COSIMA Hackathon v2.0 -- Tuesday January 24th, 2023

Does that mean @angus-g will take care of How to contribute guide · Issue #189 · COSIMA/cosima-recipes · GitHub?

Don’t know if there is a registration, but I’d like to join from Hobart!

2 Likes

Like the post near the top that says Hobart to count you in for food! That’s the “registration”. We keep things simple in COSIMA.

2 Likes

look what I have gotten on my avatar!

1 Like

You must be one of the cool kids @navidcy

Hey, how do we create a new task under COSIMA Hackathon v2.0 👩🏽‍💻🧑🏻‍💻👩🏼‍💻 · GitHub ?

My good COSIMA resolution for the year is to try to implement a workflow to do models validations and inter comparisons using Pavel’s Enkf-C (GitHub - sakov/enkf-c: EnKF code for DA with large-scale layered geophysical models.).
I am half-way through for the coastal EMS model, but would like to make it more model agnostic (and COSIMA compatible first!).

1 Like

@CloLanglais create an issue at the COSIMA recipe repo and add the label “Hackathon v2.0” / add it into Project “Hackathon v2.0”

Less than 24hrs for Hackathon v2.0! The seconds are ticking!

We urge all of you who will join virtually or remotely to dress up! The theme is:

ocean :ocean::beach_umbrella: - finding Nemo :blowfish::tropical_fish: - "Under the sea":mermaid:t4:

We will give out a prize :trophy: for the “Most oceanic attire”. The prize will be an honourary title + you’ll get prioritize in your Pull Request review + potentially other GitHub-related honours we will think by tomorrow!

(Note to @Disney: it should be corrected to “Under the sea surface” since “under the sea” implies below the sea floor!)

If you’ve never used COSIMA cookbook before then we’d urge you to have a look at this user-friendly guide our team has put together! It would be very helpful if you already have the basic set up before the Hackathon v2.0 begins tomorrow (eg, way to open a Jupyter notebook on Gadi and stuff of this sort).

2 Likes

Looks like the ARE is currently down. I (and probably others) have let NCI know. Let’s hope it’s fixed by tomorrow…

1 Like

It’s not a big deal. I never used ARE in my life. There are many other alternatives.

1 Like

Working again now anyway

2 Likes

Canberra hackathon cake is going in the oven. We have 18 people attending in Canberra, so if anyone else feels like making morning/afternoon tea snacks I’m sure it will be appreciated also!

6 Likes

And don’t forget to get your dress ups out for tomorrow. Or an ocean themed tshirt. Or your boogie board. …

Since we are at the subject, I just want to point out that the cosima-recipes repository has only 20 stars. Doesn’t it feel a bit low count to you?

This is how I arrived at the docks but my scales fall off when I’m out of the water

5 Likes

The project board that Navid is looking at now is at: COSIMA Hackathon v2.0 👩🏽‍💻🧑🏻‍💻👩🏼‍💻 · GitHub

The ARE link through which most of us will be working with our notebooks: https://are.nci.org.au/pun/sys/dashboard

Okay so I have a code that is able to compute a mass stream function for T and S and in principle should work for any 2 scalar variables. The key things remaining are to 1) make it cleaner, 2) make it faster (it is quite slow and I’d like to make it work on the 1/10th degree model) and 3) make it work for arbitrary variables (like depth, latitude, density) and 4) comment it nicely. Is anyone willing to take a look with me after the tutorials (maybe in a zoom breakout). Particularly keen for people that a) feel like solving some fun coding puzzles and b) know the capabilities of dask/xarray for binning/sorting/conditional averaging.

This sounds interesting to me, but now I’ve committed to Test/document how recent dask improvements affect particularly sticky calculations · Issue #191 · COSIMA/cosima-recipes · GitHub. Might have some time later today though!

1 Like