Python 中的数据透视表（Pandas）

Question

I'm very new to python (using pandas). Kindly help.

There are two columns in my dataframe - weight conversion (float) and sales_units (int). I simply want to group by (sum) the sales_units by weight_conversion.

Sample Data

weight_conversion   sales_units
0.1                   1
0.1                   2
50                    100
50                    200
96.1                  20
314.4                  2
500                   100
500                   200

            

I have tried two ways:

1. GROUP BY IN PANDAS :

df.groupby(['weight_conversion'])['Sales_Unit'].sum()

2. PIVOT TABLE IN PANDAS:

   df.pivot_table(index = 'weight_conversion', values='Sales_Unit', aggfunc ='sum')

Required Output: I need a simple pivot table where I have rows as weight_conversion along with sum of sales units.

The output I'm getting in Python Pandas is as follows (so weird): weight_conversion

0       3300000000000000000000000000034000000000000000...
0.1     0000100001000000000000001000000020050000000000...
0.2     0000000000000000000000000000001000000001100000...
0.3                           000000000000000000000300000
0.4     0000000000100000000000000000000000000000000001...
                            ...                        
90                                        000000000102009
92      0000200011000000000000010001000000000000000020...
92.1                            0000001000000000000000003
96      2000000000000000000000000000000000001100000000...
96.1    0000000000000000000000000000000000000000000000...

Name: Sales_Unit, Length: 96, dtype: object

sample output

weight_conversion   sales_units
0.1                   3
50                    300
96.1                  20
314.4                  2
500                   300

Please help.**

Answer 1

I dont think you need PIVOT table for the sample output you have given. Below worked as for the required output you mentioned

 df = pd.DataFrame({"weight_conversion":[0.1,0.1,50,50,96.1,314.4,500,500], "sales_units":[1,2,100,200,20,2,100,200]}) df.groupby('weight_conversion').agg({'sales_units':'sum'}).reset_index() Output: