# Gaussian Process

**URL:** <https://discourse.julialang.org/t/gaussian-process/37940>\
**Category:** New to Julia\
**Tags:** question\
**Created:** [April 21, 2020, 2:46am UTC](https://discourse.julialang.org/t/gaussian-process/37940 "2020-04-21T02:46:19Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Praise](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/praise/32/13948_2.png) [@Praise](https://discourse.julialang.org/u/Praise)\
**Post date:** [April 21, 2020, 2:46am UTC](https://discourse.julialang.org/t/gaussian-process/37940/1 "2020-04-21T02:46:19Z")

</div>

Hi, I would like to perform Gaussian Process Regression on a certain set of data I have in a csv file. Basically, I have a number of columns in my csv file. Now, based on what I have looked for, I have found two ways of doing GPR in Julia.

1: Using flux and zygote  
2: Using GaussianProcesses

I am not sure how to fit the dataset using either of these methods and thats what I am trying to figure out. Any help would be appreciated. I am not asking for a complete answer, just something to start with would be really helpful.

---

<div class="post-metadata">

**Author:** ![SPV](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/spv/32/2606_2.png) [@SPV](https://discourse.julialang.org/u/SPV)\
**Post date:** [April 21, 2020, 2:58am UTC](https://discourse.julialang.org/t/gaussian-process/37940/2 "2020-04-21T02:58:54Z")

</div>

For GP’s, there are two main Julia packages.

There is [GaussianProcesses.jl](https://github.com/STOR-i/GaussianProcesses.jl) and [Stheno](https://github.com/willtebbutt/Stheno.jl). Take a look at their repo, there is a MWE in both readme’s. There is [CSV.jl](https://github.com/JuliaData/CSV.jl) for opening .csv files. Once you have the data loaded, it should be straight forward to adapt the examples from both packages for your purposes.

I am not sure, what you mean by using Flux to do GP Regression. If you are looking for something like GPflow or GPyTorch, they are listed above.