[FIXED] Aggregating a String Field as a Record Indepdent of other Aggregations in GroupBy

Issue

I have data that looks like df1

Store	owner
Store1	A
Store1	B
Store2	B
Store2	C

and df2

Store	Sold_Item	Price
Store1	Apple	2
Store1	Orange	3
Store1	Apple	2
Store2	Apple	3
Store	Banana	2

I would like to get something like

Store	Owners	Num_Sales	Revenue
Store1	A	3	7
	B
Store2	B	2	5
	C

I have tried something like:

df = df1.merge(df2, how='inner', on='Store')
df_stats = df.groupby(['Store']).agg({"Sold_Item": np.count, "Price": 'sum'})

but I cannot what to do with the Owner field.

Any advice?

Solution

Shouldn't the last store of df2 be "Store2 ?

out = (
    df1.merge(df2, on="Store", how="inner")
        .groupby(["Store", "owner"], as_index=False)
        .agg({"Sold_Item": "count", "Price": "sum"})  
)

out.loc[out["Store"].duplicated(), out.columns.difference(["owner"])] = None

# out.to_excel("file.xlsx", index=False) # or `to_csv`

Output (in Excel) :

Store	owner	Sold_Item	Price
Store1	A	3	7
	B
Store2	B	2	5
	C

Answered By - Timeless