# Performance issues?

**URL:** https://discourse.julialang.org/t/performance-issues/46489
**Category:** New to Julia
**Tags:** question, flux
**Created:** [September 12, 2020, 12:34am UTC](https://discourse.julialang.org/t/performance-issues/46489 "2020-09-12T00:34:57Z")
**Posts on this page:** 1
**Showing post:** 11

<div class="post-metadata">

### Author: ![mcabbott](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/mcabbott/32/6603_2.png) [@mcabbott](https://discourse.julialang.org/u/mcabbott)
#### Post date: [September 12, 2020, 7:13pm UTC](https://discourse.julialang.org/t/performance-issues/46489/11 "2020-09-12T19:13:01Z")

</div>

You _should_ be able to use `batched_mul` for this, but someone needs to merge [#191](https://github.com/FluxML/NNlib.jl/pull/191) first:

```julia
julia> using NNlib # v0.7.4

julia> batched_mul(rand(2,3,5), rand(3,4,1))
ERROR: DimensionMismatch("batch size mismatch")

```

Edit – in reply to a deleted comment.

And xref [a recent thread](https://discourse.julialang.org/t/permuteddimsarray-slower-than-permutedims/46401/4). However, whether this is the bottleneck here or not I’ve no idea.

---

_[View the full topic](https://discourse.julialang.org/t/performance-issues/46489)._