# Unbalanced data in multiple object detection

**URL:** <https://discuss.pytorch.org/t/unbalanced-data-in-multiple-object-detection/88104>\
**Category:** vision\
**Created:** [July 6, 2020, 12:06pm UTC](https://discuss.pytorch.org/t/unbalanced-data-in-multiple-object-detection/88104 "2020-07-06T12:06:17Z")\
**Posts on this page:** 1\
**Showing post:** 2

<div class="post-metadata">

**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\
**Post date:** [July 7, 2020, 8:11am UTC](https://discuss.pytorch.org/t/unbalanced-data-in-multiple-object-detection/88104/2 "2020-07-07T08:11:11Z")

</div>

Balancing a mutli-label dataset is a bit tricky, as a simple weighted sampling might even increase the imbalance, e.g. if the majority classes are often associated with the oversampled minority classes.

There are a few approaches, such as [SCUMBLE, REMEDIAL](https://arxiv.org/pdf/1802.05033.pdf), [MLSOL](https://ecmlpkdd2019.org/downloads/paper/624.pdf), MLSMOTE etc.

I don’t know, what the current state of the art is, but I’m sure @rasbt would know. 😉

---

_[View the full topic](https://discuss.pytorch.org/t/unbalanced-data-in-multiple-object-detection/88104)._