# Data scattering with DistributedDataParallel

**URL:** <https://discuss.pytorch.org/t/data-scattering-with-distributeddataparallel/89148>\
**Category:** distributed\
**Created:** [July 14, 2020, 12:46pm UTC](https://discuss.pytorch.org/t/data-scattering-with-distributeddataparallel/89148 "2020-07-14T12:46:52Z")\
**Posts on this page:** 1\
**Showing post:** 2

<div class="post-metadata">

**Author:** ![Anakin](https://discuss.pytorch.org/letter_avatar_proxy/v4/letter/a/eb8c5e/32.png) [@Anakin](https://discuss.pytorch.org/u/Anakin)\
**Post date:** [July 14, 2020, 1:59pm UTC](https://discuss.pytorch.org/t/data-scattering-with-distributeddataparallel/89148/2 "2020-07-14T13:59:21Z")

</div>

I made a typo for passing devices : instead of `torch.cuda.set_device(args.local_rank)`, I passed wrong parameter to `torch.cuda.set_device(range(2))`.

After fixing this typo, I still have the same problem as posted as [How to scatter list data on multiple GPUs](https://discuss.pytorch.org/t/running-model-on-multiple-gpus-runtimeerror-caught-runtimeerror-in-replica-0-on-device-0/88687/2)

Thanks for any inputs.

---

_[View the full topic](https://discuss.pytorch.org/t/data-scattering-with-distributeddataparallel/89148)._