Pandas 的基本操作

原始数据结构如下：

code date close prince
0 a 1988-01-04 525.11 23
1 a 1988-01-05 532.04 24
2 a 1988-01-06 527.89 25
3 a 1988-01-07 538.28 26
4 a 1988-01-08 540.28 27
5 a 1988-01-11 548.18 28
6 a 1988-01-12 552.76 29
7 b 1988-02-16 647.56 51
8 b 1988-02-17 648.95 52
9 b 1988-02-19 658.75 53
10 b 1988-02-22 663.03 54
11 b 1988-02-23 658.32 55
12 c 1988-05-06 645.69 22
13 c 1988-05-09 639.77 23
14 c 1988-05-10 635.47 24
15 c 1988-05-11 634.05 25
16 c 1988-05-12 636.68 26
17 c 1988-05-13 650.66 27
18 c 1988-05-16 657.57 28
19 c 1988-05-17 658.65 29
20 c 1988-05-18 670.31 30
21 c 1988-05-19 678.77 31
22 c 1988-05-20 678.24 32
23 c 1988-05-24 698.29 33
24 c 1988-05-25 705.51 34
25 c 1988-05-26 703.54 35
26 c 1988-05-27 716.85 36
27 d 1988-07-06 703.15 61
28 d 1988-07-07 707.02 62
29 d 1988-07-08 701.41 63
30 d 1988-07-11 702.52 64
31 d 1988-07-12 696.11 65
32 d 1988-07-13 692.79 66
33 d 1988-07-14 687.36 67
34 d 1988-07-15 690.86 68
35 d 1988-07-18 678.95 69

1， 累加某一列值（部分输出）

def __f1(x):
    x['bp'] = x['close'].cumsum()
    return x
df = pd.read_excel("chg.xlsx")
df= df.groupby(['code']).apply(__f1)
print df

code date close prince bp
0 a 1988-01-04 525.11 23 525.11
1 a 1988-01-05 532.04 24 1057.15
2 a 1988-01-06 527.89 25 1585.04
3 a 1988-01-07 538.28 26 2123.32
4 a 1988-01-08 540.28 27 2663.60
5 a 1988-01-11 548.18 28 3211.78

2, shift 函数:在x或y轴上进行平移（部分输出）

def __f1(x):
    x['bp'] = x['close'] - x['close'].shift(1)
    return x
df = pd.read_excel("chg.xlsx")
df= df.groupby(['code']).apply(__f1)
print df

code date close prince bp
0 a 1988-01-04 525.11 23 NaN
1 a 1988-01-05 532.04 24 6.93
2 a 1988-01-06 527.89 25 -4.15
3 a 1988-01-07 538.28 26 10.39
4 a 1988-01-08 540.28 27 2.00
5 a 1988-01-11 548.18 28 7.90
6 a 1988-01-12 552.76 29 4.58
7 b 1988-02-16 647.56 51 NaN

3，排序

def __f1(x):
    x['bp'] = x['close'].sort_values(ascending=False)
    return x
df = pd.read_excel("chg.xlsx")
df= df.groupby(['code']).apply(__f1)
print df

code date close prince bp
0 a 1988-01-04 525.11 23 525.11
1 a 1988-01-05 532.04 24 532.04
2 a 1988-01-06 527.89 25 527.89
3 a 1988-01-07 538.28 26 538.28