I have a dataframe that looks like this:
| ID | Company_Id | value | Approve or Reject |
|---|---|---|---|
| 1A | 3412asd | value-1 | Approve |
| 2B | 2345tyu | value-2 | Approve |
| 3C | 9800bvd | value-3 | Approve |
| 2B | 2345tyu | value-1 | Approve |
Note that ID can repeat with different 'value'. ID, Company_ID are indices.
Now I need the output to be:
| ID | Company_Id | value-1 | value-2 | value-3 |
|---|---|---|---|---|
| 1A | 3412asd | Approve | NULL | NULL |
| 2B | 2345tyu | Approve | Approve | NULL |
| 3C | 9800bvd | NULL | NULL | Approve |
CodePudding user response：
pyspark pivot
df.groupBy('ID', 'Company_Id').pivot('value').agg(first('Approve or Reject')).show()