AWS DMS migration task Postgres > Redshift problem with NaN

Hi, I am trying to migrate a table from Postgres to Redshift using a migration task

Simplified table structure:

the avg values sometimes is a NaN in Postgres I created a transformation rule : where schema name is like 'public' and Source table name is like 'my_table', and Column name is like 'avg', change-data-type > real8

This fails and in the stl_load_errors i can see error because of "+nan" values. (screenshot attached)

(Edit : To be clear in the Postgres database the value is NaN not +nan)

If I replace the datatype in my transformation rule from real8 to string it works.

real8 is supposed(?) to handle NaN so I'm not sure what I am missing here.

Any help is greatly appreciated! Thank you!