Series的基础操作

——————Series的基础操作——————
1.导入库

>>> import pandas as pd
>>> import numpy as np

2.初步生成Series

>>> s1 = pd.Series([2,3,4,6],index = ['a','b','c','d'])#index是索引
>>> print(s1)
a    2
b    3
c    4
d    6
dtype: int64
>>> print(s1[1])
3

3.两个Series简单运算

>>> s2 = pd.Series([3,4,5,6],index = ['a','b','c','d'])
>>> print(s1 + s2)
a     5
b     7
c     9
d    12
dtype: int64

4.简单切片

>>> s3 = s1[0:2]
>>> print(s3)
a    2
b    3
dtype: int64
>>> s4 = s1[s1>5]
>>> print(s4)
d    6
dtype: int64

5.利用Series创建字典

>>> s5 = pd.Series({'a':1,'b':2})
>>> print(s5)
a    1
b    2
dtype: int64

6.对Series进行遍历

#对Series进行遍历的时候，输出值
>>> for i in s5:
    print(i) 
1
2

#分别获取键和值
>>> print(s5.index)
Index(['a', 'b'], dtype='object')
>>> print(s5.index[0])
a
>>> print(s5.values)
[1 2]

#关于整数索引
```python
>>> s6 = pd.Series(np.arange(20))
>>> print(s6)
0      0
1      1
2      2
3      3
4      4
5      5
6      6
7      7
8      8
9      9
10    10
11    11
12    12
13    13
14    14
15    15
16    16
17    17
18    18
19    19
dtype: int32
>>> s7 = s6[10:].copy()
>>> print(s7)
10    10
11    11
12    12
13    13
14    14
15    15
16    16
17    17
18    18
19    19
dtype: int32
>>> print(s7[10])#默认为按键（标签）的方式
10
>>> print(s7[11])#print(s7[-1])会报错
11

#关于loc
#建议使用loc（按照标签）和iloc（按照索引index）来获取
>>> print(s7.loc[10])#解释为标签(键)
10
>>> print(s7.iloc[9])#解释为index（索引）
19

6.数据对齐

#pandas在进行两个Series对象的运算时，会按照索引进行对齐，然后计算
>>> s8 = pd.Series([12,23,34],index = ['c','a','d'])
>>> s9 = pd.Series([11,20,10],index = ['d','c','a'])
>>> s10 = s8 + s9
>>> print(s10)
a    33
c    32
d    45
dtype: int64
#所以不用管排序的事情，只要标签相同，就可以运算了
#pandas的索引允许重复，但是一般设置为不重复的
>>> s11 = pd.Series([12,23,34],index = ['c','a','d'])
>>> s12 = pd.Series([11,20,10,20],index = ['d','c','a','b'])
>>> s13 = s11 + s12
>>> print(s13)
a    33.0
b     NaN
c    32.0
d    45.0
dtype: float64

7.简单处理缺失值

>>> s11 = pd.Series([12,23,34],index = ['c','a','d'])
>>> s12 = pd.Series([11,20,10],index = ['c','a','b'])
>>> s13 = s12.add(s11)#使用add（）和 + 的效果一样，但是我们可以在使用add（）时加一个参数
>>> print(s13)
a    43.0
b     NaN
c    23.0
d     NaN
dtype: float64

#加入参数后
>>> s13 = s12.add(s11,fill_value = 0)
>>> print(s13)
a    43.0
b    10.0
c    23.0
d    34.0
dtype: float64

#判断是否有缺失数据
>>> s11 = pd.Series([12,23,34],index = ['c','a','d'])
>>> s12 = pd.Series([11,20,10],index = ['c','a','b'])
>>> s13 = s12.add(s11)
>>> print(s13.isnull())
a    False
b     True
c    False
d     True
dtype: bool

#进行布尔运算
>>> s14 = s13[s13.notnull()]
>>> print(s14)
a    43.0
c    23.0
dtype: float64

#扔掉缺失值
>>> s11 = pd.Series([12,23,34],index = ['c','a','d'])
>>> s12 = pd.Series([11,20,10],index = ['c','a','b'])
>>> s13 = s12.add(s11)
>>> s15 = s13.dropna()
>>> print(s15)
a    43.0
c    23.0
dtype: float64

#给缺失值赋值
>>> s11 = pd.Series([12,23,34],index = ['c','a','d'])
>>> s12 = pd.Series([11,20,10],index = ['c','a','b'])
>>> s13 = s12.add(s11)
>>> s16 = s13.fillna(0)
>>> print(s16)
a    43.0
b     0.0
c    23.0
d     0.0
dtype: float64

#为了不影响数据的总体趋势，可以填充为平均值
>>> s17 = s13.fillna(s13.mean())
>>> print(s17)
a    43.0
b    33.0
c    23.0
d    33.0
dtype: float64