ValueError when running ./gpu_squad_base.sh

[Omnis23]

Hello, I am somewhat new to text embedding models. I have successfully ran sudo./prepro_squad.sh after some troubleshooting, and have moved on to the next step in the SQuAD2.0 training/testing. Unfortunately, I continue receiving this error and am having a hard time finding out whether or not it has to do with a .records file or something else that I am not familiar with at the moment?

And if it is some .records file, how should I go about setting that up properly?

Here is what the output is:

ValueError: Tensor conversion requested dtype string for Tensor with dtype float32: 'Tensor("arg0:0", shape=(), dtype=float32, device=/device:CPU:0)'

If someone knows how to troubleshoot this in some way, please let me know. Thank you!!

[YJ-007]

Did you fix this proplem？ I've got the same error...

[Omnis23]

Yeah I was still experiencing this issue..

On Wed, May 27, 2020 at 3:23 AM YJ-007 [email protected] wrote:

Did you fix this proplem？ I've got the same error...

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/zihangdai/xlnet/issues/260#issuecomment-634508624, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFTLRY776XZ5KJBFF5656DDRTTEZXANCNFSM4MCMH5PA .

[heeler-deer]

hey,I've got the same error...