numpy&pandas補充經常使用示例
Numpy
【數組切片】html
In [115]: a = np.arange(12).reshape((3,4))
In [116]: a
Out[116]:
array([[ 0, 1, 2, 3],
[ 4, 5, 6, 7],
[ 8, 9, 10, 11]])
In [117]: a[1:,1:3]
Out[117]:
array([[ 5, 6],
[ 9, 10]])
In [118]:
【布爾值索引】找出數據中大於5的元素python
In [134]: a = [ random.randint(0,10) for i in range(20) ]
In [135]: a
Out[135]: [4, 4, 9, 2, 2, 5, 6, 4, 3, 9, 5, 7, 10, 4, 9, 10, 6, 10, 3, 8]
In [136]: _
Out[136]: [4, 4, 9, 2, 2, 5, 6, 4, 3, 9, 5, 7, 10, 4, 9, 10, 6, 10, 3, 8]
In [137]: a = np.array(a)
In [138]: a
Out[138]:
array([ 4, 4, 9, 2, 2, 5, 6, 4, 3, 9, 5, 7, 10, 4, 9, 10, 6,
10, 3, 8])
In [139]: a>5
Out[139]:
array([False, False, True, False, False, False, True, False, False,
True, False, True, True, False, True, True, True, True,
False, True])
In [140]: a[a>5]
Out[140]: array([ 9, 6, 9, 7, 10, 9, 10, 6, 10, 8])
In [141]:
【布爾值索引】sql
- 找出數組中大於5的偶數元素json
- 找出數組中大於5的書 或 偶數vim
In [143]: b = a[a>5] In [144]: b Out[144]: array([ 9, 6, 9, 7, 10, 9, 10, 6, 10, 8]) In [145]: b[b%2==0] Out[145]: array([ 6, 10, 10, 6, 10, 8]) In [146]: a[(a>5) & (a%2==0)] Out[146]: array([ 6, 10, 10, 6, 10, 8]) In [147]: In [147]: a[(a>5) and (a%2==0)] --------------------------------------------------------------------------- ValueError Traceback (most recent call last) <ipython-input-147-fee222ad41de> in <module> ----> 1 a[(a>5) and (a%2==0)] ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all() In [148]: In [148]: a[(a>5) | (a%2==0)] Out[148]: array([ 4, 4, 9, 2, 2, 6, 4, 9, 7, 10, 4, 9, 10, 6, 10, 8]) In [149]: a[(a>5) or (a%2==0)] --------------------------------------------------------------------------- ValueError Traceback (most recent call last) <ipython-input-149-732531989282> in <module> ----> 1 a[(a>5) or (a%2==0)] ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all() In [150]:
【花式索引（索引位置無規律）】數組
# 一位數組花式索引示例
In [153]: a = np.arange(10,20)
In [154]: a
Out[154]: array([10, 11, 12, 13, 14, 15, 16, 17, 18, 19])
In [155]: a[[1,3,4,9]]
Out[155]: array([11, 13, 14, 19])
In [156]:
# 二維數組花式索引示例一（取第二行的第二列和第三列）
In [165]: a
Out[165]:
array([[ 0, 1, 2, 3, 4],
[ 5, 6, 7, 8, 9],
[10, 11, 12, 13, 14],
[15, 16, 17, 18, 19]])
In [166]: a[2,[2,3]]
Out[166]: array([12, 13])
In [167]:
# 二維數組花式索引示例二 （行：取第一行和第三行，列：取第一列和第三列）
In [167]: a
Out[167]:
array([[ 0, 1, 2, 3, 4],
[ 5, 6, 7, 8, 9],
[10, 11, 12, 13, 14],
[15, 16, 17, 18, 19]])
In [168]: a[[1,3],[2,4]] # 注意：在對行和列同時使用花式索引時，解釋效果不同
Out[168]: array([ 7, 19])
In [169]: a[[1,3],:]
Out[169]:
array([[ 5, 6, 7, 8, 9],
[15, 16, 17, 18, 19]])
In [170]: a[[1,3],:][:,[2,4]]
Out[170]:
array([[ 7, 9],
[17, 19]])
In [171]:
【數值取整問題】dom
| 向0取整(int) | 四捨五入(round) | 向上取整(math.ceil) | 向下取整(math.floor) | |
| 1.7 | 1 | 2 | 2 | 1 |
| -1.7 | -1 | -2 | -1 | -2 |
| 1.3 | 1 | 1 | 2 | 1 |
| -1.3 | -1 | -1 | -1 | -2 |
In [53]: a = np.arange(-5.5,5.5) In [54]: a Out[54]: array([-5.5, -4.5, -3.5, -2.5, -1.5, -0.5, 0.5, 1.5, 2.5, 3.5, 4.5]) In [55]: np.trunc(a) # numpy向0取整 Out[55]: array([-5., -4., -3., -2., -1., -0., 0., 1., 2., 3., 4.]) In [56]: np.round(a) # numpy四捨五入取整 Out[56]: array([-6., -4., -4., -2., -2., -0., 0., 2., 2., 4., 4.]) In [57]: np.rint(a) # numpy.rint() 等價於 numpy.round() Out[57]: array([-6., -4., -4., -2., -2., -0., 0., 2., 2., 4., 4.]) In [58]: np.ceil(a) # numpy向上取整 Out[58]: array([-5., -4., -3., -2., -1., -0., 1., 2., 3., 4., 5.]) In [59]: np.floor(a) # numpy向下取整 Out[59]: array([-6., -5., -4., -3., -2., -1., 0., 1., 2., 3., 4.]) In [60]:
numpy.modf()：把整數和小數分開ide
In [62]: a Out[62]: array([-5.5, -4.5, -3.5, -2.5, -1.5, -0.5, 0.5, 1.5, 2.5, 3.5, 4.5]) In [63]: x,y = np.modf(a) In [64]: x Out[64]: array([-0.5, -0.5, -0.5, -0.5, -0.5, -0.5, 0.5, 0.5, 0.5, 0.5, 0.5]) In [65]: y Out[65]: array([-5., -4., -3., -2., -1., -0., 0., 1., 2., 3., 4.]) In [66]: x+y Out[66]: array([-5.5, -4.5, -3.5, -2.5, -1.5, -0.5, 0.5, 1.5, 2.5, 3.5, 4.5]) In [67]:
numpy.nanexcel
In [85]: np.nan == np.nan Out[85]: False In [86]: np.nan is np.nan Out[86]: True In [87]:
In [77]: a = np.arange(5) In [78]: b = a/a /usr/bin/ipython3:1: RuntimeWarning: invalid value encountered in true_divide #!/usr/local/python3.6/bin/python3.6 In [79]: b Out[79]: array([nan, 1., 1., 1., 1.]) In [80]: np.*nan? np.isnan np.nan In [81]: np.isnan(b) Out[81]: array([ True, False, False, False, False]) In [82]: b[np.isnan(b)] Out[82]: array([nan]) In [83]: b[~np.isnan(b)] # 取反 Out[83]: array([1., 1., 1., 1.]) In [84]:
numpy.infcode
In [97]: np.inf == np.inf Out[97]: True In [98]: np.inf is np.inf Out[98]: True In [99]:
In [89]: a = np.arange(3,8) In [90]: a Out[90]: array([3, 4, 5, 6, 7]) In [91]: b = [1,0,1,0,1] In [92]: c = a/b /usr/bin/ipython3:1: RuntimeWarning: divide by zero encountered in true_divide #!/usr/local/python3.6/bin/python3.6 In [93]: c Out[93]: array([ 3., inf, 5., inf, 7.]) In [94]: c[c!=np.inf] Out[94]: array([3., 5., 7.]) In [96]: c[~np.isinf(c)] Out[96]: array([3., 5., 7.]) In [97]:
numpy.maximum 和 numpy.minimum
In [102]: a Out[102]: array([3, 4, 5, 6, 7]) In [103]: b Out[103]: array([2, 5, 3, 7, 4]) In [104]: np.maximum(a,b) Out[104]: array([3, 5, 5, 7, 7]) In [105]: np.minimum(a,b) Out[105]: array([2, 4, 3, 6, 4]) In [106]:
Pandas
series整數索引問題，推薦多使用iloc
In [137]: s1 =pd.Series(np.arange(10)) In [138]: s2 = s1[5:].copy() In [139]: s1 Out[139]: 0 0 1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9 dtype: int64 In [140]: s2 Out[140]: 5 5 6 6 7 7 8 8 9 9 dtype: int64 In [141]: In [147]: s2[5] Out[147]: 5 In [148]: s2.loc[5] Out[148]: 5 In [149]: s2.iloc[0] Out[149]: 5 In [150]:
series數據對齊
In [150]: a = pd.Series([12,23,34],['c','a','d']) In [151]: b = pd.Series([11,30,9],['d','c','a']) In [152]: a+b Out[152]: a 32 c 42 d 45 dtype: int64 In [153]:
In [153]: b = pd.Series([11,30,9,100],['d','c','a','b']) In [154]: a+b Out[154]: a 32.0 b NaN # 缺失值 c 42.0 d 45.0 dtype: float64 In [155]: a.add(b, fill_value=0) Out[155]: a 32.0 b 100.0 c 42.0 d 45.0 dtype: float64 In [156]:
series缺失值處理
In [158]: s = a+b In [159]: s Out[159]: a 32.0 b NaN c 42.0 d 45.0 dtype: float64 In [160]: s.isnull() Out[160]: a False b True c False d False dtype: bool In [161]: s.notnull() Out[161]: a True b False c True d True dtype: bool In [162]:
# 扔掉缺失值 In [162]: s.dropna() Out[162]: a 32.0 c 42.0 d 45.0 dtype: float64 In [163]: # 設置缺失值 In [163]: s.fillna(0) Out[163]: a 32.0 b 0.0 c 42.0 d 45.0 dtype: float64 In [164]:
# 設置缺失值（均值） In [166]: s.fillna(s.mean()) Out[166]: a 32.000000 b 39.666667 c 42.000000 d 45.000000 dtype: float64 In [167]:
DataFrame建立示例
In [169]: pd.DataFrame({'one':[1,2,3],'two':[10,20,30]})
Out[169]:
one two
0 1 10
1 2 20
2 3 30
In [170]: pd.DataFrame({'one':[1,2,3],'two':[10,20,30]},index=['A','B','C'])
Out[170]:
one two
A 1 10
B 2 20
C 3 30
In [171]: df = _
In [172]: df
Out[172]:
one two
A 1 10
B 2 20
C 3 30
In [173]: pd.DataFrame({'one':pd.Series([1,2,3],index=['A','B','C']),'two':pd.Series([10,20,30,40],index=['B','A','C','D'])})
Out[173]:
one two
A 1.0 20
B 2.0 10
C 3.0 30
D NaN 40
In [174]:
讀文件建立DataFrame
In [181]: !vim demo.csv
In [182]: !cat demo.csv
'one','two','three'
'A',1,2,3
'B',4,5,6
'C',7,8,9
In [183]: pd.read_csv('demo.csv')
Out[183]:
'one' 'two' 'three'
'A' 1 2 3
'B' 4 5 6
'C' 7 8 9
In [184]:
DateFrame索引問題：獲取單個元素
In [218]: df Out[218]: one two A 1.0 20 B 2.0 10 C 3.0 30 D NaN 40 In [219]: df['two']['C'] Out[219]: 30 In [220]: df.two.C Out[220]: 30 In [221]: df.loc['C','two'] Out[221]: 30 In [222]: df.iloc[2,1] Out[222]: 30 In [223]:
DateFrame索引問題：獲取多個元素
In [234]: df Out[234]: one two A 1.0 20 B 2.0 10 C 3.0 30 D NaN 40 In [235]: df['one'] Out[235]: A 1.0 B 2.0 C 3.0 D NaN Name: one, dtype: float64 In [236]: df.loc['B'] Out[236]: one 2.0 two 10.0 Name: B, dtype: float64 In [237]: df.loc['B',] Out[237]: one 2.0 two 10.0 Name: B, dtype: float64 In [238]: df.loc['B',:] Out[238]: one 2.0 two 10.0 Name: B, dtype: float64 In [239]: In [239]: df.iloc[1] Out[239]: one 2.0 two 10.0 Name: B, dtype: float64 In [240]: df.iloc[1,] Out[240]: one 2.0 two 10.0 Name: B, dtype: float64 In [241]: df.iloc[1,:] Out[241]: one 2.0 two 10.0 Name: B, dtype: float64 In [242]: df.iloc[1,:1] Out[242]: one 2.0 Name: B, dtype: float64 In [243]:
DataFrame數據對齊
In [243]: df
Out[243]:
one two
A 1.0 20
B 2.0 10
C 3.0 30
D NaN 40
In [244]: df = pd.DataFrame({'two':[1,2,3,4],'one':[4,5,6,7]},index=['C','D','B','A'])
In [245]: df2 = _243
In [246]: df
Out[246]:
two one
C 1 4
D 2 5
B 3 6
A 4 7
In [247]: df2
Out[247]:
one two
A 1.0 20
B 2.0 10
C 3.0 30
D NaN 40
In [248]: df+df2
Out[248]:
one two
A 8.0 24
B 8.0 13
C 7.0 31
D NaN 42
In [249]:
DataFrame缺失值處理
In [268]: df Out[268]: one two A 1.0 20.0 B 2.0 10.0 C 3.0 30.0 D NaN 40.0 In [269]: df.fillna(0) Out[269]: one two A 1.0 20.0 B 2.0 10.0 C 3.0 30.0 D 0.0 40.0 In [270]: df.dropna() Out[270]: one two A 1.0 20.0 B 2.0 10.0 C 3.0 30.0 In [271]: df.loc['D','two'] = np.nan In [272]: df.loc['B','two'] = np.nan In [273]: df Out[273]: one two A 1.0 20.0 B 2.0 NaN C 3.0 30.0 D NaN NaN In [274]: df.dropna() # 刪除含有NaN的行 Out[274]: one two A 1.0 20.0 C 3.0 30.0 In [275]: df.dropna(how='all') # 整行都是NaN才刪除該行 Out[275]: one two A 1.0 20.0 B 2.0 NaN C 3.0 30.0 In [276]: df.dropna(how='any') # 默認 Out[276]: one two A 1.0 20.0 C 3.0 30.0 In [277]:
axis指定刪除整列
In [282]: df Out[282]: one two A 1.0 20.0 B 2.0 10.0 C 3.0 30.0 D 4.0 10.0 In [283]: df.iloc[2,1] = np.nan In [284]: df Out[284]: one two A 1.0 20.0 B 2.0 10.0 C 3.0 NaN D 4.0 10.0 In [285]: df.dropna(axis=1) # 刪除含有NaN的列 Out[285]: one A 1.0 B 2.0 C 3.0 D 4.0 In [286]: df.dropna(axis=0) # 默認 Out[286]: one two A 1.0 20.0 B 2.0 10.0 D 4.0 10.0 In [287]:
DataFrame排序
# 按值排序 In [17]: df Out[17]: two one C 1 4.0 D 2 NaN B 3 6.0 A 4 7.0 In [18]: df.mean() Out[18]: two 2.500000 one 5.666667 dtype: float64 In [19]: df.mean(axis=1) Out[19]: C 2.5 D 2.0 B 4.5 A 5.5 dtype: float64 In [20]: df.sort_values(by='one') Out[20]: two one C 1 4.0 B 3 6.0 A 4 7.0 D 2 NaN In [21]: df.sort_values(by='one',ascending=False) Out[21]: two one A 4 7.0 B 3 6.0 C 1 4.0 D 2 NaN In [22]: In [23]: df.sort_values(by='B',axis=1) Out[23]: two one C 1 4.0 D 2 NaN B 3 6.0 A 4 7.0 In [24]: df.sort_values(by='B',axis=1,ascending=False) Out[24]: one two C 4.0 1 D NaN 2 B 6.0 3 A 7.0 4 In [25]: # 按標籤排序 In [68]: df Out[68]: two one C 1 4.0 D 2 NaN B 3 6.0 A 4 7.0 In [69]: df2 Out[69]: four C 50 A 60 D 70 B 80 In [70]: df3 = pd.concat([df, df2], axis=1, join_axes=[df.index]) # concat合併df In [71]: df3 Out[71]: two one four C 1 4.0 50 D 2 NaN 70 B 3 6.0 80 A 4 7.0 60 In [72]: df3.sort_index() Out[72]: two one four A 4 7.0 60 B 3 6.0 80 C 1 4.0 50 D 2 NaN 70 In [73]: df3.sort_index(ascending=False) Out[73]: two one four D 2 NaN 70 C 1 4.0 50 B 3 6.0 80 A 4 7.0 60 In [74]: df3.sort_index(ascending=False, axis=1) Out[74]: two one four C 1 4.0 50 D 2 NaN 70 B 3 6.0 80 A 4 7.0 60 In [75]: df3.sort_index(axis=1) Out[75]: four one two C 50 4.0 1 D 70 NaN 2 B 80 6.0 3 A 60 7.0 4 In [76]:
pandas批量解析時間對象
In [83]: pd.to_datetime(['2001-01-01','2010-Apr-09','02/04/2019','2019/02/03']) Out[83]: DatetimeIndex(['2001-01-01', '2010-04-09', '2019-02-04', '2019-02-03'], dtype='datetime64[ns]', freq=None) In [84]:
pandas生成時間對象
In [90]: pd.date_range?
In [91]: pd.date_range(start='2018-01-01',end='2018-02-01')
Out[91]:
DatetimeIndex(['2018-01-01', '2018-01-02', '2018-01-03', '2018-01-04',
'2018-01-05', '2018-01-06', '2018-01-07', '2018-01-08',
'2018-01-09', '2018-01-10', '2018-01-11', '2018-01-12',
'2018-01-13', '2018-01-14', '2018-01-15', '2018-01-16',
'2018-01-17', '2018-01-18', '2018-01-19', '2018-01-20',
'2018-01-21', '2018-01-22', '2018-01-23', '2018-01-24',
'2018-01-25', '2018-01-26', '2018-01-27', '2018-01-28',
'2018-01-29', '2018-01-30', '2018-01-31', '2018-02-01'],
dtype='datetime64[ns]', freq='D')
In [92]: pd.date_range(start='2018-01-01',periods=30)
Out[92]:
DatetimeIndex(['2018-01-01', '2018-01-02', '2018-01-03', '2018-01-04',
'2018-01-05', '2018-01-06', '2018-01-07', '2018-01-08',
'2018-01-09', '2018-01-10', '2018-01-11', '2018-01-12',
'2018-01-13', '2018-01-14', '2018-01-15', '2018-01-16',
'2018-01-17', '2018-01-18', '2018-01-19', '2018-01-20',
'2018-01-21', '2018-01-22', '2018-01-23', '2018-01-24',
'2018-01-25', '2018-01-26', '2018-01-27', '2018-01-28',
'2018-01-29', '2018-01-30'],
dtype='datetime64[ns]', freq='D')
In [93]:
In [96]: pd.date_range(start='2018-01-01',periods=30,freq='H')
Out[96]:
DatetimeIndex(['2018-01-01 00:00:00', '2018-01-01 01:00:00',
'2018-01-01 02:00:00', '2018-01-01 03:00:00',
'2018-01-01 04:00:00', '2018-01-01 05:00:00',
'2018-01-01 06:00:00', '2018-01-01 07:00:00',
'2018-01-01 08:00:00', '2018-01-01 09:00:00',
'2018-01-01 10:00:00', '2018-01-01 11:00:00',
'2018-01-01 12:00:00', '2018-01-01 13:00:00',
'2018-01-01 14:00:00', '2018-01-01 15:00:00',
'2018-01-01 16:00:00', '2018-01-01 17:00:00',
'2018-01-01 18:00:00', '2018-01-01 19:00:00',
'2018-01-01 20:00:00', '2018-01-01 21:00:00',
'2018-01-01 22:00:00', '2018-01-01 23:00:00',
'2018-01-02 00:00:00', '2018-01-02 01:00:00',
'2018-01-02 02:00:00', '2018-01-02 03:00:00',
'2018-01-02 04:00:00', '2018-01-02 05:00:00'],
dtype='datetime64[ns]', freq='H')
In [97]: pd.date_range(start='2018-01-01',periods=30,freq='W')
Out[97]:
DatetimeIndex(['2018-01-07', '2018-01-14', '2018-01-21', '2018-01-28',
'2018-02-04', '2018-02-11', '2018-02-18', '2018-02-25',
'2018-03-04', '2018-03-11', '2018-03-18', '2018-03-25',
'2018-04-01', '2018-04-08', '2018-04-15', '2018-04-22',
'2018-04-29', '2018-05-06', '2018-05-13', '2018-05-20',
'2018-05-27', '2018-06-03', '2018-06-10', '2018-06-17',
'2018-06-24', '2018-07-01', '2018-07-08', '2018-07-15',
'2018-07-22', '2018-07-29'],
dtype='datetime64[ns]', freq='W-SUN')
In [98]: pd.date_range(start='2018-01-01',periods=30,freq='W-MON')
Out[98]:
DatetimeIndex(['2018-01-01', '2018-01-08', '2018-01-15', '2018-01-22',
'2018-01-29', '2018-02-05', '2018-02-12', '2018-02-19',
'2018-02-26', '2018-03-05', '2018-03-12', '2018-03-19',
'2018-03-26', '2018-04-02', '2018-04-09', '2018-04-16',
'2018-04-23', '2018-04-30', '2018-05-07', '2018-05-14',
'2018-05-21', '2018-05-28', '2018-06-04', '2018-06-11',
'2018-06-18', '2018-06-25', '2018-07-02', '2018-07-09',
'2018-07-16', '2018-07-23'],
dtype='datetime64[ns]', freq='W-MON')
In [99]:
In [99]: pd.date_range(start='2018-01-01',periods=30,freq='B')
Out[99]:
DatetimeIndex(['2018-01-01', '2018-01-02', '2018-01-03', '2018-01-04',
'2018-01-05', '2018-01-08', '2018-01-09', '2018-01-10',
'2018-01-11', '2018-01-12', '2018-01-15', '2018-01-16',
'2018-01-17', '2018-01-18', '2018-01-19', '2018-01-22',
'2018-01-23', '2018-01-24', '2018-01-25', '2018-01-26',
'2018-01-29', '2018-01-30', '2018-01-31', '2018-02-01',
'2018-02-02', '2018-02-05', '2018-02-06', '2018-02-07',
'2018-02-08', '2018-02-09'],
dtype='datetime64[ns]', freq='B')
In [100]: dt = _
In [101]: type(dt)
Out[101]: pandas.core.indexes.datetimes.DatetimeIndex
In [102]: dt[0]
Out[102]: Timestamp('2018-01-01 00:00:00', freq='B')
In [105]: dt[0].to_pydatetime()
Out[105]: datetime.datetime(2018, 1, 1, 0, 0)
In [106]:
時間間隔指定很是靈活
In [107]: pd.date_range(start='2018-01-01',periods=30,freq='1h20min')
Out[107]:
DatetimeIndex(['2018-01-01 00:00:00', '2018-01-01 01:20:00',
'2018-01-01 02:40:00', '2018-01-01 04:00:00',
'2018-01-01 05:20:00', '2018-01-01 06:40:00',
'2018-01-01 08:00:00', '2018-01-01 09:20:00',
'2018-01-01 10:40:00', '2018-01-01 12:00:00',
'2018-01-01 13:20:00', '2018-01-01 14:40:00',
'2018-01-01 16:00:00', '2018-01-01 17:20:00',
'2018-01-01 18:40:00', '2018-01-01 20:00:00',
'2018-01-01 21:20:00', '2018-01-01 22:40:00',
'2018-01-02 00:00:00', '2018-01-02 01:20:00',
'2018-01-02 02:40:00', '2018-01-02 04:00:00',
'2018-01-02 05:20:00', '2018-01-02 06:40:00',
'2018-01-02 08:00:00', '2018-01-02 09:20:00',
'2018-01-02 10:40:00', '2018-01-02 12:00:00',
'2018-01-02 13:20:00', '2018-01-02 14:40:00'],
dtype='datetime64[ns]', freq='80T')
In [108]:
pandas時間序列切片/截取
In [121]: series = pd.Series(np.arange(1000),index=pd.date_range(start='2017-01-01',periods=1000))
In [122]: series
Out[122]:
2017-01-01 0
2017-01-02 1
2017-01-03 2
2017-01-04 3
2017-01-05 4
2017-01-06 5
2017-01-07 6
2017-01-08 7
2017-01-09 8
2017-01-10 9
2017-01-11 10
2017-01-12 11
2017-01-13 12
2017-01-14 13
2017-01-15 14
2017-01-16 15
2017-01-17 16
2017-01-18 17
2017-01-19 18
2017-01-20 19
2017-01-21 20
2017-01-22 21
2017-01-23 22
2017-01-24 23
2017-01-25 24
2017-01-26 25
2017-01-27 26
2017-01-28 27
2017-01-29 28
2017-01-30 29
...
2019-08-29 970
2019-08-30 971
2019-08-31 972
2019-09-01 973
2019-09-02 974
2019-09-03 975
2019-09-04 976
2019-09-05 977
2019-09-06 978
2019-09-07 979
2019-09-08 980
2019-09-09 981
2019-09-10 982
2019-09-11 983
2019-09-12 984
2019-09-13 985
2019-09-14 986
2019-09-15 987
2019-09-16 988
2019-09-17 989
2019-09-18 990
2019-09-19 991
2019-09-20 992
2019-09-21 993
2019-09-22 994
2019-09-23 995
2019-09-24 996
2019-09-25 997
2019-09-26 998
2019-09-27 999
Freq: D, Length: 1000, dtype: int64
In [123]:
In [123]: series.index
Out[123]:
DatetimeIndex(['2017-01-01', '2017-01-02', '2017-01-03', '2017-01-04',
'2017-01-05', '2017-01-06', '2017-01-07', '2017-01-08',
'2017-01-09', '2017-01-10',
...
'2019-09-18', '2019-09-19', '2019-09-20', '2019-09-21',
'2019-09-22', '2019-09-23', '2019-09-24', '2019-09-25',
'2019-09-26', '2019-09-27'],
dtype='datetime64[ns]', length=1000, freq='D')
In [124]: series.head()
Out[124]:
2017-01-01 0
2017-01-02 1
2017-01-03 2
2017-01-04 3
2017-01-05 4
Freq: D, dtype: int64
In [125]: series.tail()
Out[125]:
2019-09-23 995
2019-09-24 996
2019-09-25 997
2019-09-26 998
2019-09-27 999
Freq: D, dtype: int64
In [126]: series['2018-03']
Out[126]:
2018-03-01 424
2018-03-02 425
2018-03-03 426
2018-03-04 427
2018-03-05 428
2018-03-06 429
2018-03-07 430
2018-03-08 431
2018-03-09 432
2018-03-10 433
2018-03-11 434
2018-03-12 435
2018-03-13 436
2018-03-14 437
2018-03-15 438
2018-03-16 439
2018-03-17 440
2018-03-18 441
2018-03-19 442
2018-03-20 443
2018-03-21 444
2018-03-22 445
2018-03-23 446
2018-03-24 447
2018-03-25 448
2018-03-26 449
2018-03-27 450
2018-03-28 451
2018-03-29 452
2018-03-30 453
2018-03-31 454
Freq: D, dtype: int64
In [128]: series['2018-12':'2019-01-10']
Out[128]:
2018-12-01 699
2018-12-02 700
2018-12-03 701
2018-12-04 702
2018-12-05 703
2018-12-06 704
2018-12-07 705
2018-12-08 706
2018-12-09 707
2018-12-10 708
2018-12-11 709
2018-12-12 710
2018-12-13 711
2018-12-14 712
2018-12-15 713
2018-12-16 714
2018-12-17 715
2018-12-18 716
2018-12-19 717
2018-12-20 718
2018-12-21 719
2018-12-22 720
2018-12-23 721
2018-12-24 722
2018-12-25 723
2018-12-26 724
2018-12-27 725
2018-12-28 726
2018-12-29 727
2018-12-30 728
2018-12-31 729
2019-01-01 730
2019-01-02 731
2019-01-03 732
2019-01-04 733
2019-01-05 734
2019-01-06 735
2019-01-07 736
2019-01-08 737
2019-01-09 738
2019-01-10 739
Freq: D, dtype: int64
In [129]:
resample從新採樣並作計算
In [132]: series.resample('M').sum()
Out[132]:
2017-01-31 465
2017-02-28 1246
2017-03-31 2294
2017-04-30 3135
2017-05-31 4185
2017-06-30 4965
2017-07-31 6076
2017-08-31 7037
2017-09-30 7725
2017-10-31 8928
2017-11-30 9555
2017-12-31 10819
2018-01-31 11780
2018-02-28 11466
2018-03-31 13609
2018-04-30 14085
2018-05-31 15500
2018-06-30 15915
2018-07-31 17391
2018-08-31 18352
2018-09-30 18675
2018-10-31 20243
2018-11-30 20505
2018-12-31 22134
2019-01-31 23095
2019-02-28 21686
2019-03-31 24924
2019-04-30 25035
2019-05-31 26815
2019-06-30 26865
2019-07-31 28706
2019-08-31 29667
2019-09-30 26622
Freq: M, dtype: int64
In [133]: series.resample('M').mean()
Out[133]:
2017-01-31 15.0
2017-02-28 44.5
2017-03-31 74.0
2017-04-30 104.5
2017-05-31 135.0
2017-06-30 165.5
2017-07-31 196.0
2017-08-31 227.0
2017-09-30 257.5
2017-10-31 288.0
2017-11-30 318.5
2017-12-31 349.0
2018-01-31 380.0
2018-02-28 409.5
2018-03-31 439.0
2018-04-30 469.5
2018-05-31 500.0
2018-06-30 530.5
2018-07-31 561.0
2018-08-31 592.0
2018-09-30 622.5
2018-10-31 653.0
2018-11-30 683.5
2018-12-31 714.0
2019-01-31 745.0
2019-02-28 774.5
2019-03-31 804.0
2019-04-30 834.5
2019-05-31 865.0
2019-06-30 895.5
2019-07-31 926.0
2019-08-31 957.0
2019-09-30 986.0
Freq: M, dtype: float64
In [134]:
pandas文件讀取
In [14]: pd.read_csv('601318.csv', index_col='date', parse_dates=['date'])
Out[14]:
Unnamed: 0 open close high low volume code
date
2007-03-01 0 21.878 20.473 22.302 20.040 1977633.51 601318
2007-03-02 1 20.565 20.307 20.758 20.075 425048.32 601318
2007-03-05 2 20.119 19.419 20.202 19.047 419196.74 601318
2007-03-06 3 19.253 19.800 20.128 19.143 297727.88 601318
2007-03-07 4 19.817 20.338 20.522 19.651 287463.78 601318
2007-03-08 5 20.171 20.093 20.272 19.988 130983.83 601318
2007-03-09 6 20.084 19.922 20.171 19.559 160887.79 601318
2007-03-12 7 19.821 19.563 19.821 19.471 145353.06 601318
2007-03-13 8 19.607 19.642 19.804 19.524 102319.68 601318
2007-03-14 9 19.384 19.664 19.734 19.161 173306.56 601318
2007-03-15 10 19.918 19.673 20.342 19.603 152521.90 601318
2007-03-16 11 19.686 19.782 20.106 19.428 227547.24 601318
2007-03-20 12 20.478 20.031 20.530 19.909 222026.87 601318
2007-03-21 13 20.040 19.734 20.128 19.646 136728.32 601318
2007-03-22 14 19.887 19.848 20.093 19.791 167509.84 601318
2007-03-23 15 19.839 19.760 19.922 19.563 139810.14 601318
2007-03-26 16 19.778 20.101 20.215 19.769 223266.79 601318
2007-03-27 17 20.036 20.088 20.285 19.966 139338.19 601318
2007-03-28 18 20.084 20.382 20.522 19.944 258263.69 601318
2007-03-29 19 20.482 20.740 21.349 20.338 461986.18 601318
2007-03-30 20 20.548 20.587 20.946 20.443 144617.20 601318
2007-04-02 21 20.587 21.174 21.309 20.587 231445.03 601318
2007-04-03 22 21.187 21.095 21.335 20.959 132712.04 601318
2007-04-04 23 21.099 20.911 21.222 20.806 122454.69 601318
2007-04-05 24 20.915 20.968 21.003 20.653 122865.38 601318
2007-04-06 25 20.863 21.007 21.419 20.784 195208.52 601318
2007-04-09 26 21.042 22.582 22.705 20.872 462770.21 601318
2007-04-10 27 22.316 23.112 23.488 22.316 407823.90 601318
2007-04-11 28 23.138 23.427 24.145 23.016 243446.50 601318
2007-04-12 29 23.619 23.383 25.378 23.169 159270.43 601318
... ... ... ... ... ... ... ...
2017-11-06 2533 64.690 64.010 64.700 62.920 908570.00 601318
2017-11-07 2534 64.300 65.370 66.570 64.300 1173565.00 601318
2017-11-08 2535 65.400 64.610 66.350 64.320 867820.00 601318
2017-11-09 2536 64.500 66.330 66.390 64.400 708669.00 601318
2017-11-10 2537 66.000 69.890 69.950 65.930 1254060.00 601318
2017-11-13 2538 70.100 70.150 70.570 69.480 752207.00 601318
2017-11-14 2539 70.690 70.420 71.290 69.770 801748.00 601318
2017-11-15 2540 69.980 69.200 70.430 68.590 1009459.00 601318
2017-11-16 2541 68.800 73.010 73.110 68.750 1163764.00 601318
2017-11-17 2542 72.700 75.270 75.320 71.800 1580393.00 601318
2017-11-20 2543 74.780 75.710 76.490 74.070 1141281.00 601318
2017-11-21 2544 75.130 78.440 79.680 75.130 1445569.00 601318
2017-11-22 2545 79.500 77.450 79.960 76.580 1293487.00 601318
2017-11-23 2546 76.600 74.320 78.440 73.700 1576210.00 601318
2017-11-24 2547 74.150 74.620 75.460 72.710 1317843.00 601318
2017-11-27 2548 74.700 73.550 74.900 71.550 1637232.00 601318
2017-11-28 2549 72.700 72.730 73.540 71.880 786469.00 601318
2017-11-29 2550 73.540 72.420 74.190 71.260 875004.00 601318
2017-11-30 2551 71.370 69.920 71.670 69.550 1163733.00 601318
2017-12-01 2552 69.650 68.100 70.180 67.910 1393046.00 601318
2017-12-04 2553 67.600 69.390 70.350 67.370 1159283.00 601318
2017-12-05 2554 68.900 71.200 71.500 68.780 1692539.00 601318
2017-12-06 2555 70.900 69.400 71.100 68.000 1245607.00 601318
2017-12-07 2556 69.350 68.640 69.810 67.600 859703.00 601318
2017-12-08 2557 68.940 71.490 71.860 68.660 1095632.00 601318
2017-12-11 2558 71.200 73.250 73.310 70.820 1139927.00 601318
2017-12-12 2559 73.250 71.210 73.560 71.170 777900.00 601318
2017-12-13 2560 71.210 72.120 72.620 70.200 865117.00 601318
2017-12-14 2561 72.120 71.010 72.160 70.600 676186.00 601318
2017-12-15 2562 70.690 70.380 71.440 70.050 735547.00 601318
[2563 rows x 7 columns]
In [15]: df = _
In [16]: df.index
Out[16]:
DatetimeIndex(['2007-03-01', '2007-03-02', '2007-03-05', '2007-03-06',
'2007-03-07', '2007-03-08', '2007-03-09', '2007-03-12',
'2007-03-13', '2007-03-14',
...
'2017-12-04', '2017-12-05', '2017-12-06', '2017-12-07',
'2017-12-08', '2017-12-11', '2017-12-12', '2017-12-13',
'2017-12-14', '2017-12-15'],
dtype='datetime64[ns]', name='date', length=2563, freq=None)
In [17]:
若是csv文件沒有首行（列名）
pd.read_csv('601318.csv', header=None, names=['A','B','C','D','E','F','G','H'])
若是csv文件裏有一些缺失的值，好比有的是NaN，有的是None，那麼如何讓pandas正確解釋？
na_values參數指定列表，即列表裏的字符串都會被解釋成numpy.nan
pd.read_csv('601318.csv', na_values=['None','none','nan','NaN'])
In [8]: pd.read_*? pd.read_clipboard pd.read_csv pd.read_excel pd.read_feather pd.read_fwf pd.read_gbq pd.read_hdf pd.read_html pd.read_json pd.read_msgpack pd.read_parquet pd.read_pickle pd.read_sas pd.read_sql pd.read_sql_query pd.read_sql_table pd.read_stata pd.read_table In [9]:
相關文章
- 1. 經常使用Windows命令、經常使用 Cmd命令（補充）
- 2. python經常使用標準庫 --補充
- 3. 數組經常使用方法補充
- 4. xshell經常使用命令(待補充)
- 5. html總結（經常使用插件，補充知識，經典例題）
- 6. Hibernate 補充 ManyToOne、OneToMany、OneToOne的使用例
- 7. SQLalchemy使用補充
- 8. Linux find經常使用用法示例
- 9. mysql中經常使用sql（持續補充中）
- 10. python經常使用模塊補充hashlib configparser logging，subprocess模塊
- 更多相關文章...
- • 使用 CSS 顯示 XML - XML 教程
- • 使用 XSLT 顯示 XML - XML 教程
- • Composer 安裝與使用
- • 常用的分佈式事務解決方案
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
最新文章
歡迎關注本站公眾號,獲取更多信息
相關文章