# Implementing multiple skip connections in Flux.ml?

**URL:** https://discourse.julialang.org/t/implementing-multiple-skip-connections-in-flux-ml/23378
**Category:** Machine Learning
**Tags:** flux
**Created:** [April 21, 2019, 6:09pm UTC](https://discourse.julialang.org/t/implementing-multiple-skip-connections-in-flux-ml/23378 "2019-04-21T18:09:42Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![dopamine](https://avatars.discourse-cdn.com/v4/letter/d/e9a140/32.png) [@dopamine](https://discourse.julialang.org/u/dopamine)
#### Post date: [April 21, 2019, 6:09pm UTC](https://discourse.julialang.org/t/implementing-multiple-skip-connections-in-flux-ml/23378/1 "2019-04-21T18:09:42Z")

</div>

Hello,  
I’m trying to implement u-net ([see here for architecture](https://lmb.informatik.uni-freiburg.de/people/ronneber/u-net/)) like architecture using Flux.ml, which requires multiple skip connections. What is the most efficient way to implement multiple skip connections? I’ve seen [this answer on the discourse](https://discourse.julialang.org/t/best-way-to-implement-shortcut-connections-for-feed-forward-neural-networks-in-flux-jl/10633), but I’m a little bit lost on putting multiple concatenating skip connections.

Thanks!