# Flux.Conv kernel definition

**URL:** <https://discourse.julialang.org/t/flux-conv-kernel-definition/85109>\
**Category:** General Usage\
**Tags:** flux\
**Created:** [August 1, 2022, 5:10pm UTC](https://discourse.julialang.org/t/flux-conv-kernel-definition/85109 "2022-08-01T17:10:14Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![cirobr](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/cirobr/32/219994_2.png) [@cirobr](https://discourse.julialang.org/u/cirobr)\
**Post date:** [August 1, 2022, 5:10pm UTC](https://discourse.julialang.org/t/flux-conv-kernel-definition/85109/1 "2022-08-01T17:10:14Z")

</div>

Consider the below command line, which defines kernel size as 5x5 and has seven output channels:

> layer1 = Flux.Conv((5,5), 3 =\> 7, relu; bias = false)

Questions:

1. Which kernel(s), given that there are seven output channels, are effectively used on the above example? Are they pre-built?
2. How do I specify my own (5,5) kernel? How do I specify my own set of seven (5,5) kernels?

Thanks.