So I have an external system that I get some data from. It comes in this format:
+----+-----------+------------+-------+
| id | sessionId | element | value |
+----+-----------+------------+-------+
| 1 | 0 | extrainfo1 | 11 |
| 1 | 0 | extrainfo2 | 12 |
| 2 | 0 | extrainfo1 | 21 |
| 2 | 0 | extrainfo2 | 22 |
| 2 | 1 | extrainfo1 | 31 |
| 2 | 1 | extrainfo2 | 32 |
+----+-----------+------------+-------+
And I need it in this format:
+----+-----------+------------+------------+
| id | sessionId | extrainfo1 | extrainfo2 |
+----+-----------+------------+------------+
| 1 | 0 | 11 | 12 |
| 2 | 0 | 21 | 22 |
| 2 | 1 | 31 | 32 |
+----+-----------+------------+------------+
I have tried
sf_pivot = sessionfields.pivot(index='id', columns=['sessionId','element'], values='value')
But that returns this:
+-----------+------------+------------+------------+------------+
| sessionId | 0 | | 1 | |
+-----------+------------+------------+------------+------------+
| element | extrainfo1 | extrainfo2 | extrainfo1 | extrainfo2 |
| id | | | | |
| 1 | 11 | 12 | | |
| 2 | 21 | 22 | 31 | 32 |
+-----------+------------+------------+------------+------------+
I tried to unstack that but didn't seem to work.
To have sessionId as a value in index or value when doing the pivot doesn't seem to work either, I get duplicate error on that.
