runtimeerror

[sdws258]

python 3.7
pytorch 1.10.1+cu111
gcc 5.4.0

I have modified the cloneable.h file according to the FAQs section, but I still encounter the following error when the program is running. Moreover, I have tried to run this code under gcc==7.5.0, the same error appears. Please tell me how can i fix it?

[sdws258]

[hzhwcmhf]

@sdws258 Try removing this line

DA-Transformer/fs_plugins/custom_ops/dag_loss.py

Line 54 in 8ef9c8f

extra_include_paths=[os.path.join(module_path, "../../cub")],

If it works well, I will update a patch to fix the issue

[bbo0924]

@sdws258 Try removing this line

DA-Transformer/fs_plugins/custom_ops/dag_loss.py

Line 54 in 8ef9c8f

extra_include_paths=[os.path.join(module_path, "../../cub")],

If it works well, I will update a patch to fix the issue

It looks good to me

[sdws258]

@sdws258 Try removing this line

DA-Transformer/fs_plugins/custom_ops/dag_loss.py

Line 54 in 8ef9c8f

extra_include_paths=[os.path.join(module_path, "../../cub")],

If it works well, I will update a patch to fix the issue

yeah, it works

[sdws258]

I'm sorry that I have another question:
I run the DAG on IWSLT14 ENDE raw data, it appears one warning and causes one error:

Moreover, i find there is no result of IWSLT14 in DAG paper. Please tell me the result of it and how to fix the above problem in DAG.

[hzhwcmhf]

max-source-positions and max-target-positions specify the max length of the samples. You should set it according to your dataset.
If you want to train with a sample whose target length is 132, max-target-positions should be set at least \lambda * 132, i.e., 1056 or larger if lambda=8.
Moreover, dropping some examples does not satisfy the length limitation is normal behavior (your 1st screenshot). But we usually don't drop valid or test samples for fair evaluation (2nd screenshot).

We do not have an official result on IWSLT14 for now. It will be appreciated if you can train a model and tell us your result.

[myaxxxxx]

Hi, I also encountered the same question, and I solved it by using "python /path/to/fairseq_cli/dagprocess.py" and "python /path/to/fairseq_cli/train.py". I recommend you to use this command.