如何确定每个人每次都住在不同城市的地方？

Question

Here is a small set of the dataset that I am currently working on.

FirstName  LastName   cities       occupation         time
---------------------------------------------------------------
---------------------------------------------------------------
Alice      Oumi       Queens       software engineer  1/1/2019
Alice      Oumi       New York     software engineer  12/3/2018
Sam        Charles    Santa Clara  Engineer           2/5/2017
Sam        Charles    Santa Monica Engineer           8/9/2018
Sam        Charles    Santa Clara  Engineer           12/12/2019
Alice      Oumi       New York     software engineer  1/2/2017

As you see above, the same person could be living in a same place but for a different duration of a time. I want to make clean this dataset that should what places did Alice and Sam live. For example, instead of having 2 rows of Alice living in New York, I only need to have one. Something similar to the following table

FirstName  LastName   cities         FirstTime    SecondTime
---------------------------------------------------------------
---------------------------------------------------------------
Alice      Oumi       Queens         1/1/2019     NA
Alice      Oumi       New York       1/2/2017     12/3/2018
Sam        Charles    Santa Clara    2/5/2017     12/12/2019
Sam        Charles    Santa Monica   8/9/2018     NA 

I am kinda new to python and trying to learn. but i have tried to use for loops using iterrows() but didn't work. What can use to achieve this table?

Thank you so much in advance

Answer 1

You can do that as follows:

# number the times a person lived in the same city (with the same occupation)
df['sequence']= df.groupby(['FirstName', 'LastName', 'cities', 'occupation']).cumcount()+1
# now create the "pivot" table
result= df.set_index(['FirstName', 'LastName', 'cities', 'occupation', 'sequence']).unstack()
# rename the columns
result.columns= ['FirstTime', 'SecondTime']

# reset the index (it was just needed for "pivoting"
result.reset_index(inplace=True)

The result looks like:

Out[483]: 
  FirstName LastName                 cities         occupation  FirstTime  SecondTime
0     Alice     Oumi               New York  software engineer  12/3/2018    1/2/2017
1     Alice     Oumi                 Queens  software engineer   1/1/2019         NaN
2       Sam  Charles            Santa Clara           Engineer   2/5/2017  12/12/2019
3       Sam  Charles  Santa Monica Engineer           8/9/2018       None         NaN