MLIP School 2026 — Dataset craft for machine-learned interatomic potentials using ACEpotentials.jl

You will build a silicon dataset from scratch, fit ACE potentials, and learn why a low RMSE can still be a bad potential.

The hosted services were retired on 8 September 2026. Everything here still works, but the compute now runs on your machine rather than ours.

curl -O https://mlipschool.uk/local.yml
docker compose -f local.yml up -d

That brings up the oracle on localhost:8001 and the ACE workbench on localhost:8002, which is where the notebooks already look. Nothing local checks keys, so any key string works — but every notebook still hard-stops on an empty API key field, so paste something into it before running a service cell.

The notebook pages below still open and read, and their released solutions still reveal, but their service calls will fail until the stack above is running.

The opening talk is a five-minute read of the slides from the room, published flat so it keeps working after the session.

Two ways to work

In your browser retired

The hosted services are retired. These pages still load, but they cannot label structures or fit potentials any more — the oracle and workbench are gone for good, not merely slow to wake up. Use On your own machine instead.

Work through the day-one exercises in order: E1, E2, E3, then C. E1x is an optional extension — take it after E1 if you are ahead. It resumes nothing and spends none of your labels, so skipping it costs you nothing.

Keep one exercise open at a time. Each of these pages carries its own Python runtime and uses about 1.5 GB of memory, so close the tab for the exercise you have finished before opening the next. Two open at once is fine on most laptops; three is not.

Day two — bring your own system

A separate one-hour session the following day. It resumes nothing from day one: you bring a system of your own and run the same loop on it.

On your own machine

Run:

uvx mograder student https://mlipschool.uk/mograder.toml

Run the whole stack locally. The hosted services are retired, so this is the only way to run these exercises now: both the oracle that labels your structures and the workbench that fits them, on your own machine. About 1.3 GB to download.

curl -O https://mlipschool.uk/local.yml
docker compose -f local.yml up -d

Then set the notebook's Oracle field to http://localhost:8001 and ACE workbench to http://localhost:8002. Nothing local checks keys, so any key string works, and the label budget is your own rather than the hosted allowance. This works with the notebook you run yourself, not with one opened from this site.

Licence: The oracle image contains the MACE-MH-1 model, licensed under the Academic Software Licence (ASL): academic and non-commercial use only. https://github.com/gabor1/ASL

Credits. Authored by James Kermode, with feedback and support from Facundo Costa, Roman Shantsila and Svenja Janke.

AI declaration. Support from AI coding agents setting up the infrastructure, but I designed and (with help from Roman, Facundo and Svenja) fact checked everything and take responsibility for scientific decisions as well as for the accuracy of what’s there.