# Unbelievable Error: Help ME!

**URL:** <https://discuss.huggingface.co/t/unbelievable-error-help-me/129971>\
**Category:** 🤗Transformers\
**Created:** [December 8, 2024, 12:11pm UTC](https://discuss.huggingface.co/t/unbelievable-error-help-me/129971 "2024-12-08T12:11:11Z")\
**Posts on this page:** 1\
**Showing post:** 2

<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:** [December 8, 2024, 12:38pm UTC](https://discuss.huggingface.co/t/unbelievable-error-help-me/129971/2 "2024-12-08T12:38:18Z")

</div>

There seem to be several patterns… I hope it’s not a new bug.

> [@ValueError: Trainer: evaluation requires an eval\_dataset](https://discuss.huggingface.co/t/valueerror-trainer-evaluation-requires-an-eval-dataset/40670):
>
> I’m trying to do a finetuning without an evaluation dataset. For that, I’m using the following code: training\_args = TrainingArguments( output\_dir=resume\_from\_checkpoint, evaluation\_strategy="epoch", per\_device\_train\_batch\_size=1, ) def compute\_metrics(pred: EvalPrediction): labels = pred.label\_ids preds = pred.predictions.argmax(-1) f1 = f1\_score(labels, preds, average="weighted") acc = accuracy\_score(labels, preds, average="…

> You can avoid this error by not specifying `evaluation_strategy` in `TrainingArguments`.

> <https://stackoverflow.com/questions/76310533/how-to-fix-trainer-evaluation-requires-an-eval-dataset-in-huggingface-transfo>

---

_[View the full topic](https://discuss.huggingface.co/t/unbelievable-error-help-me/129971)._