《利用Python 进行数据分析》pandas 总结

1.In [1]: from pandas import Series,DataFrame

In [2]: import pandas as pd
In [4]: ss=pd.Series

In [5]: df=pd.DataFrame

In [6]: obj=ss([3,2,1,4])

In [8]: obj
Out[8]:
0 3
1 2
2 1
3 4
dtype: int64

In [9]: obj.values
Out[9]: array([3, 2, 1, 4], dtype=int64)

In [10]: obj.index
Out[10]: Int64Index([0, 1, 2, 3], dtype=’int64’)

In [13]: obj2=ss([3,2,2,6666],index=[‘a’,’b’,’c’,’d’])

In [14]: obj2
Out[14]:
a 3
b 2
c 2
d 6666
dtype: int64

2.
In [15]: obj2[obj2>3]
Out[15]:
d 6666
dtype: int64

In [16]: obj2+3
Out[16]:
a 6
b 5
c 5
d 6669
dtype: int64

In [17]: import numpy as np

In [18]: np.sqrt(obj2)
Out[18]:
a 1.732051
b 1.414214
c 1.414214
d 81.645576
dtype: float64

3.In [19]: sadada={‘love1’:2232,’love2’:33232,’love3’:88988}

In [20]: sadada{‘love1’}
File “”, line 1
sadada{‘love1’}
^
SyntaxError: invalid syntax

In [21]: sadada[‘love1’]
Out[21]: 2232

In [22]: obj3=ss(sadada)

In [23]: obj3
Out[23]:
love1 2232
love2 33232
love3 88988
dtype: int64

4.In [30]: obj4=ss(sadada,index=states)

In [31]: obj4
Out[31]:
CC NaN
bb NaN
aa NaN
dtype: float64

In [32]: pd.isnull(obj4)(用来表示数值是否缺失)
Out[32]:
CC True
bb True
aa True
dtype: bool

5.
In [33]: obj3
Out[33]:
love1 2232
love2 33232
love3 88988
dtype: int64

In [34]: obj3.name=’love number’

In [35]: obj3
Out[35]:
love1 2232
love2 33232
love3 88988
Name: love number, dtype: int64

In [36]: obj3.index.name=’love man‘

In [37]: obj3
Out[37]:
love man
love1 2232
love2 33232
love3 88988
Name: love number, dtype: int64

6.
In [39]: data={‘name’:[‘Bob’,’KIM’,’Apple’,’Gogo’,’feiiw’],’year’:[1222,1333,1414,1996,1898],’pop’:[2,24242,33,44,11151]}

In [40]: frame=df(data)

In [41]: frame
Out[41]:
name pop year
0 Bob 2 1222
1 KIM 24242 1333
2 Apple 33 1414
3 Gogo 44 1996
4 feiiw 11151 1898

In [42]: frame=df(data,columns=[‘pop’,’name’,’year’])

In [43]: frame
Out[43]:
pop name year
0 2 Bob 1222
1 24242 KIM 1333
2 33 Apple 1414
3 44 Gogo 1996
4 11151 feiiw 1898

7.
In [44]: frame.columns
Out[44]: Index([u’pop’, u’name’, u’year’], dtype=’object’)

In [45]: frame[‘pop’]
Out[45]:
0 2
1 24242
2 33
3 44
4 11151
Name: pop, dtype: int64

8.
In [47]: frame.year
Out[47]:
0 1222
1 1333
2 1414
3 1996
4 1898
Name: year, dtype: int64

In [48]: frame.name
Out[48]:
0 Bob
1 KIM
2 Apple
3 Gogo
4 feiiw
Name: name, dtype: object

In [49]: frame.pop
Out[49]: