# Flux - LSTM - Issue with input format for multiple features

**URL:** <https://discourse.julialang.org/t/flux-lstm-issue-with-input-format-for-multiple-features/55272>\
**Category:** Machine Learning\
**Tags:** flux, machine-learning\
**Created:** [February 14, 2021, 7:24pm UTC](https://discourse.julialang.org/t/flux-lstm-issue-with-input-format-for-multiple-features/55272 "2021-02-14T19:24:37Z")\
**Posts on this page:** 1\
**Showing post:** 8

<div class="post-metadata">

**Author:** ![Mateusz\_K](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/mateusz_k/32/31832_2.png) [@Mateusz\_K](https://discourse.julialang.org/u/Mateusz_K)\
**Post date:** [November 1, 2022, 6:57am UTC](https://discourse.julialang.org/t/flux-lstm-issue-with-input-format-for-multiple-features/55272/8 "2022-11-01T06:57:43Z")

</div>

I found this thread and I think it’s not accurate. First of all Flux which uses NNLib dispatches to both 2D and 3D API exposed by CUDNN (that’s the case since long time). The 3D version is just not documented but you can pass three dimensional array through LSTM/GRU or any RNN. It’s something CUDNN supports and that is correctly dispatched to. I don’t know about non CUDAs applications.

Secondly, it states in documentation not to use map or broadcast dot operator with reccurent layers. Because order is not guaranteed and you want to send your input in sequencial order.

---

_[View the full topic](https://discourse.julialang.org/t/flux-lstm-issue-with-input-format-for-multiple-features/55272)._