# Argument was ignored by\`Longformer\` model, stsb task

**URL:** <https://discuss.huggingface.co/t/argument-was-ignored-by-longformer-model-stsb-task/110191>\
**Category:** Beginners\
**Created:** [October 5, 2024, 1:43am UTC](https://discuss.huggingface.co/t/argument-was-ignored-by-longformer-model-stsb-task/110191 "2024-10-05T01:43:54Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![averyyu99](https://avatars.discourse-cdn.com/v4/letter/a/4da419/32.png) [@averyyu99](https://discuss.huggingface.co/u/averyyu99)\
**Post date:** [October 5, 2024, 1:43am UTC](https://discuss.huggingface.co/t/argument-was-ignored-by-longformer-model-stsb-task/110191/1 "2024-10-05T01:43:54Z")

</div>

Dear Community, I am wondering if someone could help me with the ‘warnings’ that I encountered while using the run\_glue.py. My model is “allenai/longformer-base-4096”, and the task name is “stsb”, and it said :

“The following columns in the test set don’t have a corresponding argument in `LongformerForSequenceClassification.forward` and have been ignored: idx, sentence1, sentence2. If idx, sentence1, sentence2 are not expected by `LongformerForSequenceClassification.forward`, you can safely ignore this message.”

My training dataset are json files that has idx, sentence1, sentence2 and a label(converted to float and treated as float). If those field were not used, then why are my model is running? Any insights or suggesions is greatly appreciated! Thank you.

---

<div class="post-metadata">

**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](https://discuss.huggingface.co/u/John6666)\
**Post date:** [October 5, 2024, 5:09am UTC](https://discuss.huggingface.co/t/argument-was-ignored-by-longformer-model-stsb-task/110191/2 "2024-10-05T05:09:22Z")

</div>

When the software is working properly, the warnings are only displayed to give users and developers a clue as to the cause of errors or undesired results.  
If it works fine, ignore it.  
If not, the software stops with an error. Of course, there are cases where broken data can be obtained, but these are rare nowadays.

In this case, I think it is safe to ignore it. There was a similar case in a past post.

> [@The following columns in the training set don't have a corresponding argument](https://discuss.huggingface.co/t/the-following-columns-in-the-training-set-dont-have-a-corresponding-argument/23624):
>
> Hello, I am trying to fine tune a T5 model using xsum dataset but I am getting the following. Should I just change the sames “summary, document and id” to something else? Or that’s due to something else? The following columns in the training set don’t have a corresponding argument in T5ForConditionalGeneration.forward and have been ignored: summary, document, id. Many thanks!