panda库2

 >>> a=pd.Series([1,2],index=['a','b'])
 >>> a
 a    1
 b    2
 dtype: int64
 >>> b=pd.Series(['b','a'])
 >>> b
 0    b
 1    a
 dtype: object

 >>> b.index
 RangeIndex(start=0, stop=2, step=1)
 >>> b.values
 array(['b', 'a'], dtype=object)
 >>> a/2
 a    0.5
 b    1.0
 dtype: float64

 >>> dic={'zhang':1,'li':2}
 >>> d=pd.Series(dic)  　　　　　　　　　　　　　　　　　　　　参数的形式是字典，numpy中参数是列表
 >>> d
 li       2
 zhang    1
 dtype: int64
 >>> frame=pd.DataFrame('name':['zhang','li'],'age':[12,13],'addr':['beijing','shanghai'])
 SyntaxError: invalid syntax
 >>> dic={'name':['zhang','li'],'age':[12,13],'addr':['beijing','shanghai']}
 >>> frame=pd.DataFrame(dic)                             参数是字典
 >>> frame
        addr  age   name
 0   beijing   12  zhang
 1  shanghai   13     li
 >>> frame.columns                                       index，columns两个关键字属性
 Index(['addr', 'age', 'name'], dtype='object')
 >>> frame.index
 RangeIndex(start=0, stop=2, step=1)
 >>> frame2=pd.DataFrame(np.arange(16).reshape((4,4)),colums=['name','age','addr'],index=['a','b','c'])

 >>> frame2
    like  name  age  addr
 a     0     1    2     3
 b     4     5    6     7
 c     8     9   10    11
 d    12    13   14    15
 >>> frame.name      　　　　　　　　　　　　　　　　　　　　指定列名字
 0    zhang
 1       li
 Name: name, dtype: object

>>> frame2.ix[2]      查看某行  ix【】
like 8
name 9
age 10
addr 11
Name: c, dtype: int32
>>> frame2.ix[2,3]
11

 >>> frame2.index.name='id';frame2.columns.name='item'     对标头的name属性指定
 >>> frame2
 item  like  name  age  addr
 id
 a        0     1    2     3
 b        4     5    6     7
 c        8     9   10    11
 d       12    13   14    15
 >>> frame2['new']=12    　　　　　　　　　　　　　　　　　　添加新列
 >>> frame2
 item  like  name  age  addr  new
 id
 a        0     1    2     3   12
 b        4     5    6     7   12
 c        8     9   10    11   12
 d       12    13   14    15   12
 >>> frame2['new']
 id
 a    12
 b    12
 c    12
 d    12
 Name: new, dtype: int64

 >>> frame2['new']['b']    　　　　根据列行找到元素
 12
 >>> frame2.isin([2])
 item   like   name    age   addr    new
 id
 a     False  False   True  False  False
 b     False  False  False  False  False
 c     False  False  False  False  False
 d     False  False  False  False  False
 >>> del frame['new']

 >>> del frame2['new']　　　　　　　　　　删除列
 >>> frame2
 item  like  name  age  addr
 id
 a        0     1    2     3
 b        4     5    6     7
 c        8     9   10    11
 d       12    13   14    15

 >>> frame2[frame2<8]　　　　　　　　　　　　找到小于8的所有元素
 item  like  name  age  addr
 id
 a      0.0   1.0  2.0   3.0
 b      4.0   5.0  6.0   7.0
 c      NaN   NaN  NaN   NaN
 d      NaN   NaN  NaN   NaN
 >>> frame.T　　　　　　　　　　　　　　　　对表进行转置
             0         1
 addr  beijing  shanghai
 age        12        13
 name    zhang        li
 >>> frame2.T
 id    a  b   c   d
 item
 like  0  4   8  12
 name  1  5   9  13
 age   2  6  10  14
 addr  3  7  11  15

 >>> frame2.idxmin()       找到索引的最小值  idxmin（）
 item
 like    a
 name    a
 age     a
 addr    a
 dtype: object
 >>> frame2.idxmax()
 item
 like    d
 name    d
 age     d
 addr    d
 dtype: object
 >>> frame2.index.is_unique
 True
 >>> frame2.reindex(['one','two','three','four'])
 item   like  name  age  addr
 id
 one     NaN   NaN  NaN   NaN
 two     NaN   NaN  NaN   NaN
 three   NaN   NaN  NaN   NaN
 four    NaN   NaN  NaN   NaN
 >>> frame2
 item  like  name  age  addr
 id
 a        0     1    2     3
 b        4     5    6     7
 c        8     9   10    11
 d       12    13   14    15

 >>> frame2.drop('a')   　　　　　　　　删除行根据索引
 item  like  name  age  addr
 id
 b        4     5    6     7
 c        8     9   10    11
 d       12    13   14    15

>>> frame2.drop(['name'],axis=1)      　　删除列
item like age addr
id
a 0 2 3
b 4 6 7
c 8 10 11
d 12 14 15

 >>> frame2
 item  like  name  age  addr    每行-series
 id
 a        0     1    2     3
 b        4     5    6     7
 c        8     9   10    11
 d       12    13   14    15
 >>> ser=[1,2,3,4]
 >>> frame2-ser
 item  like  name  age  addr
 id
 a       -1    -1   -1    -1
 b        3     3    3     3
 c        7     7    7     7
 d       11    11   11    11

 >>> np.sqrt(frame2)  求所有的sqrt
 item      like      name       age      addr
 id
 a     0.000000  1.000000  1.414214  1.732051
 b     2.000000  2.236068  2.449490  2.645751
 c     2.828427  3.000000  3.162278  3.316625
 d     3.464102  3.605551  3.741657  3.872983

 >>> f=lambda x:x.max()-x.min()     对frame  f（x）默认是对一列的所有值中寻找
 >>> frame2.apply(f)
 item
 like    12
 name    12
 age     12
 addr    12
 dtype: int64
 >>> def f(x):
     return pd.Series([x.min(),x.max()],index=['min','max'])

 >>> frame.apply(f)
          addr  age   name
 min   beijing   12     li
 max  shanghai   13  zhang
 >>> frame2.sum()
 item
 like    24
 name    28
 age     32
 addr    36
 dtype: int64
 >>> frame.mean()
 age    12.5
 dtype: float64
 >>> frame2.mean()
 item
 like    6.0
 name    7.0
 age     8.0
 addr    9.0
 dtype: float64
 >>> frame2.describe()
 item        like       name        age       addr
 count   4.000000   4.000000   4.000000   4.000000
 mean    6.000000   7.000000   8.000000   9.000000
 std     5.163978   5.163978   5.163978   5.163978
 min     0.000000   1.000000   2.000000   3.000000
 25%     3.000000   4.000000   5.000000   6.000000
 50%     6.000000   7.000000   8.000000   9.000000
 75%     9.000000  10.000000  11.000000  12.000000
 max    12.000000  13.000000  14.000000  15.000000
 >>> frame.describe()
              age
 count   2.000000
 mean   12.500000
 std     0.707107
 min    12.000000
 25%    12.250000
 50%    12.500000
 75%    12.750000
 max    13.000000
 >>> frame2.sort_index()
 item  like  name  age  addr
 id
 a        0     1    2     3
 b        4     5    6     7
 c        8     9   10    11
 d       12    13   14    15
 >>> frame.sort_index()       以索引进行排序
        addr  age   name
 0   beijing   12  zhang
 1  shanghai   13     li
 >>> frame
        addr  age   name
 0   beijing   12  zhang
 1  shanghai   13     li
 >>> ser.sort_index()
 Traceback (most recent call last):
   File "<pyshell#95>", line 1, in <module>
     ser.sort_index()
 AttributeError: 'list' object has no attribute 'sort_index'
 >>> ser
 [1, 2, 3, 4, 5]