gudovskiy / autodo Goto Github PK

Hi! Thanks for your inspiring work.

I am confused about this description in your paper. It's true that we simultaneously optimize conditional distribution and joint distribution, but I think using asymmetric KL loss is ok because the optimizations are in separate steps. Could you please give a further explanation to help me understand it better?

Thanks a lot!

Hi,
thanks for sharing your implementation. I have some questions about it:

1. Does it also work on tabular data?
2. Is the code tailored to the datasets used in the paper or can one apply it to any data?
3. Is it possible to identify the noisy instances (return the noisy IDs or the clean set)?

Thanks!