CODEWITHSUNDEEP
pythonpandas
AlliDeacon
AlliDeacon

Reputation: 1495

Pandas divide by value not giving correct result

I have a pandas dataframe 'pivoted' : 

Reason                CE CS DG DR IC  IO    
Warehouse Month                                                                
01        01           9  4  4  0  1   8

I also have a variable that accumulates the total of the issues. It's an "int" type variable. In this case it is 626

When I run the following line: total_percentages = pivoted/total_issues

I'm not getting the correct (or expected) results:

01        01     0.021143  0.009397  0.009397  0.000000  0.002349  0.018793

I would expect this:

01        01     0.014376  0.006389 0.006389  0.00000   0.001597  0.012780

My full code:

issue_df = pd.read_sql(issue_query, cnxn)
issue_df.rename(columns={'00001' : 'Invoices', 'OBWHID' : 'Warehouse', 'OBRTRC':'Reason', 'INV_MONTH':'Month', '00005':'Date'}, inplace=True)


pivoted = pd.pivot_table(issue_df, index=["Warehouse", "Month"], values=["Invoices"], columns=['Reason'], aggfunc='count', fill_value=0)

pivoted.loc['Column Total'] = pivoted.sum()
print(pivoted.dtypes)

#Percentages of Warehouse Returns by Month
warehouse_percentages = pivoted[:] = 100 * pivoted[:].div(pivoted[:].sum(axis=1), axis=0)
print(warehouse_percentages)
print(total_issues)
total_percentages = pivoted.div(total_issues)`

Upvotes: 1

Views: 339

Answers (1)

miradulo
miradulo

Reputation: 29710

With the line 

warehouse_percentages = pivoted[:] = 100 * pivoted[:].div(pivoted[:].sum(axis=1), axis=0)

you are re-assigning all of the values of pivoted to be the result of the right hand side of the operation, which is why your expected output isn't matching what you print pivoted as before the line above.

Thus, if you don't intend on modifying pivoted with this operation, remove pivoted[:] - I'm not sure if you think that without the copy you'll modify pivoted but it is not necessary - Pandas operations almost always by default do not operate in place, but return another object. 

warehouse_percentages = 100* pivoted.div(pivoted.sum(axis=1), axis=0)

Upvotes: 2

Related Questions