关于pandas:Pandas入门教程四

groupby分组

import pandas as pdimport numpy as npdf=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)})print(df)grouped=df.groupby('A')print('-'*30)print(grouped.count())print('-'*30)grouped=df.groupby(['A','B'])print(grouped.count())print('-'*30)# 通过函数分组def get_letter_type(letter):    if letter.lower() in 'aeiou':        return 'a'    else:        return 'b'grouped=df.groupby(get_letter_type,axis=1)print(grouped.count())    

     A      B         C         D0  foo    one  1.429387  0.6435691  bar    one -0.858448 -0.2130342  foo    two  0.375644  0.2145843  bar  three  0.042284 -0.3304814  foo    two -1.421967  0.7681765  bar    two  1.293483 -0.3990036  foo    one -1.101385 -0.2363417  foo  three -0.852603 -1.718694------------------------------     B  C  DA           bar  3  3  3foo  5  5  5------------------------------           C  DA   B          bar one    1  1    three  1  1    two    1  1foo one    2  2    three  1  1    two    2  2------------------------------   a  b0  1  31  1  32  1  33  1  34  1  35  1  36  1  37  1  3

se=pd.Series([1,2,3,4,5],[6,9,8,9,8])print(se)se.groupby(level=0)

6    19    28    39    48    5dtype: int64<pandas.core.groupby.generic.SeriesGroupBy object at 0x111b00040>

# 分组求和grouped=se.groupby(level=0).sum()print(grouped)

6    18    89    6dtype: int64

df2=pd.DataFrame({'X':['A','B','A','B'],'Y':[1,2,3,4]})print(df2)

   X  Y0  A  11  B  22  A  33  B  4

# 按X分组,并查问A列的数据grp=df2.groupby('X').get_group('A')print(grp)

   X  Y0  A  12  A  3

Pandas 多级索引

arrays = [['bar', 'bar', 'baz', 'baz', 'foo', 'foo', 'qux', 'qux'],          ['one', 'two', 'one', 'two', 'one', 'two', 'one', 'two']]index=pd.MultiIndex.from_arrays(arrays,names=['first','second'])print(index)

MultiIndex([('bar', 'one'),            ('bar', 'two'),            ('baz', 'one'),            ('baz', 'two'),            ('foo', 'one'),            ('foo', 'two'),            ('qux', 'one'),            ('qux', 'two')],           names=['first', 'second'])

s=pd.Series(np.random.randn(8),index=index)print(s)

first  secondbar    one       0.120979       two      -0.440384baz    one       0.515106       two      -0.019882foo    one       1.149595       two      -0.369984qux    one      -0.930438       two       0.146044dtype: float64

# 分组求和grouped=s.groupby(level='first')print(grouped.sum())

firstbar   -0.319405baz    0.495224foo    0.779611qux   -0.784394dtype: float64

grouped=df.groupby(['A','B'])print(grouped.size())

A    B    bar  one      1     three    1     two      1foo  one      2     three    1     two      2dtype: int64

print(df)

     A      B         C         D0  foo    one  1.429387  0.6435691  bar    one -0.858448 -0.2130342  foo    two  0.375644  0.2145843  bar  three  0.042284 -0.3304814  foo    two -1.421967  0.7681765  bar    two  1.293483 -0.3990036  foo    one -1.101385 -0.2363417  foo  three -0.852603 -1.718694

print(grouped.describe().head())

              C                                                              \          count      mean       std       min       25%       50%       75%   A   B                                                                         bar one     1.0 -0.858448       NaN -0.858448 -0.858448 -0.858448 -0.858448       three   1.0  0.042284       NaN  0.042284  0.042284  0.042284  0.042284       two     1.0  1.293483       NaN  1.293483  1.293483  1.293483  1.293483   foo one     2.0  0.164001  1.789526 -1.101385 -0.468692  0.164001  0.796694       three   1.0 -0.852603       NaN -0.852603 -0.852603 -0.852603 -0.852603                           D                                                    \                max count      mean       std       min       25%       50%   A   B                                                                         bar one   -0.858448   1.0 -0.213034       NaN -0.213034 -0.213034 -0.213034       three  0.042284   1.0 -0.330481       NaN -0.330481 -0.330481 -0.330481       two    1.293483   1.0 -0.399003       NaN -0.399003 -0.399003 -0.399003   foo one    1.429387   2.0  0.203614  0.622191 -0.236341 -0.016364  0.203614       three -0.852603   1.0 -1.718694       NaN -1.718694 -1.718694 -1.718694   

                75%       max  A   B                          bar one   -0.213034 -0.213034      three -0.330481 -0.330481      two   -0.399003 -0.399003  foo one    0.423592  0.643569      three -1.718694 -1.718694  

grouped=df.groupby('A')grouped['C'].agg([np.sum,np.mean,np.std])

<div>
<style scoped>

.dataframe tbody tr th:only-of-type {    vertical-align: middle;}.dataframe tbody tr th {    vertical-align: top;}.dataframe thead th {    text-align: right;}

</style>

</div>

字符串操作

import pandas as pdimport numpy as nps=pd.Series(['A','b','c','D',np.nan])print(s)# 转小写print(s.str.lower())# 转大写print(s.str.upper())# 每个字符的长度print(s.str.len())

0      A1      b2      c3      D4    NaNdtype: object0      a1      b2      c3      d4    NaNdtype: object0      A1      B2      C3      D4    NaNdtype: object0    1.01    1.02    1.03    1.04    NaNdtype: float64

index=pd.Index([' Index','ru ',' men'])# 去掉空格print(index.str.strip())# 去掉右边的空格print(index.str.lstrip())# 去掉左边的空格print(index.str.rstrip())

Index(['Index', 'ru', 'men'], dtype='object')Index(['Index', 'ru ', 'men'], dtype='object')Index([' Index', 'ru', ' men'], dtype='object')

df=pd.DataFrame(np.random.randn(3,2),columns=['A a','B b'],index=range(3))print(df)

        A a       B b0  3.005273  0.4866961  1.093889  1.0542302 -2.846352  0.302465

# 列替换print(df.columns.str.replace(' ','_'))

Index(['A_a', 'B_b'], dtype='object')

s=pd.Series(['a_b_C','c_d_e','f_g_h'])print(s)

0    a_b_C1    c_d_e2    f_g_hdtype: object

print(s.str.split('_'))

0    [a, b, C]1    [c, d, e]2    [f, g, h]dtype: object

print(s.str.split('_',expand=True,n=1))

   0    10  a  b_C1  c  d_e2  f  g_h

s = pd.Series(['A','rumen','ru','rumen','xiao','zhan'])

print(s.str.contains('ru'))

0    False1     True2     True3     True4    False5    Falsedtype: bool

s=pd.Series(['a','a|b','a|c'])print(s)

0      a1    a|b2    a|cdtype: object

print(s.str.get_dummies(sep='|'))

   a  b  c0  1  0  01  1  1  02  1  0  1

索引

s=pd.Series(np.arange(5),np.arange(5)[::-1],dtype='int64')

s

4    03    12    21    30    4dtype: int64

print(s[s>2])

1    30    4dtype: int64

# isin查问索引在某个范畴print(s.isin([1,3,4]))

4    False3     True2    False1     True0     Truedtype: bool

# 依据索引查问数据print(s[s.isin([1,3,4])])

3    11    30    4dtype: int64

# 结构一个联结索引的数据s=pd.Series(np.arange(6),index=pd.MultiIndex.from_product([[1,2],['a','b','c']]))

print(s)

1  a    0   b    1   c    22  a    3   b    4   c    5dtype: int64

print(s.iloc[s.index.isin([(1,'b'),(2,'c')])])

1  b    12  c    5dtype: int64

# 结构一个工夫序列dates=pd.date_range('20200920',periods=8)print(dates)

DatetimeIndex(['2020-09-20', '2020-09-21', '2020-09-22', '2020-09-23',               '2020-09-24', '2020-09-25', '2020-09-26', '2020-09-27'],              dtype='datetime64[ns]', freq='D')

df=pd.DataFrame(np.random.randn(8,4),index=dates,columns=['A','B','C','D'])print(df)

                   A         B         C         D2020-09-20 -1.218522  2.067088  0.015009  0.1587802020-09-21 -0.546837 -0.601178 -0.894882  0.1720372020-09-22  0.189848 -0.910520  0.196186 -0.0734952020-09-23 -0.566892  0.899193 -0.450925  0.6332532020-09-24  0.038838  1.577004  0.580927  0.6090502020-09-25  1.562094  0.020813 -0.618859 -0.5152122020-09-26 -1.333947  0.275765  0.139325  1.1242072020-09-27 -1.271748  1.082302  1.036805 -1.041206

# 查问A列数据print(df['A'])

2020-09-20   -1.2185222020-09-21   -0.5468372020-09-22    0.1898482020-09-23   -0.5668922020-09-24    0.0388382020-09-25    1.5620942020-09-26   -1.3339472020-09-27   -1.271748Freq: D, Name: A, dtype: float64

# 查问小于0的数字,大于0的值默认被置为NaNdf.where(df<0)

<div>
<style scoped>

.dataframe tbody tr th:only-of-type {    vertical-align: middle;}.dataframe tbody tr th {    vertical-align: top;}.dataframe thead th {    text-align: right;}

</style>

</div>

# 查问小于0的数字,大于0的值变成正数print(df.where(df<0,-df))

                   A         B         C         D2020-09-20 -1.218522 -2.067088 -0.015009 -0.1587802020-09-21 -0.546837 -0.601178 -0.894882 -0.1720372020-09-22 -0.189848 -0.910520 -0.196186 -0.0734952020-09-23 -0.566892 -0.899193 -0.450925 -0.6332532020-09-24 -0.038838 -1.577004 -0.580927 -0.6090502020-09-25 -1.562094 -0.020813 -0.618859 -0.5152122020-09-26 -1.333947 -0.275765 -0.139325 -1.1242072020-09-27 -1.271748 -1.082302 -1.036805 -1.041206

# 查问小于0的数据,大于0的置为1000print(df.where(df<0,1000))

                      A            B            C            D2020-09-20    -1.218522  1000.000000  1000.000000  1000.0000002020-09-21    -0.546837    -0.601178    -0.894882  1000.0000002020-09-22  1000.000000    -0.910520  1000.000000    -0.0734952020-09-23    -0.566892  1000.000000    -0.450925  1000.0000002020-09-24  1000.000000  1000.000000  1000.000000  1000.0000002020-09-25  1000.000000  1000.000000    -0.618859    -0.5152122020-09-26    -1.333947  1000.000000  1000.000000  1000.0000002020-09-27    -1.271748  1000.000000  1000.000000    -1.041206

# 结构一个10行3列的数据df=pd.DataFrame(np.random.randn(10,3),columns=list('abc'))print(df)

          a         b         c0  1.761415  0.528009 -0.3472711 -0.682149  0.353312  0.3372292  1.080733 -0.272290  1.0203353 -0.979681 -1.753745  0.8363874  0.243748  2.085531 -0.9933185 -1.041006  1.518130 -0.0873836 -1.400354 -0.095196  3.0436397 -0.835144  0.926415 -1.2171028  0.326098  1.079906  0.1568849  1.836618 -1.288516 -2.492620

# 查问a>b的数据print(df.query('a>b'))

          a         b         c0  1.761415  0.528009 -0.3472712  1.080733 -0.272290  1.0203353 -0.979681 -1.753745  0.8363879  1.836618 -1.288516 -2.492620

# 查问c>b>a的数据print(df.query('(c<b) & (b<a)'))

          a         b         c0  1.761415  0.528009 -0.3472719  1.836618 -1.288516 -2.492620