Skip to content

Liquid densities#388

Open
kuryla wants to merge 4 commits intoddmms:mainfrom
kuryla:liquid_densities
Open

Liquid densities#388
kuryla wants to merge 4 commits intoddmms:mainfrom
kuryla:liquid_densities

Conversation

@kuryla
Copy link
Collaborator

@kuryla kuryla commented Feb 21, 2026

Pre-review checklist for PR author

PR author must check the checkboxes below when creating the PR.

Summary

Benchmark of 61 organic liquid densities. Densities are computed from a NPT molecular dynamics run and errors are obtained with respect to the experimental densities.

Linked issue

Resolves #

Progress

  • Calculations
  • Analysis
  • Application
  • Documentation

Testing

MACE-OMOL, MACE-POLAR-1-M, MACE-POLAR-1-L, UMA-S-1P1-OMOL

New decorators/callbacks

@kuryla kuryla added the new benchmark Proposals and suggestions for new benchmarks label Feb 21, 2026
@joehart2001 joehart2001 added the lr label Feb 21, 2026
@alinelena
Copy link
Collaborator

Hi @kuryla would be nice to have a calculator even if we do not plan to run the calculations, mainly as a reference of what we used.

Computational cost
------------------

Low: tests are likely to take several days to run on GPU.
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
Low: tests are likely to take several days to run on GPU.
Very high: tests are likely to take several days to run on GPU.

?

bad: 0.4
unit: g/cm^3
tooltip: Mean Absolute Error for all systems
level_of_theory: Experimental density
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
level_of_theory: Experimental density
level_of_theory: Experimental

Comment on lines +25 to +26
AU_TO_G_CM3 = 1e24 / units.mol
G_CM3_TO_AU = 1 / AU_TO_G_CM3
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
AU_TO_G_CM3 = 1e24 / units.mol
G_CM3_TO_AU = 1 / AU_TO_G_CM3
G_CM3_TO_AU = units.mol / 1e24

Can we not just go straight to this?

Comment on lines +34 to +38
download_s3_data(
filename="liquid_densities.zip",
key="inputs/molecular_dynamics/liquid_densities/liquid_densities.zip",
)
/ "liquid_densities"
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is this supposed to be here/has this been uploaded?

I think we probably want this in outputs/... or something, since this is more set up as calculation inputs, not analysis inputs, but I also don't know if we want to do the analysis download as part of the script. I think I'd rather check for the data in calcs/.../outputs, and if it's not there, suggest downloading is using ml_peg download

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

lr new benchmark Proposals and suggestions for new benchmarks

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants