# Big accuracy gap with DDP

**URL:** <https://discuss.pytorch.org/t/big-accuracy-gap-with-ddp/160455>\
**Category:** distributed\
**Created:** [August 31, 2022, 9:58am UTC](https://discuss.pytorch.org/t/big-accuracy-gap-with-ddp/160455 "2022-08-31T09:58:06Z")\
**Posts on this page:** 1\
**Showing post:** 9

<div class="post-metadata">

**Author:** ![Pavel](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/pavel/32/61230_2.png) [@Pavel](https://discuss.pytorch.org/u/Pavel)\
**Post date:** [June 12, 2023, 10:39pm UTC](https://discuss.pytorch.org/t/big-accuracy-gap-with-ddp/160455/9 "2023-06-12T22:39:10Z")

</div>

Hi! I’ve faced the same issue. I’ve provided details on the solution in [this post](https://discuss.pytorch.org/t/distributeddataparallel-training-not-efficient/121624/11)

---

_[View the full topic](https://discuss.pytorch.org/t/big-accuracy-gap-with-ddp/160455)._