7-pandas分组计算-白红宇

7-pandas分组计算

阅读量：6083 次

发布时间：2019-06-20

本文共 5708 字，大约阅读时间需要 19 分钟。

#encoding:utf8import numpy as npimport pandas as pd'''分组计算：    拆分：根据什么进行分组    应用：每个分组进行怎样的计算    合并：每个分组的计算结果合并起来'''df = pd.DataFrame(    {        'key1':['a','a','b','b','a'],        'key2':['one','two','one','two','one'],        'data1':np.random.randint(1,10,5),        'data2':np.random.randint(1,10,5),    })print(df)'''   data1  data2 key1 key20      3      9    a  one1      3      5    a  two2      7      3    b  one3      1      1    b  two4      2      1    a  one'''print(df['data1'])'''0    71    72    43    14    8'''#计算data1列按照key1字段聚合求平均值print(df['data1'].groupby(df['key1']).mean())'''   data1  data2 key1 key20      1      3    a  one1      4      4    a  two2      3      9    b  one3      7      3    b  two4      1      3    a  onekey1a    2b    5'''#除了可以在数据内按照聚合，也可以自定义聚合#这里列表中元素表示data1列的元素位置聚合#data1列下的第1,3,4个元素聚合，第2,5个元素聚合，然后求平均值key = [1,2,1,1,2]print(df)print(df['data1'].groupby(key).mean())'''   data1  data2 key1 key20      9      9    a  one1      5      2    a  two2      4      8    b  one3      2      4    b  two4      1      3    a  one1    52    3'''#分组也可以是多级列表#在groupby中按照元素索引顺序进行排组的依据先后print(df)print(df['data1'].groupby([df['key1'],df['key2']]).sum())print(df['data1'].groupby([df['key2'],df['key1']]).sum())'''   data1  data2 key1 key20      9      8    a  one1      3      4    a  two2      7      8    b  one3      4      1    b  two4      3      7    a  onea[9,3,3]---->one[9,3]....key1  key2a     one     12      two      3b     one      7      two      4one[9,7,3]---->a[9,3]....key2  key1one   a       12      b        7two   a        3      b        4'''#查看分组的个数print(df['data1'].groupby([df['key1'],df['key2']]).size())'''key1  key2a     one     2      two     1b     one     1      two     1'''#按照key1进行分组#生成的是一个DataFrameprint(df.groupby('key1').sum())'''      data1  data2key1a        16     12b        10      6'''#对分组之后可进行索引的选取print(df.groupby('key1').sum()['data1'])'''key1a    14b    15'''#当然也可以进行多级分组，然后转换为DataFramemean = df.groupby(['key1','key2']).sum()print(mean)print(mean.unstack())'''           data1  data2key1 key2a    one      17      6     two       1      3b    one       7      3     two       1      6     data1     data2key2   one two   one twokey1a       17   1     6   3b        7   1     3   6'''#groupby支持迭代for name,group in df.groupby('key1'):    print(name)    print(group)'''a   data1  data2 key1 key20      9      8    a  one1      8      8    a  two4      7      7    a  oneb   data1  data2 key1 key22      6      1    b  one3      5      8    b  two'''#也可以对groupby转换为字典print(dict(list(df.groupby('key1'))))print(dict(list(df.groupby('key1')))['a'])print(dict(list(df.groupby('key1')))['b'])'''{'a':    data1  data2 key1 key20      9      4    a  one1      4      3    a  two4      5      6    a  one, 'b':    data1  data2 key1 key22      3      7    b  one3      2      8    b  two}   data1  data2 key1 key20      9      4    a  one1      4      3    a  two4      5      6    a  one   data1  data2 key1 key22      3      7    b  one3      2      8    b  two'''#按照列类型进行分组print(df.groupby(df.dtypes,axis=1).sum())'''0      4   aone1     15   atwo2     10   bone3      8   btwo4     18   aone'''#以上都是按照列表进行分组#下面用其他分组形式来进行分组#通过字典进行分组df = pd.DataFrame(    np.random.randint(1,10,(5,5)),    columns=list('abcde'),    index=['Alice','Bob','Candy','Dark','Emily'])#看一下处理非数字df.ix[1,1:3] = np.NaNprint(df)'''       a    b    c  d  eAlice  3  3.0  7.0  7  9Bob    4  NaN  NaN  3  4Candy  9  5.0  1.0  4  1Dark   6  3.0  9.0  9  9Emily  8  4.0  2.0  3  6'''mapping = {    'a':'red',    'b':'red',    'c':'blue',    'd':'orange',    'e':'blue'}grouped = df.groupby(mapping,axis=1)print(grouped.sum())'''       a    b    c  d  eAlice  4  5.0  2.0  2  2Bob    3  NaN  NaN  8  1Candy  6  7.0  6.0  9  7Dark   7  8.0  1.0  4  3Emily  8  6.0  8.0  3  3       blue  orange   redAlice   4.0     2.0   9.0Bob     1.0     8.0   3.0Candy  13.0     9.0  13.0Dark    4.0     4.0  15.0Emily  11.0     3.0  14.0可以看对Nan和数字分组计算是按照Nan=0来处理的'''print(grouped.size())print(grouped.count())'''blue      2orange    1red       2       blue  orange  redAlice     2       1    2Bob       1       1    1Candy     2       1    2Dark      2       1    2Emily     2       1    2Nan是没有统计个数的'''#通过函数进行分组df = pd.DataFrame(    np.random.randint(1,10,(5,5)),    columns=list('abcde'),    index=['Alice','Bob','Candy','Dark','Emily'])#按照行索引def _group_by(idx):    print(idx)    return idxprint(df.groupby(_group_by).size())print(df.groupby(_group_by).count())'''AliceBobCandyDarkEmily       a  b  c  d  eAlice  1  1  1  1  1Bob    1  1  1  1  1Candy  1  1  1  1  1Dark   1  1  1  1  1Emily  1  1  1  1  1按照行来进行分组的'''#按照行索引长度def _group_by2(idx):    print(idx)    return len(idx)print(df.groupby(_group_by2).size())print(df.groupby(_group_by2).count())'''AliceBobCandyDarkEmily   a  b  c  d  e3  1  1  1  1  14  1  1  1  1  15  3  3  3  3  3'''#通过索引级别进行分组columns = pd.MultiIndex.from_arrays(    [        ['China','USA','China','USA','China'],        ['A','A','B','C','B']    ],    names=['country','index'])df = pd.DataFrame(np.random.randint(1,10,(5,5)),columns=columns)print(df)'''country China USA China USA Chinaindex       A   A     B   C     B0           9   5     4   5     21           8   5     4   6     92           2   7     3   1     33           2   1     6   5     54           5   3     6   9     4'''print(df.groupby(level='country',axis=1).sum())'''country  China  USA0           17   121           19   122           19    83           21   134           18    9'''print(df.groupby(level='index',axis=1).sum())'''index   A   B  C0       7   7  61      11  12  32      12  11  93      14  11  44      17  15  8'''

转载于:https://my.oschina.net/eddylinux/blog/1529788

你可能感兴趣的文章

Do not request Window.FEATURE_SUPPORT_ACTION_BAR

为什么Windows会蓝屏（Blue Screen Of Death）？

查看>>