# When are MARL replay buffers zero padded?

**URL:** <https://discuss.ray.io/t/when-are-marl-replay-buffers-zero-padded/3763>\
**Category:** RLlib\
**Created:** [October 8, 2021, 10:31pm UTC](https://discuss.ray.io/t/when-are-marl-replay-buffers-zero-padded/3763 "2021-10-08T22:31:36Z")\
**Posts on this page:** 1\
**Showing post:** 8

<div class="post-metadata">

**Author:** ![Aceticia](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/aceticia/32/1506_2.png) [@Aceticia](https://discuss.ray.io/u/Aceticia)\
**Post date:** [October 12, 2021, 2:19am UTC](https://discuss.ray.io/t/when-are-marl-replay-buffers-zero-padded/3763/8 "2021-10-12T02:19:59Z")

</div>

@mannyv Hi, sorry about the spamming, I’ve found my misunderstanding about the code. Normally tuple spaces are handled by flattening, but I disabled my preprocessor because of [this previous issue](https://discuss.ray.io/t/unpack-obs-doesnt-know-to-expect-1-hot/3644) I had. Therefore the padding couldn’t be done correctly. The reproduction script is likely also due to the previous issue I had that hadn’t been fixed yet.

So I think a solution to this should be instead of disabling the preprocessor, I should write a custom pre-processor that only flattens tuple into vectors? Is this correct?

---

_[View the full topic](https://discuss.ray.io/t/when-are-marl-replay-buffers-zero-padded/3763)._