Evelin Carvalho Freire de Amorim comments

Results 7 comments of


                                            Evelin Carvalho Freire de Amorim

Potential bug that causes IndexError in inference process

I changed the SRL code. However, I am not sure if the change is correct, so I reported [here](https://github.com/CogComp/SRL-English/issues/9#issue-1506139218) the issue and modification I made.

Different instance type in srl.py from utils and the SRL output

Thanks for the reply. I made some changes to SRL code. Since I am trying to run your code on top of a different dataset, I need to go through...

`AcceleratorState` object has no attribute `distributed_type`.

I am sorry I did not specify ```CorefTrainer```. I am using a custom trainer (you can check in this [link](https://github.com/evelinamorim/Seq2seqCoref/blob/main/trainer.py) ). This custom trainer is a subclass of the ```Seq2SeqTrainer```....

Segmentation fault (core dumped) when execute model.to("xpu")

@gujinghui , as @gekeleda said, the same result is produced no matter the order of the imports. @jgong5 I executed with gdb and the following output was produced. ``` Starting...

Segmentation fault (core dumped) when execute model.to("xpu")

Thanks for the quick reply! The backtrace is the following: https://gist.github.com/evelinamorim/40dce656614394bad491955b7bc274a9#file-error_intel_pytorch_extension

Segmentation fault (core dumped) when execute model.to("xpu")

@gujinghui Thanks again for the quick reply! The change of version of oneAPI Base Toolkit worked! Thanks!

Requirements txt is broken

Try to update your setuptools==45. I had a similar issue and it worked for me.