# Hierarchical two-layer Gaussian process

**URL:** <https://discourse.pymc.io/t/hierarchical-two-layer-gaussian-process/5715>\
**Category:** Questions\
**Tags:** gaussian\_process\
**Created:** [August 30, 2020, 10:42pm UTC](https://discourse.pymc.io/t/hierarchical-two-layer-gaussian-process/5715 "2020-08-30T22:42:38Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![jroberayalas](https://avatars.discourse-cdn.com/v4/letter/j/7feea3/32.png) [@jroberayalas](https://discourse.pymc.io/u/jroberayalas)\
**Post date:** [August 30, 2020, 10:42pm UTC](https://discourse.pymc.io/t/hierarchical-two-layer-gaussian-process/5715/1 "2020-08-30T22:42:38Z")

</div>

Hi everyone,

I’m new in PyMC3 and I was wondering if there is a way to implement a hierarchical two-layer Gaussian process to implement something similar to what is described in section 3.1 of this [paper](https://arxiv.org/pdf/2005.08837.pdf). I’ve already check this other [post](https://discourse.pymc.io/t/using-a-gaussian-process-at-intermediate-level-in-hierarchy/5102), where a GP is used at an intermediate level in hierarchy, but I’d like to extend this to a two-layer GP. Any help/guidance is highly appreciated.

---

<div class="post-metadata">

**Author:** ![ml-ME-gatech](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/ml-me-gatech/32/4730_2.png) [@ml-ME-gatech](https://discourse.pymc.io/u/ml-ME-gatech)\
**Post date:** [January 21, 2022, 11:51am UTC](https://discourse.pymc.io/t/hierarchical-two-layer-gaussian-process/5715/2 "2022-01-21T11:51:07Z")

</div>

Any luck with this? I’m trying to do the same thing here.