Issues

• [Feature request] TorchSim calculator

  ICAMS/grace-tensorpotential#35 · Filter by author akwarii opened on Jul 15, 2026

• Possible inconsistency in reported mae/depa train/test metrics

  ICAMS/grace-tensorpotential#30 · Filter by author hprats opened on May 10, 2026
  2

• GPU utilization at large atomic scales

  ICAMS/grace-tensorpotential#28 · Filter by author Seunghyo-Noh opened on Apr 18, 2026
  1

• QhullError (QH6154) in compute_convexhull_dist due to degenerate/flat composition space

  ICAMS/grace-tensorpotential#26 · Filter by author yury-lysogorskiy opened on Feb 11, 2026
  

• User specified output directory in CLI

  ICAMS/grace-tensorpotential#24 · Filter by author harveydevereux opened on Feb 5, 2026
  

• Incompatibility with PyTorch

  ICAMS/grace-tensorpotential#23 · Filter by author ElliottKasoar opened on Jan 20, 2026
  

• How to obtain a float32 LAMMPS-runnable pretrained model

  ICAMS/grace-tensorpotential#22 · Filter by author alphalm4 opened on Dec 16, 2025
  1

• finetuning problem

  ICAMS/grace-tensorpotential#20 · Filter by author Asif-Iqbal-Bhatti opened on Nov 27, 2025
  3

• LAMMPS fix 'sgcmc' in MC package not found

  ICAMS/grace-tensorpotential#18 · Filter by author tisabe opened on Nov 12, 2025
  2

• Conda-forge architecture flag appears incorrect

  ICAMS/grace-tensorpotential#13 · Filter by author liamhuber opened on Sep 12, 2025

• kokkos suppoting issue and GRACE/FS GPU running in lammps

  ICAMS/grace-tensorpotential#10 · Filter by author Seunghyo-Noh opened on Jul 17, 2025
  3

  

• The GRACE-2L-OAM_28Jan25 was changed

  ICAMS/grace-tensorpotential#7 · Filter by author jan-janssen opened on Mar 5, 2025

• Python 3.13 support

  ICAMS/grace-tensorpotential#6 · Filter by author janosh opened on Feb 19, 2025
  1

• Obtaining atomic features

  ICAMS/grace-tensorpotential#5 · Filter by author siamak-attarian opened on Dec 17, 2024