Hi Ashley,

First of all apologies for not seeing this until now. It looks like your message was held because you're not a member of the mailing list and for some reason the list owners didn't get a notification about the held message. I'm cc'ing you directly in the reply along with the list but to ensure that you'll see future messages and that your messages don't get held you should subscribe to the yt-dev list:

https://mail.python.org/mm3/mailman3/lists/yt-dev.python.org/

That's great to hear you're interested, I think GSOC will be a good opportunity for the summer between finishing your bachelors and starting your PhD, especially if you're interested in a computational project for your PhD.

We already had one other potential student express interest in that project, which is fine, but if both of you end up wanting to apply for this project then we might want to think about whether we can break up the project into two separate chunks that do not depend on each other.

Here's a list of materials that you might want to familiarize yourself with as background material for the project:

* YTEP-0032: http://ytep.readthedocs.io/en/latest/YTEPs/YTEP-0032.html
* The talk Meagan Lang and I gave at SciPy 2017: https://www.youtube.com/watch?v=pkZgQIGac6I

The project you are interested in will require an understanding of methods for interpolating data from an N-body simulation onto various grid types. To understand the motivation and the algorithms you'll be using, it would be good to get some background on smoothed particle hydrodynamics (although it seems you already have some background), KDTrees and the k-nearest neighbors problem. Some materials that might be helpful:

* The paper describing the SPLASH visualization software (http://users.monash.edu.au/~dprice/splash/): https://doi.org/10.1071/AS07022
* "Smoothed Particle Hydrodynamics and Magnetohydroynamics" by Dan Price: https://arxiv.org/abs/1012.1885 (in particular Sections 1, 2, and 4)
* cykdtree, the cython KDTree library by Meagan Lang we've been using for this project: https://github.com/cykdtree/cykdtree
* If your university library has it, the discussion in "Computer Simulation using Particles" by Hockney and Eastwood about particle deposition methods

SPLASH is a piece of Fortran SPH visualization software. We don't make use of SPLASH ourselves much, but the paper describing it has some excellent discussion on how to visualize SPH data in practice.

You've already submitted your first pull request (thanks again!) so the next step is to start writing your proposal. You're still quite early so there is plenty of time for that step. I'd also be happy to review your proposal early to maximize your changes of success.

Hope that helps!

-Nathan


On Mon, Feb 19, 2018 at 12:15 PM, Ashley Kelly <a.j.kelly@durham.ac.uk> wrote:
Hi all,

I'm currently 4rth student at Durham University (UK) studying physics (MPhys) and in October I will be undertaking a phd in the field of theoretical astrophysics (SPH simulations - likely working on improving feedback models for next gen sph codes). My phd destination is currently undecided, I'm waiting to receive all of my offers before making a final decision, but I will know before the end of March.

It will probably not be surprising that I am really interested in the sph smoothing project, 'Interpolating particle data onto grids'. I actually have a reasonable amount of experience with this kind of stuff, I've worked with a variety of sph codes, and even written some toy ones myself. I'm just following the list of things to do to apply on https://github.com/yt-project/gsoc-2018  and this is number 3.

I have 4 phd interviews in the next 10 days, which with travel and prep, take up a large amount of my time, as such I will probably start my application and take a look at open issues in a weeks time!

Cheers,
Ashley Kelly
_______________________________________________
yt-dev mailing list -- yt-dev@python.org
To unsubscribe send an email to yt-dev-leave@python.org
_______________________________________________
yt-dev mailing list -- yt-dev@python.org
To unsubscribe send an email to yt-dev-leave@python.org