How to unload a Jar for UDF without restart spark context?

vonjack
New Contributor II

In the scala notebook of databricks, I created a temporary function with a certain Jar and class name. 

Then I want to update the Jar. But without restart the context, I can not reload the new Jar, the temporary function always reuses the old classes.

I noticed that for normal function it support refresh, but temporary function doesn't.

Does anyone know the way to resolve this issue?