Shamane Siri

Oh I get it. So if we use dynamic RNN encoder without bucketing dynamic rnn function will padded the input to a fixed size right?

I cannot over-fit the model for the lighweight data set. I think it's more to do with the dialog text file. If I enter few question and answer line by...

What do you mean by reversed inputs ? Is that about the input sequence ? But I 'v red in a paper that by keeping fixed size inputs we can...

So it's like there will be no information passing with the padding when we get it as (In-Pd-De)? Because 0 symbols don't pass anything ? Yeah then it's correct.

I am interested to work on this. Can we use RAG-end2end ?

+1 GitHub - shamanez On Sat, Nov 20, 2021, 23:30 RM ***@***.***> wrote: > +1 > > — > You are receiving this because you commented. > Reply to this...

@Sharathnasa I got this. I am using 1080 Ti . 11GB

@akanimax your codes are perfect. I even followed the VDB code which is very efficient and informative.

@albertchristianto what about the accuracy? It is working or just converging?

@albertchristianto check whether he used something like weight clipping or gradient norm. Because for me this cord is error free algorithmic wise.