Dong Qian

Hi, thanks for your reply. We have to use kl annealing to alleviate kl vanishing issue in this model. I think it is the same. best

Hi Zekang, Thanks for providing the source code. I just followed this post. Suppose I don't want to use the older version of `transformers`, due to python environment issue, then...

> Follow up question, the variable "responses" in "generate.py" ln: 293. What is the purpose of it? Hi Evram, Based on my understanding, the evaluation task is to generate a...

> > Hi Zekang, > > Thanks for providing the source code. > > I just followed this post. Suppose I don't want to use the older version of `transformers`,...

> > > Follow up question, the variable "responses" in "generate.py" ln: 293. What is the purpose of it? > > > > > > Hi Evram, > > Based...

Hi, I am trying to reproduce this model as the baseline and curious about basic network architectures applied in the experiment. p(f) is from N(0, I) -- q(f | x_{1:T})...

Hi @dblueeye, thank you for pointing out. You mean we can still improve the implementation for SMC?