MycChiu

I see, thanks for the answer. I actually managed to achieve this with `tf.dynamic_partition` on the batch index and then use the results to gather different branches' tensors, but since...

I see, thanks for the suggestions, should I close this issue for now then?

@ronghanghu Yeah, that's essentially what I am trying to do now, but adding the requirement of BPTT to the mix makes this a lot more complicated. *EDIT:* also, do you...

@ronghanghu Have you tried putting the dropout mask in a variable as well? Use that variable as the mask throughout the process, then just assign a new one in the...

Okay, I have played a bit with `sess.partial_run`, and the main advantage is that we can do conditional branching in the middle with python code, but every tensor can only...

well...after some profiling, I realized that `tf.dynamic_stitch` actually doesn't have GPU implementation, so the slowness was actually due to `tf.dynamic_stitch` constantly transferring all states back and forth between CPU and...

Yes, I just ran the test again with the newly compiled .so file, and it worked fine. I am really new to c/c++, so I am not sure about this,...

@cloofa Hi, sorry for the long delay, I didn't realize my chrome logged me out of github. Your make output seems alright, but I wonder which Nvidia card are you...

@NickShahML were you able to build with `make` without errors? or were you using the `.so` files provided with the repo? The only time I ran into this issue while...

@NickShahML @cloofa , I think I found the culprit! The latest commit should (hopefully) solve the undefined symbol error.