add orb support... tricky

[alinelena]

No description provided.

[oerc0122]

Could be simplified. Depends if you think it's more or less readable.

[ElliottKasoar]

Mostly for reference but just to note this may pin torch to 2.2.0, which at the moment is fine, but could be a problem in the future.

(I need to double check but I think even uninstalled extras' dependencies are taken into account when installing)

[ElliottKasoar]

I think the installation is all working now with uv, but since I don't think we're dropping 3.9 quite yet, the match statement still needs changing.

The README and getting started lists should also be updated.

[ElliottKasoar]

I think we want a test for the calculator init e.g. parameters are set, and (invalid) model path works as expected, but generally looks good!

[ElliottKasoar]

Rebased, and updated docs. Setting the model_path needs fixing, and then a couple more tests to be added.

Then the MacOS dm-tree error needs fixing.

[alinelena]

Rebased, and updated docs. Setting the model_path needs fixing, and then a couple more tests to be added.

Then the MacOS dm-tree error needs fixing.

i wonder if we shall to a try catch... and then raise accordingly?

[ElliottKasoar]

It's possible there are slightly neater versions for setting the model, but I think it now uses the default model correctly, as well as using the pretrained model labels, or a loaded model (which is tested with their XS checkpoint file).

Also installed cmake on our Mac, so hopefully those tests are fixed.

[ElliottKasoar]

I can't officially approve, but I think it looks good

[harveydevereux]

Also looks good to me