panda库------对数据进行操作---合并，转换，拼接

 >>> frame2
        addr  age   name
 0   beijing   12  zhang
 1  shanghai   24     li
 2  hangzhou   24    cao
 >>> frame1
        addr   name
 0   beijing  zhang
 1  shanghai     li
 2  hangzhou    cao
 3  shenzhen    han
 >>> pd.merge(frame1,frame2)   以name列为连接进行拼接
        addr   name  age
 0   beijing  zhang   12
 1  shanghai     li   24
 2  hangzhou    cao   24
 >>> pd.merge(frame1,frame2,on='name')     指定  列  和拼接方式
      addr_x   name    addr_y  age
 0   beijing  zhang   beijing   12
 1  shanghai     li  shanghai   24
 2  hangzhou    cao  hangzhou   24
 >>> pd.merge(frame1,frame2,on='name',how='outer')
      addr_x   name    addr_y   age
 0   beijing  zhang   beijing  12.0
 1  shanghai     li  shanghai  24.0
 2  hangzhou    cao  hangzhou  24.0
 3  shenzhen    han       NaN   NaN
 >>> pd.merge(frame1,frame2,on='name',how='inner')
      addr_x   name    addr_y  age
 0   beijing  zhang   beijing   12
 1  shanghai     li  shanghai   24
 2  hangzhou    cao  hangzhou   24
 >>> pd.merge(frame1,frame2,on='name',how='left')
      addr_x   name    addr_y   age
 0   beijing  zhang   beijing  12.0
 1  shanghai     li  shanghai  24.0
 2  hangzhou    cao  hangzhou  24.0
 3  shenzhen    han       NaN   NaN
 >>> pd.merge(frame1,frame2,on='name',how='right')
      addr_x   name    addr_y  age
 0   beijing  zhang   beijing   12
 1  shanghai     li  shanghai   24
 2  hangzhou    cao  hangzhou   24
 >>> pd.merge(frame1,frame2,on='name',left_index=True)
      addr_x   name    addr_y  age
 0   beijing  zhang   beijing   12
 1  shanghai     li  shanghai   24
 2  hangzhou    cao  hangzhou   24
 >>> pd.merge(frame1,frame2,on='name',right_index=True)
      addr_x   name    addr_y  age
 0   beijing  zhang   beijing   12
 1  shanghai     li  shanghai   24
 2  hangzhou    cao  hangzhou   24
 >>> pd.merge(frame1,frame2,on='addr',right_index=True)
        addr name_x  age name_y
 0   beijing  zhang   12  zhang
 1  shanghai     li   24     li
 2  hangzhou    cao   24    cao

 >>> frame1.columns=['addr1','name1']
 >>> frame1.join(frame2)
       addr1  name1      addr   age   name    修改掉重复的列名称，然后join（）
 0   beijing  zhang   beijing  12.0  zhang
 1  shanghai     li  shanghai  24.0     li
 2  hangzhou    cao  hangzhou  24.0    cao
 3  shenzhen    han       NaN   NaN    NaN

 >>> array1
 array([[0, 1, 2],
        [3, 4, 5],
        [6, 7, 8]])
 >>> array1=np.arange(9).reshape((3,3))+6
 >>> array2=np.arange(9).reshape((3,3))
 >>> array1
 array([[ 6,  7,  8],
        [ 9, 10, 11],
        [12, 13, 14]])
 >>> np.concatenate([array1,array2],axis=1)   np模块中对元组进行concatenate（）
 array([[ 6,  7,  8,  0,  1,  2],
        [ 9, 10, 11,  3,  4,  5],
        [12, 13, 14,  6,  7,  8]])
 >>> np.concatenate([array1,array2],axis=0)
 array([[ 6,  7,  8],
        [ 9, 10, 11],
        [12, 13, 14],
        [ 0,  1,  2],
        [ 3,  4,  5],
        [ 6,  7,  8]])
 >>>
 >>> np.concatenate([array1,array2])
 array([[ 6,  7,  8],
        [ 9, 10, 11],
        [12, 13, 14],
        [ 0,  1,  2],
        [ 3,  4,  5],
        [ 6,  7,  8]])

 >>> ser1=pd.Series(np.random.rand(4))   pd模块中也有concat（）
 >>> ser1
 0    0.998915
 1    0.117503
 2    0.747180
 3    0.641508
 dtype: float64
 >>> ser1=pd.Series(np.random.rand(4)*100)
 >>> ser1
 0     8.818592
 1    42.317816
 2    43.274021
 3    23.245148
 dtype: float64
 >>> ser2=pd.Series(np.random.rand(4)*100,index=[5,6,7,8])
 >>> ser2
 5    58.416554
 6    11.840838
 7    38.146851
 8     0.135517
 dtype: float64
 >>> pd.concat([ser1,ser2])
 0     8.818592
 1    42.317816
 2    43.274021
 3    23.245148
 5    58.416554
 6    11.840838
 7    38.146851
 8     0.135517
 dtype: float64
 >>> pd.concat([ser1,ser2],axis=1)
            0          1
 0   8.818592        NaN
 1  42.317816        NaN
 2  43.274021        NaN
 3  23.245148        NaN
 5        NaN  58.416554
 6        NaN  11.840838
 7        NaN  38.146851
 8        NaN   0.135517

 >> pd.concat([ser1,ser2],axis=1,keys=[1,2])
            1          2
 0   8.818592        NaN
 1  42.317816  79.632793
 2  43.274021  96.700070
 3  23.245148  64.573269
 4        NaN  68.629709
 >>> ser2.index=[2,4,5,6]
 >>> ser2
 2    79.632793
 4    96.700070
 5    64.573269
 6    68.629709
 dtype: float64
 >>> ser1.combine_first(ser2)    对缺额的数据进行填充  combin_first()
 0     8.818592
 1    42.317816
 2    43.274021
 3    23.245148
 4    96.700070
 5    64.573269
 6    68.629709
 dtype: float64

 >>> ser1
 0    a
 1    b
 2    c
 3    d
 dtype: object
 >>> ser2
 2    0
 4    1
 5    2
 6    3
 dtype: int32
 >>> ser2.combine_first(ser1)   ser1在后
 0    a
 1    b
 2    0
 3    d
 4    1
 5    2
 6    3
 dtype: object
 >>> ser1[:2].combine_first(ser2)  ser1在前
 0    a
 1    b
 2    0
 4    1
 5    2
 6    3
 dtype: object

 >>> frame1=pd.DataFrame({'name':['zhang','li','wang'],'age':[12,45,34],'addr':['beijing','shanghai','shenzhen']})
 >>> frame1
        addr  age   name
 0   beijing   12  zhang
 1  shanghai   45     li
 2  shenzhen   34   wang
 >>> frame1.stack()        frame的进栈和出栈
 0  addr     beijing
    age           12
    name       zhang
 1  addr    shanghai
    age           45
    name          li
 2  addr    shenzhen
    age           34
    name        wang
 dtype: object
 >>> frame1.stack().unstack()
        addr age   name
 0   beijing  12  zhang
 1  shanghai  45     li
 2  shenzhen  34   wang
 >>> frame1.stack().unstack(0)   列和索引转换
             0         1         2
 addr  beijing  shanghai  shenzhen
 age        12        45        34
 name    zhang        li      wang

 >>> longframe=pd.DataFrame({'color':['white','white','white','red','red','red','black','black','black'],'item':['ball','pen','mug','ball','pen','mug','ball','pen','mug'],'value':np.random.rand(9)})
 >>> longframe
    color  item     value        对冗余的消除，将longframe转换为wideframe
 0  white  ball  0.260358
 1  white   pen  0.543955
 2  white   mug  0.456874
 3    red  ball  0.967021
 4    red   pen  0.657271
 5    red   mug  0.984256
 6  black  ball  0.550236
 7  black   pen  0.731625
 8  black   mug  0.006728
 >>> wideframe=longframe.pivot('color','item')
 >>> wideframe
           value
 item       ball       mug       pen
 color
 black  0.550236  0.006728  0.731625
 red    0.967021  0.984256  0.657271
 white  0.260358  0.456874  0.543955
 >>> frame1
        addr  age   name
 0   beijing   12  zhang
 1  shanghai   12     li
 2   beijing   12   wang
 >>> del frame['addr']
 Traceback (most recent call last):
   File "<pyshell#103>", line 1, in <module>
     del frame['addr']
 NameError: name 'frame' is not defined
 >>> del frame1['addr']
 >>> frame1
    age   name
 0   12  zhang
 1   12     li
 2   12   wang