ACEsuit/mace

Issues

• Can we train the periodic and nonperiodic structures together in the training set ?

  #580 opened 10 hours ago by jinfeng-data
  18

• universal loss (at least in multi-head-interface branch) does not support per-config weight

  #558 opened 16 days ago by bernstei
  0

• Automatically add parameters to optimizer

  #525 opened 2 months ago by LarsSchaaf
  0

• Move to ruff for linting and formating?

  #534 opened a month ago by CompRhys
  5

• Improve plotting script

  #524 opened 2 months ago by ilyes319
  0

• Change stress to virials internally

  #527 opened 2 months ago by ilyes319
  0

• stress errors reported during fitting and at end should be consistent

  #442 opened 10 hours ago by bernstei
  5

• models created from multihead fine tuning don't work in lammps

  #557 opened 10 hours ago by bernstei
  7

• cannot turn off multihead finetuning in multi-head-interface branch

  #559 opened 10 hours ago by bernstei
  18

• Issue creating LAMMPS model from EnergyDipoleMACE model

  #582 opened a day ago by lemessuriernb
  3

• Problem when finalizing the training of a pure water model

  #581 opened 2 days ago by nihil39
  8

• Some Questions about MACE for pure water and in general

  #576 opened 6 days ago by nihil39
  2

• [feature suggest] active learning in lammps

  #568 opened 7 days ago by thangckt
  1

• How to use multi-GPUs training with PBS system

  #567 opened 7 days ago by jinfeng-data
  21

• I got Warning pytorch not support instance-level annotations.

  #573 opened 7 days ago by RBERLOP
  5

• Remove `e3nn` version pin

  #555 opened 19 days ago by Andrew-S-Rosen
  2

• help message for `--pair_repulsion` is wrong

  #565 opened 13 days ago by njzjz
  1

• Colab Tutorial Link Does not work

  #563 opened 14 days ago by RBERLOP
  2

• Bug on loading finetuning model

  #545 opened a month ago by hspark1212
  2

• Problems with parallelization on CPU (Using LAMMPS)

  #520 opened 2 months ago by MarcoRavalli
  1

• Segmentation fault during LAMMPS-MACE-CPU run

  #519 opened 2 months ago by fcannizzaro-byte
  1

• Tripeptides Data for MACE-OFF Model

  #538 opened a month ago by tkreiman
  0

• ASE documentation update?

  #531 opened a month ago by wcwitt
  4

• Issue on integrating lammps with MACE potential

  #507 opened 2 months ago by aishwaryo
  10

• CUDA out of memory from MD with MP-0 pre-trained model using 'big' geometry.

  #487 opened 2 months ago by turbosonics
  16

• Error while using multihead interface

  #513 opened 2 months ago by jungsdao
  2

• How to use LAMMPS to train a larger system?

  #476 opened 3 months ago by stargolike
  6

• Could you please share with more examples that use MACE in LAMMPS?

  #497 opened 2 months ago by waywaygirl
  5

• LAMMPS-GPU (mace) installed Failed

  #503 opened 2 months ago by waywaygirl
  3

• Deployment of the MACE Model Without Using ASE

  #511 opened 2 months ago by Hsuchein
  5

• Flexibility in choosing which checkpoint to save during multi-head training.

  #479 opened 2 months ago by LarsSchaaf
  6

• Restart of the training using --restart_latest doesn't properly work - resolved -

  #509 opened 2 months ago by turbosonics
  0

• How to use multi-training without slurm system?

  #458 opened 3 months ago by stargolike
  18

• installation

  #453 opened 2 months ago by niloufareabi
  19

• MD crash with finetuned model from MACE-MP0

  #499 opened 2 months ago by jungsdao
  4

• How to recreate/find MP ids in your hdf5 train/val splits

  #501 opened 2 months ago by DeNeutoy
  0

• Error in online data loading for large datasets

  #465 opened 2 months ago by John2021-hub
  1

• Error accessing Git repository

  #483 opened 2 months ago by turbosonics
  2

• Vectorized Hessians seem not to work

  #488 opened 2 months ago by Nilsgoe
  10

• Machines other than A100 compiling LAMMPS

  #496 opened 2 months ago by stargolike
  4

• Inquiry about implement of `pair_repulsion` in the mace_mp function

  #494 opened 2 months ago by hspark1212
  2

• First experiment using lammps with a pretrained potential (mace mp0)

  #473 opened 3 months ago by nihil39
  6

• Dependencies pyproject.toml

  #491 opened 2 months ago by galjos
  4

• model not returning stress in the latest commit on `multi-head-interface`

  #481 opened 2 months ago by CheukHinHoJerry
  3

• Does MACE satisfy E(3) equivariance or just SE(3) equivariance?

  #469 opened 3 months ago by QuantumLab-ZY
  0

• Problem in Compiling Lammps for GPU version

  #463 opened 3 months ago by XiaoTongDeng
  3

• How to Convert VASP Output Files to MACE XYZ Format for Training and Testing Sets

  #456 opened 3 months ago by obaica
  0

• multi-head-finetune validation batches not deterministic

  #450 opened 3 months ago by bernstei
  3

• during-training valid loss is wrong

  #444 opened 3 months ago by bernstei
  11

• np.random.choice for selecting pre-fit subset for multihead fine tune doesn't accept a list of Atoms

  #447 opened 3 months ago by bernstei
  0