Closed zihanlalala closed 1 year ago
@Zettelkasten maybe can help.
Can you show your modified config? Maybe upload it to a Gist and link here.
Yes, here is the full config: https://gist.github.com/zihanlalala/9883b3efaad0739bcf58ffd049e6e7f2
I see that you use reuse_params
and you try to share params between source embed, target embed and target output. This only works when you have the same source and target vocab. Remove those reuse_params
usages.
Thank you! It's the problem. By the way, do returnn supports tensorflow2.0 and above?
By the way, do returnn supports tensorflow2.0 and above?
Yes, it should be no problem.
Thanks.
Hi! I am running translation task with the provided script of 2022-lsh-attention, but get an error. Here is a brief error message.
It seems that the code tries to assert 42235 == 32881, where 42235 is the source vocab size and 32881 is the target vocab size. I think I might set the wrong config but have no idea what is wrong. Here is my config about vocab.
I wonder is there anything wrong with my config? I have followed the setup of 2022-lsh-attention and only replaced the dataset and vocab setting in the provided script.
Thank you.