# Design a three-dimensional ConvolutionLayer?

**URL:** <https://community.wolfram.com/t/design-a-three-dimensional-convolutionlayer/10542>\
**Category:** Questions\
**Tags:** data-science, image-processing, machine-learning, neural-networks\
**Created:** [May 16, 2017, 9:46pm UTC](https://community.wolfram.com/t/design-a-three-dimensional-convolutionlayer/10542 "2017-05-16T21:46:07Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![lgmendes](https://avatars.discourse-cdn.com/v4/letter/l/f14d63/32.png) [@lgmendes](https://community.wolfram.com/u/lgmendes)\
**Post date:** [May 16, 2017, 9:46pm UTC](https://community.wolfram.com/t/design-a-three-dimensional-convolutionlayer/10542/1 "2017-05-16T21:46:07Z")

</div>

Is it possible to design a three-dimensional ConvolutionLayer in version 11.1.1?

---

<div class="post-metadata">

**Author:** ![sghuisman](https://yyz1.discourse-cdn.com/wolfram/user_avatar/community.wolfram.com/sghuisman/32/27850_2.png) [@sghuisman](https://community.wolfram.com/u/sghuisman)\
**Post date:** [May 17, 2017, 8:32am UTC](https://community.wolfram.com/t/design-a-three-dimensional-convolutionlayer/10542/2 "2017-05-17T08:32:35Z")

</div>

It suggest it is possible (especially the last note under details & options). Use the option “Dimensionality” found in the Details & Options section in the documentation of ConvolutionLayer…

---

<div class="post-metadata">

**Author:** ![dabrowsa](https://yyz1.discourse-cdn.com/wolfram/user_avatar/community.wolfram.com/dabrowsa/32/23592_2.png) [@dabrowsa](https://community.wolfram.com/u/dabrowsa)\
**Post date:** [May 20, 2017, 5:33pm UTC](https://community.wolfram.com/t/design-a-three-dimensional-convolutionlayer/10542/3 "2017-05-20T17:33:12Z")

</div>

That Dimensionality option would be handy for PoolingLayer also.

---

<div class="post-metadata">

**Author:** ![lgmendes](https://avatars.discourse-cdn.com/v4/letter/l/f14d63/32.png) [@lgmendes](https://community.wolfram.com/u/lgmendes)\
**Post date:** [August 18, 2017, 4:55pm UTC](https://community.wolfram.com/t/design-a-three-dimensional-convolutionlayer/10542/4 "2017-08-18T16:55:44Z")

</div>

Just one update given by the Wolfram support that may be useful:

_Unfortunately, ConvolutionLayer can currently handle tensors of up to rank 3 right now:_

> **[ConvolutionLayer: Apply a convolution operation—Wolfram Documentation](https://reference.wolfram.com/language/ref/ConvolutionLayer.html)**
>
> ConvolutionLayer is a type of network layer that applies a convolution operation to the input.

_It is not currently set up to handle rank 4 tensors as would be required for RGB-D data of 3D images._