ts_utc = ts.tz_localize('UTC')


ts_utc

2012-01-01 00:00:00+00:00    225
2012-01-01 00:00:01+00:00    354
2012-01-01 00:00:02+00:00    438
2012-01-01 00:00:03+00:00    440
2012-01-01 00:00:04+00:00      9
2012-01-01 00:00:05+00:00    179
2012-01-01 00:00:06+00:00    396
2012-01-01 00:00:07+00:00    200
2012-01-01 00:00:08+00:00    413
2012-01-01 00:00:09+00:00    490
2012-01-01 00:00:10+00:00     37
2012-01-01 00:00:11+00:00     57
2012-01-01 00:00:12+00:00     33
2012-01-01 00:00:13+00:00    388
2012-01-01 00:00:14+00:00     44
2012-01-01 00:00:15+00:00     95
2012-01-01 00:00:16+00:00      8
2012-01-01 00:00:17+00:00      1
2012-01-01 00:00:18+00:00    307
2012-01-01 00:00:19+00:00    332
2012-01-01 00:00:20+00:00     20
2012-01-01 00:00:21+00:00     84
2012-01-01 00:00:22+00:00    309
2012-01-01 00:00:23+00:00    308
2012-01-01 00:00:24+00:00     67
2012-01-01 00:00:25+00:00    245
2012-01-01 00:00:26+00:00    180
2012-01-01 00:00:27+00:00      9
2012-01-01 00:00:28+00:00    126
2012-01-01 00:00:29+00:00    232
                            ... 
2012-01-01 00:01:10+00:00    409
2012-01-01 00:01:11+00:00    355
2012-01-01 00:01:12+00:00     70
2012-01-01 00:01:13+00:00    266
2012-01-01 00:01:14+00:00    118
2012-01-01 00:01:15+00:00    325
2012-01-01 00:01:16+00:00    214
2012-01-01 00:01:17+00:00      3
2012-01-01 00:01:18+00:00    143
2012-01-01 00:01:19+00:00     28
2012-01-01 00:01:20+00:00     56
2012-01-01 00:01:21+00:00    120
2012-01-01 00:01:22+00:00     99
2012-01-01 00:01:23+00:00    102
2012-01-01 00:01:24+00:00     71
2012-01-01 00:01:25+00:00    464
2012-01-01 00:01:26+00:00    489
2012-01-01 00:01:27+00:00    404
2012-01-01 00:01:28+00:00    356
2012-01-01 00:01:29+00:00    197
2012-01-01 00:01:30+00:00    390
2012-01-01 00:01:31+00:00    345
2012-01-01 00:01:32+00:00    115
2012-01-01 00:01:33+00:00    377
2012-01-01 00:01:34+00:00    388
2012-01-01 00:01:35+00:00     39
2012-01-01 00:01:36+00:00    406
2012-01-01 00:01:37+00:00    408
2012-01-01 00:01:38+00:00    410
2012-01-01 00:01:39+00:00    256
Freq: S, Length: 100, dtype: int32


ts_utc.tz_convert('US/Eastern')

2011-12-31 19:00:00-05:00    225
2011-12-31 19:00:01-05:00    354
2011-12-31 19:00:02-05:00    438
2011-12-31 19:00:03-05:00    440
2011-12-31 19:00:04-05:00      9
2011-12-31 19:00:05-05:00    179
2011-12-31 19:00:06-05:00    396
2011-12-31 19:00:07-05:00    200
2011-12-31 19:00:08-05:00    413
2011-12-31 19:00:09-05:00    490
2011-12-31 19:00:10-05:00     37
2011-12-31 19:00:11-05:00     57
2011-12-31 19:00:12-05:00     33
2011-12-31 19:00:13-05:00    388
2011-12-31 19:00:14-05:00     44
2011-12-31 19:00:15-05:00     95
2011-12-31 19:00:16-05:00      8
2011-12-31 19:00:17-05:00      1
2011-12-31 19:00:18-05:00    307
2011-12-31 19:00:19-05:00    332
2011-12-31 19:00:20-05:00     20
2011-12-31 19:00:21-05:00     84
2011-12-31 19:00:22-05:00    309
2011-12-31 19:00:23-05:00    308
2011-12-31 19:00:24-05:00     67
2011-12-31 19:00:25-05:00    245
2011-12-31 19:00:26-05:00    180
2011-12-31 19:00:27-05:00      9
2011-12-31 19:00:28-05:00    126
2011-12-31 19:00:29-05:00    232
                            ... 
2011-12-31 19:01:10-05:00    409
2011-12-31 19:01:11-05:00    355
2011-12-31 19:01:12-05:00     70
2011-12-31 19:01:13-05:00    266
2011-12-31 19:01:14-05:00    118
2011-12-31 19:01:15-05:00    325
2011-12-31 19:01:16-05:00    214
2011-12-31 19:01:17-05:00      3
2011-12-31 19:01:18-05:00    143
2011-12-31 19:01:19-05:00     28
2011-12-31 19:01:20-05:00     56
2011-12-31 19:01:21-05:00    120
2011-12-31 19:01:22-05:00     99
2011-12-31 19:01:23-05:00    102
2011-12-31 19:01:24-05:00     71
2011-12-31 19:01:25-05:00    464
2011-12-31 19:01:26-05:00    489
2011-12-31 19:01:27-05:00    404
2011-12-31 19:01:28-05:00    356
2011-12-31 19:01:29-05:00    197
2011-12-31 19:01:30-05:00    390
2011-12-31 19:01:31-05:00    345
2011-12-31 19:01:32-05:00    115
2011-12-31 19:01:33-05:00    377
2011-12-31 19:01:34-05:00    388
2011-12-31 19:01:35-05:00     39
2011-12-31 19:01:36-05:00    406
2011-12-31 19:01:37-05:00    408
2011-12-31 19:01:38-05:00    410
2011-12-31 19:01:39-05:00    256
Freq: S, Length: 100, dtype: int32


prng = pd.period_range('1990Q1', '2000Q4', freq='Q-NOV')

prng

PeriodIndex(['1990Q1', '1990Q2', '1990Q3', '1990Q4', '1991Q1', '1991Q2',
             '1991Q3', '1991Q4', '1992Q1', '1992Q2', '1992Q3', '1992Q4',
             '1993Q1', '1993Q2', '1993Q3', '1993Q4', '1994Q1', '1994Q2',
             '1994Q3', '1994Q4', '1995Q1', '1995Q2', '1995Q3', '1995Q4',
             '1996Q1', '1996Q2', '1996Q3', '1996Q4', '1997Q1', '1997Q2',
             '1997Q3', '1997Q4', '1998Q1', '1998Q2', '1998Q3', '1998Q4',
             '1999Q1', '1999Q2', '1999Q3', '1999Q4', '2000Q1', '2000Q2',
             '2000Q3', '2000Q4'],
            dtype='period[Q-NOV]', freq='Q-NOV')


ts = pd.Series(np.random.randn(len(prng)), prng)
ts

1990Q1    1.796304
1990Q2    0.659808
1990Q3   -0.647755
1990Q4    1.846486
1991Q1    0.488348
1991Q2    1.830351
1991Q3   -1.658804
1991Q4    0.585780
1992Q1   -0.596026
1992Q2   -1.900346
1992Q3   -0.066638
1992Q4    0.419037
1993Q1    0.055711
1993Q2   -2.103900
1993Q3    0.229944
1993Q4    0.317348
1994Q1   -0.776638
1994Q2   -0.241438
1994Q3   -0.587104
1994Q4    0.825772
1995Q1    2.444721
1995Q2    0.803142
1995Q3    0.494378
1995Q4   -0.984900
1996Q1   -0.431641
1996Q2    0.766768
1996Q3   -1.176313
1996Q4    0.339700
1997Q1   -1.523029
1997Q2    0.512173
1997Q3    1.359914
1997Q4    0.564407
1998Q1    0.354859
1998Q2   -0.493561
1998Q3    0.514986
1998Q4   -0.156142
1999Q1    1.047135
1999Q2    0.648944
1999Q3   -1.581937
1999Q4    0.261181
2000Q1   -0.809498
2000Q2    1.102175
2000Q3    0.424905
2000Q4   -0.775245
Freq: Q-NOV, dtype: float64


ts.index = (prng.asfreq('M', 'e') + 1).asfreq('H', 's') + 9#e-end,s-start

ts.head()

1990-03-01 09:00    1.796304
1990-06-01 09:00    0.659808
1990-09-01 09:00   -0.647755
1990-12-01 09:00    1.846486
1991-03-01 09:00    0.488348
Freq: H, dtype: float64


df = pd.DataFrame({"id":[1,2,3,4,5,6], "raw_grade":['a', 'b', 'b', 'a', 'a', 'e']})

df

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
idraw_grade
01a
12b
23b
34a
45a
56e
df["grade"] = df["raw_grade"].astype("category")
df["grade"]

0    a
1    b
2    b
3    a
4    a
5    e
Name: grade, dtype: category
Categories (3, object): [a, b, e]


df["grade"].cat.categories = ["very good", "good", "very bad"]

df["grade"]

0    very good
1         good
2         good
3    very good
4    very good
5     very bad
Name: grade, dtype: category
Categories (3, object): [very good, good, very bad]


df["grade"] = df["grade"].cat.set_categories(["very bad", "bad", "medium", "good", "very good"])

df["grade"]#methods under Series .cat return a new Series by default

0    very good
1         good
2         good
3    very good
4    very good
5     very bad
Name: grade, dtype: category
Categories (5, object): [very bad, bad, medium, good, very good]


df.sort_values(by="grade")

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
idraw_gradegrade
56every bad
12bgood
23bgood
01avery good
34avery good
45avery good
df.groupby("grade").size()

grade
very bad     1
bad          0
medium       0
good         2
very good    3
dtype: int64


ts = pd.Series(np.random.randn(1000), index=pd.date_range('1/1/2000', periods=1000))
ts

2000-01-01    0.585074
2000-01-02    0.605786
2000-01-03    0.697632
2000-01-04   -0.783338
2000-01-05    1.150780
2000-01-06   -0.718491
2000-01-07    0.696745
2000-01-08    0.270574
2000-01-09    0.657496
2000-01-10   -2.613661
2000-01-11   -1.978929
2000-01-12    0.325563
2000-01-13    0.286470
2000-01-14   -0.315502
2000-01-15    0.487703
2000-01-16   -1.830420
2000-01-17    0.847074
2000-01-18   -2.363392
2000-01-19    0.139429
2000-01-20   -0.512045
2000-01-21    0.209301
2000-01-22   -0.202987
2000-01-23   -0.605512
2000-01-24    0.113967
2000-01-25   -0.546799
2000-01-26    1.758145
2000-01-27    0.299658
2000-01-28   -0.614838
2000-01-29    0.262877
2000-01-30    0.021676
                ...   
2002-08-28    0.222110
2002-08-29   -1.846013
2002-08-30   -0.094660
2002-08-31    1.281895
2002-09-01   -1.072053
2002-09-02    0.503427
2002-09-03   -0.499512
2002-09-04   -1.080912
2002-09-05   -0.780288
2002-09-06   -0.537608
2002-09-07   -0.991904
2002-09-08    0.159327
2002-09-09    0.224638
2002-09-10    2.063388
2002-09-11    1.217366
2002-09-12    0.603689
2002-09-13    0.832689
2002-09-14   -1.788089
2002-09-15   -2.183370
2002-09-16   -0.759798
2002-09-17   -0.836241
2002-09-18    0.298536
2002-09-19    1.969939
2002-09-20   -0.688728
2002-09-21   -0.964116
2002-09-22   -1.279596
2002-09-23    0.357739
2002-09-24    1.253534
2002-09-25   -0.798673
2002-09-26   -1.023241
Freq: D, Length: 1000, dtype: float64


ts.cumsum()

2000-01-01     0.585074
2000-01-02     1.190860
2000-01-03     1.888493
2000-01-04     1.105155
2000-01-05     2.255935
2000-01-06     1.537445
2000-01-07     2.234190
2000-01-08     2.504764
2000-01-09     3.162260
2000-01-10     0.548599
2000-01-11    -1.430329
2000-01-12    -1.104767
2000-01-13    -0.818296
2000-01-14    -1.133798
2000-01-15    -0.646095
2000-01-16    -2.476516
2000-01-17    -1.629442
2000-01-18    -3.992834
2000-01-19    -3.853405
2000-01-20    -4.365450
2000-01-21    -4.156149
2000-01-22    -4.359136
2000-01-23    -4.964649
2000-01-24    -4.850682
2000-01-25    -5.397481
2000-01-26    -3.639336
2000-01-27    -3.339678
2000-01-28    -3.954516
2000-01-29    -3.691639
2000-01-30    -3.669963
                ...    
2002-08-28   -12.290664
2002-08-29   -14.136676
2002-08-30   -14.231337
2002-08-31   -12.949442
2002-09-01   -14.021495
2002-09-02   -13.518068
2002-09-03   -14.017579
2002-09-04   -15.098492
2002-09-05   -15.878779
2002-09-06   -16.416387
2002-09-07   -17.408292
2002-09-08   -17.248965
2002-09-09   -17.024327
2002-09-10   -14.960940
2002-09-11   -13.743574
2002-09-12   -13.139885
2002-09-13   -12.307196
2002-09-14   -14.095285
2002-09-15   -16.278655
2002-09-16   -17.038453
2002-09-17   -17.874694
2002-09-18   -17.576157
2002-09-19   -15.606219
2002-09-20   -16.294946
2002-09-21   -17.259062
2002-09-22   -18.538658
2002-09-23   -18.180919
2002-09-24   -16.927385
2002-09-25   -17.726058
2002-09-26   -18.749299
Freq: D, Length: 1000, dtype: float64


ts.plot()#绘制所有列

<matplotlib.axes._subplots.AxesSubplot at 0x22f031729e8>

在这里插入图片描述

df = pd.DataFrame(np.random.randn(1000, 4), index=ts.index,
   .....:                   columns=['A', 'B', 'C', 'D'])

df

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}
ABCD
2000-01-01-0.4347701.7971700.354820-0.166193
2000-01-020.4202330.8232510.468701-0.582949
2000-01-030.6746681.0322301.1343460.696656
2000-01-040.9486840.188295-0.604249-0.062779
2000-01-05-1.381247-0.360335-0.3464911.072618
2000-01-060.4925101.9243410.5226050.293788
2000-01-07-1.1290930.0638740.0991830.557496
2000-01-081.142263-0.3091921.1400491.007656
2000-01-091.751273-0.7471530.795127-0.480155
2000-01-10-1.519661-1.1877340.417908-0.675147
2000-01-11-0.0961921.0953080.0946481.485943
2000-01-120.109115-0.213535-0.9272501.189941
2000-01-13-0.787367-0.9197871.2867090.894471
2000-01-14-0.5848500.7940880.533716-0.159539
2000-01-15-1.352332-0.8804460.0419340.002573
2000-01-160.3179330.9579250.8137800.952499
2000-01-170.9503170.162642-0.018575-0.940598
2000-01-18-2.0211251.5921080.219355-1.300103
2000-01-19-0.673145-1.852674-0.4928450.070786
2000-01-20-0.562802-0.5040830.980132-0.079636
2000-01-210.6939270.276601-0.5022671.824789
2000-01-220.240543-0.0490040.051460-1.093965
2000-01-230.1591810.5593770.353952-1.750909
2000-01-24-1.009695-0.1699142.214441-1.301680
2000-01-250.741394-0.206067-1.250305-2.021061
2000-01-26-1.050527-0.4487260.7448410.559876
2000-01-27-0.2689870.755171-0.865320-0.077159
2000-01-28-1.445525-0.4438870.0483990.295317
2000-01-29-0.348641-0.5708660.446533-0.745215
2000-01-30-0.8038830.7198170.035095-0.057671
...............
2002-08-28-1.400857-1.993967-0.563839-0.553431
2002-08-29-0.860120-0.252746-0.5853360.083630
2002-08-300.6772180.113083-0.507485-1.247440
2002-08-311.9019130.124469-0.4829480.093981
2002-09-011.7288611.909778-1.206848-1.324399
2002-09-021.419153-1.000495-0.117854-0.630926
2002-09-030.716920-0.8317952.443522-0.247801
2002-09-04-0.886588-0.4872400.4765271.273604
2002-09-05-2.361533-0.074533-1.0950400.087406
2002-09-06-1.225924-0.4448360.378192-0.785585
2002-09-07-1.0643950.0460030.1485250.393557
2002-09-08-0.2946590.912430-0.7957670.064672
2002-09-090.2768460.993007-0.4931920.673319
2002-09-101.6760720.102106-1.286082-1.454404
2002-09-112.1245210.0694510.4950540.148496
2002-09-120.821348-0.8807140.9339781.869043
2002-09-13-0.890738-1.2639200.128660-0.282550
2002-09-14-1.0974840.6521240.702043-0.552927
2002-09-150.1613430.1573930.851718-1.265120
2002-09-160.865516-1.196734-0.985248-1.472387
2002-09-17-0.5392481.388908-0.870515-0.671165
2002-09-181.1545110.879535-0.249820-0.393302
2002-09-191.2371630.6680460.9178170.300664
2002-09-20-0.1878010.173142-0.2253072.142230
2002-09-210.517452-0.5471581.587477-0.922776
2002-09-220.4247840.6968311.3402581.252117
2002-09-23-0.687751-0.006990-0.6072200.709964
2002-09-24-1.8113470.2004852.117700-0.468944
2002-09-25-0.431668-0.3859970.3039360.817534
2002-09-260.6789591.0619571.2528700.735550

1000 rows × 4 columns

df = df.cumsum()
plt.figure(); df.plot(); plt.legend(loc='best')

<matplotlib.legend.Legend at 0x22f052c62b0>


<Figure size 432x288 with 0 Axes>

在这里插入图片描述

JoyceCoder
关注
  • 0
    点赞
  • 踩
  • 1
    收藏
    觉得还不错? 一键收藏
  • 3
    评论
专栏目录
python遍历data、并输出结果_Python遍历pandas数据方法总结
weixin_39710106的博客
11-30 2094
前言Pandas是python的一个数据分析包，提供了大量的快速便捷处理数据的函数和方法。其中Pandas定义了Series 和 DataFrame两种数据类型，这使数据操作变得更简单。Series 是一种一维的数据结构，类似于将列表数据值与索引值相结合。DataFrame 是一种二维的数据结构，接近于电子表格或者mysql数据库的形式。在数据分析中不可避免的涉及到对数据的遍历查询和处理，比如我们...
10分钟学pandas中文版
11-05
10分钟学pandas中文版 Pandas 是 Python 中一个功能强大且广泛使用的数据处理库，它提供了高效、灵活的数据结构和操作方式。本教程将带您快速了解 Pandas 的基本概念和使用方法。 导入必要的库 在开始使用 Pandas...
非常全面的Pandas入门教程
热门推荐
weixin_44489066的博客
04-24 5万+
pandas是基于NumPy的一种数据分析工具，在机器学习任务中，我们首先需要对数据进行清洗和编辑等工作，pandas库大大简化了我们的工作量，熟练并掌握pandas常规用法是正确构建机器学习模型的第一步。 目录 如何安装pandas 如何导入pandas库和查询相应的版本信息 pandas数据类型 series教程 dataframe教程 小结 1. 如何安装Pandas 最常用的方法是通过...
Go最全Pandas常见用法总结_pandas没有head方法(4)，成为阿里P7Golang架构师到底有多难
最新发布
2401_84911273的博客
05-16 985
df.pivot_table(index=col1, values=[col2,col3], aggfunc=max)：创建一个按列col1进行分组，并计算col2和col3的最大值的数据透视表。df.isnull()：查看空置(注：空置部分会用true显示，不是空置False显示)（扩展：df[‘two’].isnull，查看“two”这一列的空置）返回DataFrame格式的数据。s.replace([1,3],[‘one’,‘three’])：用’one’代替1，用’three’代替3。
python基础：小白自学pandas--DataFrame数据读取的方法
m0_51367666的博客
07-24 1万+
目录 一、直接用方括号[]取下标读取，DataFrame[列标签][行标签] 二、用DataFrame.loc[行标签,列标签]按照行/列标签名index读取 三、用DataFrame.iloc[行序列,列序列]按所在行/列号读取 四、DataFrame.ix[] 本人小白一枚，正在自学python，pandas模块的DataFrame数据类型是一个处理表格数据的强大工具，下面是我对读取DataFrame内指定数据的方法进行了初步学习与探索。 一、直接用方括号[]取下标读取，D...
边看边撸，10分钟高效掌握Pandas技术！
Oner.wv的专栏
09-17 3187
点击上方“AI派”，选择“设为星标”最新分享，第一时间送达！作者：贾胜杰，硕士，退役军人，电气工程专业，现成功转行K12领域数据挖掘工程师，不仅在数据清理、分析和预测方向...
【Pandas指南】给我十分钟，带你光速起步Pandas
谢谢你们的关注
12-26 355
【Pandas指南】给我十分钟，带你光速起步Pandas
10分钟搞定pandas.docx
05-30
pandas 库是 Python 中一个非常流行的数据处理和分析库，提供了高效、灵活的数据结构和操作函数，广泛应用于数据科学、机器学习、金融分析、科学计算等领域。 一、 创建对象 pandas 库提供了多种方式来创建对象，...
十分钟学习Pandas
05-12
**十分钟学习Pandas** 在数据分析领域，Pandas库是不可或缺的一部分，它为Python提供了一种高效的数据结构，便于处理和分析各种类型的数据集。Pandas是数据科学家和工程师的首选工具，尤其在数据预处理、清洗和探索...
学习pandas的数据
09-09
"学习pandas的数据"这个主题涵盖了对Pandas库的深入理解和实践应用。以下是对Pandas的一些核心知识点的详细介绍： 1. **数据结构**: Pandas主要提供两种核心数据结构——Series和DataFrame。Series是一维的数据结构...
Python数据分析之pandas学习
02-25
来源于cnblogs，介绍了数据结构，数据索引index，利用pandas查询数据，统计分析等。在pandas中有两类非常...序列的创建主要有三种方式：1）通过一维数组创建序列1.importnumpyasnp,pandasaspd2.arr1=np.arange(10)3.a
pandas(五) -- 文本处理
Blankit1的博客
06-21 1990
文本处理.–使用字符串处理函数 先将pandas的series对象转成strings对象，再使用字符串相关函数。 s = pd.Series(['A','b','C','bbhello','123',np.nan,'hj']) 0 A 1 b 2 C 3 bbhello 4 123 5 NaN 6 ...
十分钟学会pandas（入门级）
summer_original的博客
11-06 962
写给刚接触pandas的小白们，深一点的可以看这里。这也是我的第一篇博客，希望大家多多指教~ 1、在使用前，我们需要对相应模块就行导入（也就是import）import pandas as pd import numpy as np import matplotlib.pyplot as plt2、创建对象 通过传递一个列表（list）的值创建序列（series），让pandas创建一个默认的
python学习笔记（二）——Pandas十分钟入门
哇哇小仔的博客
12-02 5555
Pandas的数据结构主要分为三种： Series：一维同类型元素的数组 DataFrame：二维，大小可变的的表格结构，列与列的数据类型可以不同 Panel：三维，大小可变的数组首先引入一些需要用到的包（注意，如果使用的是jupyter notebook，因为后边需要用到画图的功能，如果希望在jupyter notebook中直接显示图像，就在import这些包之前加上一条命令： In
python数据科学系列：pandas入门详细教程
python爬虫人工智能大数据
05-06 9441
导读前2篇分别系统性介绍了numpy和matplotlib的入门基本知识，今天本文自然是要对pandas进行入门详细介绍，通过本文你将系统性了解pandas为何会有数据分析界"瑞...
十分钟搞定pandas（一）
qq826364410的专栏
10-10 739
这是关于pandas的简短介绍，主要面向新用户。可以参阅Cookbook了解更复杂的使用方法。 习惯上,我们做以下导入 Python 123In [1]: import pandas as pdIn [2]: import numpy as npIn [3]: import matpl
pandas教程：series和dataframe
好好学java
02-09 613
起步 pandas是一种Python数据分析的利器，是一个开源的数据分析包，最初是应用于金融数据分析工具而开发出来的，因此pandas为时间序列分析提供了很好的支持。pandas是PyData项目的一部分。 官网：http://pandas.pydata.org/ 官方文档：http://pandas.pydata.org/pandas-docs/stable/ 安装与导入 安装方式 Python...
Pandas 十：字符串处理
cai_and_luo的博客
05-20 646
Pandas 十：字符串处理0、读取北京2018年天气数据1、获取Series的str属性，使用各种字符串处理函数2、使用str的startswith、contains等得到bool的Series可以做条件查询3、需要多次str处理的链式操作4. 使用正则表达式的处理 前面我们已经使用了字符串的处理函数： df[“bWendu”].str.replace(“℃”, “”).astype(‘int32’) Pandas的字符串处理： 使用方法：先获取Series的str属性，然后在属性上调用函数； 只能在字符
pandas学习路线
04-02
当你开始学习pandas时，以下是一个简单的学习路线： 1. 学习Python基础知识：在使用pandas之前，建议先掌握Python的基本语法和数据结构，例如列表、字典和函数等。 2. 安装pandas库：使用pip或conda等工具安装...

“相关推荐”对你有帮助么？

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论 3
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值