# Handling Imbalanced Dataset

**URL:** <https://discuss.huggingface.co/t/handling-imbalanced-dataset/93080>\
**Category:** 🤗Datasets\
**Created:** [June 20, 2024, 12:14am UTC](https://discuss.huggingface.co/t/handling-imbalanced-dataset/93080 "2024-06-20T00:14:25Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![AmeerHesham](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/ameerhesham/32/28407_2.png) [@AmeerHesham](https://discuss.huggingface.co/u/AmeerHesham)\
**Post date:** [June 20, 2024, 12:14am UTC](https://discuss.huggingface.co/t/handling-imbalanced-dataset/93080/1 "2024-06-20T00:14:25Z")

</div>

I am trying to build an audio classification model by fine-tuning the distil-hubert model on a dataset called ( Nooon/Donate\_a\_cry ) after I finished, I noticed that the accuracy doesn’t get any better, and I think it’s because of the imbalanced data.  
So, my question is how to handle this situation because I am new to hugging face and thanks in advance 🥰