# Backprop Through Sparse Tensor Is Not Memory Efficient?

**URL:** <https://discuss.pytorch.org/t/backprop-through-sparse-tensor-is-not-memory-efficient/66409>\
**Category:** Uncategorized\
**Created:** [January 11, 2020, 10:32pm UTC](https://discuss.pytorch.org/t/backprop-through-sparse-tensor-is-not-memory-efficient/66409 "2020-01-11T22:32:39Z")\
**Posts on this page:** 1\
**Showing post:** 3

<div class="post-metadata">

**Author:** ![Kevin96](https://discuss.pytorch.org/letter_avatar_proxy/v4/letter/k/e5b9ba/32.png) [@Kevin96](https://discuss.pytorch.org/u/Kevin96)\
**Post date:** [January 14, 2020, 5:34am UTC](https://discuss.pytorch.org/t/backprop-through-sparse-tensor-is-not-memory-efficient/66409/3 "2020-01-14T05:34:34Z")

</div>

Thanks for clarification!

Hopefully, in my case, the backprop operation is quite simple. I have implemented an customized backprop operation to solve the memory issue.

---

_[View the full topic](https://discuss.pytorch.org/t/backprop-through-sparse-tensor-is-not-memory-efficient/66409)._