How to build a statement to perform a groupby operation during runtime on a Pandas DataFrame?

user_15384702 0
Viewed 36

I have a Pandas DataFrame dfs and a list headers

The list headers is assigned the column names of the DataFrame dfs during runtime.

For ex, let us consider the list gets assigned with dfs's column names as:

["Information_type", "Interface", "Type_of_Interface", "Connection_Mechanism"]

I want to perform the below groupby and agg operation on the DataFrame dfs without explicitly mentioning the column names in the groupby operation i.e. "Information_type": " ".join , "Interface": " ".join , "Type_of_Interface": " ".join , "Connection_Mechanism": " ".join:

dfs[0]=dfs[0].groupby("grp").agg({"Information_type": " ".join, "Interface": " ".join, "Type_of_Interface": " ".join, "Connection_Mechanism": " ".join})

Basically write "Information_type": " ".join , "Interface": " ".join , "Type_of_Interface": " ".join , "Connection_Mechanism": " ".join to the above line in runtime.

It would be great if such a thing would be possible, else I would have to manually edit the column names and execute the groupby and agg operation for each table!

Appreciate your help. Thanks in advance!

pandas dynamic-programming dataframe python
1 Answers

IIUC this is what you want:

#setup
df = pd.DataFrame({'a':np.random.randint(0,5,25),
                   'b':np.random.randint(0,5,25),
                   'c':np.random.randint(0,5,25), 
                   'd':np.random.randint(0,5,25)}, dtype = str)

cols = ['b','c']

df.groupby('a').agg({col: " ".join for col in cols})

Output

               b              c
a                              
0  0 0 3 3 4 2 3  3 3 4 0 4 3 2
1      2 4 1 2 1      3 0 2 1 3
2        0 0 4 2        1 3 1 3
3    2 2 4 1 3 0    3 1 1 1 2 0
4          4 2 0          2 0 3
user_137432330
Related