CODEWITHSUNDEEP
pythonpandas
1974sb
1974sb

Reputation: 69

Data frame transformation

I am very novice Python pandas user so this may be a dumb question and pandas may have simple method to do this operation

My data frame is as follows:

Name        Activity      Date
Abc.           Run.           June 3
Cde.           Walk.         Sept. 4
Abc.           Run.           June 4
Abc.           Run.           June 5
Code.         Walk.          June 3
Cde.            Run.           Sept 5

I want to convert it to

Name  run freq  walk freq 
Abc.       3.             0
Cde.        1.             1
Code.      0.            1

How do I do this in pandas?

Upvotes: 1

Views: 47

Answers (2)

Alexander
Alexander

Reputation: 109526

Group on the column, get the count of each, and then unstack the results. 

df = pd.DataFrame({'Name': ['Abc.', 'Cde.', 'Abc.', 'Abc.', 'Code.', 'Cde.'], 
                   'Activity': ['Run.', 'Walk.', 'Run.', 'Run.', 'Walk.', 'Run.'], 
                   'Date': ['June 3', 'Sept. 4', 'June 4', 'June 5', 'June 3', 'Sept 5']})

>>> df.groupby(['Name', 'Activity']).count().unstack()

          Date       
Activity  Run.  Walk.
Name                 
Abc.         3    NaN
Cde.         1      1
Code.      NaN      1

Upvotes: 1

Def_Os
Def_Os

Reputation: 5467

You want to group by name and activity and get the sum:

df.groupby(['Name', 'Activity']).sum()

Upvotes: 1

Related Questions