Just curious about how will pyro deal with the data from and into recurrent neural network? Is there any typical example or tutorials?
unfortunately, none that are up to date since pyro is moving so fast . feel free to contribute one though! do you have any specific questions on your mind?
In pytorch's RNN modules, the default input and output tensors are provided as seq first, as (seq, batch, feature). Will there be some special tricks to make such a data form work in pyro?
it depends on what you want to do; check out the dmm tutorial for one example.