Pandas: create pivot table with duplicate index and show information (no counts, just data)

Question

I have the following dataframe:

    ID    KEY   TYPE  FIELD   VALUE
0   MYID  112233  1     Height  117
1   MYID  112233  1     Weight  17
2   MYID  445566  1     Height  95
3   MYID  445566  1     Weight  39
4   MYID  112233  2     Height  118
5   MYID  112233  2     Weight  17

This has duplicate KEY values, and I need to keep the ones with the highest TYPE. The intended result is:

   KEY    TYPE Height  Weight
0  112233 2    118     17
1  445566 1    95      39

I tried accomplishing this using pd.pivot(index='KEY', columns='FIELD')['VALUE'] but it complains because I have duplicate values in KEY. I also tried using pd.pivot_table() but this is inadequate for this problem as I don't have to aggregate data, just show them.

How could I solve this?