How to groupby pandas dataframe and sum values in another column-CodePudding

I have a pandas dataframe with 3 columns (CHAR, VALUE, and WEIGHT).

CHAR column contains duplicate values which I need to group ['A', 'A', 'A', 'B', 'B', 'C'].
VALUE column has a unique value for every unique CHAR [10, 10, 10, 15, 15, 20].
WEIGHT column has various values [1, 2, 1, 4, 4, 6].

Consider an example of my initial dataframe:

I need to create a new dataframe which will have 3 columns.

CHAR which will not have any duplicates
T_VALUE (total value) which will have a sum of this CHAR's value and all its weights
T_WEIGHT (total weight) which will have a sum of this CHAR's weights

Result would look like this:

I would highly appreciate any help.

CodePudding user response：

You could use = instead:

newDF = df.groupby(['CHAR', 'VALUE'], as_index=False)['WEIGHT'].sum()
newDF['VALUE']  = newDF['WEIGHT']

CodePudding user response：

I was actually able to answer my own question. Here is the solution:

d = {'CHAR': ['A', 'A', 'A', 'B', 'B', 'C'],
     'VALUE': [10, 10, 10, 15, 15, 20], 
     'WEIGHT':  [1, 2, 1, 4, 4, 6]}
df = pandas.DataFrame(data=d)

newDF = df.groupby(['CHAR', 'VALUE'], as_index=False)['WEIGHT'].sum()
newDF['VALUE'] = newDF['VALUE']   newDF['WEIGHT']