How to select rows based on conditions for each id

Question

I have the following data frame:

Hotel_id    Month_Year      Chef_Id  Chef_is_masterchef  Transition
2400188     February-2018   4597566     1                             


        

Answer 1

Use GroupBy.cumcount for counter per groups and then subtract number of 0 values by compare by 0 and GroupBy.transform:

s = df['Chef_is_masterchef'].eq(0).groupby(df['Chef_Id']).transform('sum')
df['var'] = df.groupby('Chef_Id').cumcount().sub(s)

---

print (df)
    Hotel_id      Month_Year  Chef_Id  Chef_is_masterchef  Transition  var
0    2400614        May-2015  2297544                   0           0   -8
1    2400614       June-2015  2297544                   0           0   -7
2    2400614       July-2015  2297544                   0           0   -6
3    2400614     August-2015  2297544                   0           0   -5
4    2400614  September-2015  2297544                   0           0   -4
5    2400614    October-2015  2297544                   0           0   -3
6    2400614   November-2015  2297544                   0           0   -2
7    2400614   December-2015  2297544                   0           0   -1
8    2400614    January-2016  2297544                   1           1    0
9    2400614   February-2016  2297544                   1           0    1
10   2400614      March-2016  2297544                   1           0    2
11   3400624        May-2016  2597531                   0           0   -3
12   3400624       June-2016  2597531                   0           0   -2
13   3400624       July-2016  2597531                   0           0   -1
14   3400624     August-2016  2597531                   1           1    0
15   2400133   February-2016  4597531                   0           0   -6
16   2400133      March-2016  4597531                   0           0   -5
17   2400133      April-2016  4597531                   0           0   -4
18   2400133        May-2016  4597531                   0           0   -3
19   2400133       June-2016  4597531                   0           0   -2
20   2400133       July-2016  4597531                   0           0   -1
21   2400133     August-2016  4597531                   1           1    0
22   2400133  September-2016  4597531                   1           0    1
23   2400133    October-2016  4597531                   1           0    2
24   2400133   November-2016  4597531                   1           0    3
25   2400133   December-2016  4597531                   1           0    4
26   2400133    January-2017  4597531                   1           0    5
27   2400133   February-2017  4597531                   1           0    6
28   2400133      March-2017  4597531                   1           0    7
29   2400133      April-2017  4597531                   1           0    8
30   2400133        May-2017  4597531                   1           0    9

Last filter by Series.between:

df1 = df[df['var'].between(-3, 2)]
print (df1)

df2 = df[df['var'].between(-6, 5)]
print (df2)