Skip to content

rohanplatts/MLIP-Workflows

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

56 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

mlip-workflows

MLIP command line workflows for phonons, NEB/MEP, ShakeNBreak giving vasp-improvable results (for computational efficiency) for 5 model families are supported by this package. Furthermore, mlip fine-tuning is supported fully for MACE and ORB model families, with petmad coming soon. The former two support LoRA and REPLAY fine tuning.

Workflows

  • mlip-phonons --inputs <config-dir> for phonons, DOS, and band structures
  • mlip-neb --inputs <config-dir> for NEB / MEP runs
  • mlip-snb --inputs <config-dir> for ShakeNBreak defect workflows
  • mlip-coup --inputs <config-dir> for phonon-coupling analysis

Examples of SnB, NEB, and Phonon workflows are included.

Fine Tuning

Three families are included in fine-tuning. LoRA and REPLAY style fine tuning are available for MACE and ORB. Pet-mad is still running into some issues, I am yet to push that work.

Benchmarking Supported for any MLIP-workflow

All workflows are available to be benchmarked. That is, you can download any number of supported models (list found in SUPPORTED_MODELS.yml, but, generally, any model in the MACE, Orb, PET-MAD, Tensornet, mattersim and CHGNet model families).

If defaults.model_name is a list, the ordinary workflow command fans out once per model and runs each case in the model-specific environment from SUPPORTED_MODELS.yml. This enables you to take any workflow defined by its config.yml file, and then simply append model names to run the same workflow across however many models you like.

For the fine-tuning demos, step 4 shows this usage, with:

  • mlip-neb --inputs demo/fine_tuning/<family>/4_benchmark --report-benchmark

Setup

  • Put model files under assets/models/<family>/
  • Register model environments in SUPPORTED_MODELS.yml

For all model families you are using, install and set up their corresponding environments according to environments.md.

Once you do that, you are done and free to run these workflows. More details on running the workflows can be found here.

Caveat

Dependent on CUDA.

About

Useful for: Accelerating computational materials workflows using your choice of MLIP (many accounted for). Includes the acceleration of NEB, Phonon and Bandpath calculation, AND Model benchmarking on all processes.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors