How does COPOD handle missing values and categorical variables?

Hi @winstonll, @yzhao062:

I am new to COPOD and have a few questions:

1. Can COPOD handle missing values or do I have to impute the data first?
2. How does COPOD handle categorical variables?
3. Is there any way to pull the dimensional outlier graph as figure 2 in the paper via pyod? If not, what is the best way to get the 'variable importance' for each outlier?

Thank you!