xiao03 / nh Goto Github PK

Dear Shaw Liu,

Thanks for the code and for making it available.
I wanted to ask you about a problem I noticed in the model implementation, in the CTLSTM class, in the forward function you have written self.init_states(batch_size) at the start of every time step. I think it should be written out of this loop (before it). Because the states should be initialized once at the start of feeding the sequence. Thanks to let me know if I am wrong.

Thank you very much for the implementation of the Hawks process with pytorch. It would be better if you had some documentation to help understand it.

Thank you for you release code with pytorch implement. I have a question of you data process in utils.py.
In pad_bos function, why you add 1 in create pad_event_seqs and pad_time_seqs, what are added elements mean? And why * type_size in create pad_event_seqs.