Separate discrete and continuous variables

I know how to separate numerical and categorical data as follows:

num_data = [cname for cname in df.columns if df[cname].dtypes == 'object']
cat_data = [cname for cname in df.columns if df[cname].dtypes in ['int64', 'float64']]

Now I want to separate my numerical variables into discrete and continuous. How do I do that?