how to generate catalog files?

Hi,

Thank you for your work! It's inspiring and interesting!

1. Would you mind releasing the codes about generating catalogs for each dataset? 
2. What is the value of N when you conduct k-means clustering? As described in your suppl., N=200 is only for visualizing. 

Thank you for any help!
