۵ تابع مفید پانداز در پایتون — راهنمای کاربردی

1import pandas as pd
2import numpy as np
3df = pd.DataFrame({'DATE': [1, 2, 3, 4, 5],
4                   'VOLUME': [100, 200, 300,400,500],
5                   'PRICE': [214, 234, 253,272,291]})
6print(df)
7  
8    DATE  VOLUME  PRICE
90     1     100    214
101     2     200    234
112     3     300    253
123     4     400    272
134     5     500    291
14df.shift(1)
15DATE  VOLUME  PRICE
160   NaN     NaN    NaN
171   1.0   100.0  214.0
182   2.0   200.0  234.0
193   3.0   300.0  253.0
204   4.0   400.0  272.0
21# with fill_Value = 0
22df.shift(1,fill_value=0)
23DATE  VOLUME  PRICE
240     0       0      0
251     1     100    214
262     2     200    234
273     3     300    253
284     4     400    272

1df['PREV_DAY_PRICE'] = df['PRICE'].shift(1,fill_value=0)
2print(df)
3DATE  VOLUME  PRICE  PREV_DAY_PRICE
40     1     100    214               0
51     2     200    234             214
62     3     300    253             234
73     4     400    272             253
84     5     500    291             272

1df['LAST_3_DAYS_AVE_PRICE'] = (df['PRICE'].shift(1,fill_value=0) + 
2                               df['PRICE'].shift(2,fill_value=0) + 
3                               df['PRICE'].shift(3,fill_value=0))/3

1DATE  VOLUME  PRICE     LAST_3_DAYS_AVE_PRICE
20     1     100    214               0.000000
31     2     200    234              71.333333
42     3     300    253             149.333333
53     4     400    272             233.666667
64     5     500    291             253.000000

1DATE  VOLUME  PRICE     TOMORROW_PRICE
20     1     100    214             234
31     2     200    234             253
42     3     300    253             272
53     4     400    272             291
64     5     500    291               0

1a = pd.Index([3,3,4,2,1,3, 1, 2, 3, 4, np.nan,4,6,7])
2a.value_counts()
3#Output
43.0    4
54.0    3
61.0    2
72.0    2
87.0    1
96.0    1
10dtype: int64

1#In
2b = pd.Series(['ab','bc','cd',1,'cd','cd','bc','ab','bc',1,2,3,2,3,np.nan,1,np.nan])
3b.value_counts()
4#Out
5bc    3
6cd    3
71     3
83     2
9ab    2
102     2
11dtype: int64

1#In
2a = pd.Index([3,3,4,2,1,3, 1, 2, 3, 4, np.nan,4,6,7])
3a.value_counts(bins=4)
4#Out
5(2.5, 4.0]      7
6(0.993, 2.5]    4
7(5.5, 7.0]      2
8(4.0, 5.5]      0
9dtype: int64

1df = pd.DataFrame(np.arange(15).reshape(-1, 3), columns=['A', 'B','C'])
2print(df)
3#Out
4    A   B   C
50   0   1   2
61   3   4   5
72   6   7   8
83   9  10  11
94  12  13  14
10#mask operation to check if element is divided by 2 without any remainder. If match change the sign of the element as original
11df.mask(df % 2 == 0,-df)
12#Out
13A   B   C
140   0   1  -2
151   3  -4   5
162  -6   7  -8
173   9 -10  11
184 -12  13 -14

1import pandas as pd
2import numpy as np
3df = pd.DataFrame({'HEIGHT': [170,78,99,160,160,130,155,70,70,20],
4                   'WEIGHT': [50,60,70,80,90,90,90,50,60,70]},
5                   index=['A','B','C','D','E','F','G','H','I','J'])
6print(df)
7HEIGHT  WEIGHT
8A     170      50
9B      78      60
10C      99      70
11D     160      80
12E     160      90
13F     130      90
14G     155      90
15H      70      50
16I      70      60
17J      20      70
18dfl = df.nlargest(3,'HEIGHT')
19print(dfl)
20HEIGHT  WEIGHT
21A     170      50
22D     160      80
23E     160      90

1dfl = df.nlargest(2,'HEIGHT',keep='all')
2print(dfl)
3HEIGHT  WEIGHT
4A     170      50
5D     160      80
6E     160      90

1dfl = df.nlargest(2,'HEIGHT',keep='last')
2print(dfl)
3HEIGHT  WEIGHT
4A     170      50
5E     160      90

1dfl = df.nlargest(2,'HEIGHT',keep='first')
2print(dfl)
3HEIGHT  WEIGHT
4A     170      50
5D     160      80

1import pandas as pd
2import numpy as np
3df = pd.DataFrame({'HEIGHT': [170,78,99,160,160,130,155,70,70,20],
4                   'WEIGHT': [50,60,70,80,90,90,90,50,60,70]},
5                   index=['A','B','C','D','E','F','G','H','I','J'])
6print(df)
7HEIGHT  WEIGHT
8A     170      50
9B      78      60
10C      99      70
11D     160      80
12E     160      90
13F     130      90
14G     155      90
15H      70      50
16I      70      60
17J      20      70
18dfs = df.nsmallest(3,'WEIGHT')
19print(dfs)
20HEIGHT  WEIGHT
21A     170      50
22H      70      50
23B      78      60