| Property | Value |
|---|---|
| Size | 932 NEB relaxation trajectories |
| Reaction Types | Desorptions, Dissociations, Transfers |
| Purpose | Transition state energy calculations |
| Paper | CatTSunami (arXiv) |
| License | CC-BY-4.0 |
Overview¶
This is a validation dataset which was used to assess model performance in CatTSunami: Accelerating Transition State Energy Calculations with Pre-trained Graph Neural Networks. It is comprised of 932 NEB relaxation trajectories. There are three different types of reactions represented: desorptions, dissociations, and transfers. NEB calculations allow us to find transition states. The rate of reaction is determined by the transition state energy, so access to transition states is very important for catalysis research. For more information, check out the paper.
File Structure and Contents¶
The tar file contains 3 subdirectories: dissociations, desorptions, and transfers. As the names imply, these directories contain the converged DFT trajectories for each of the reaction classes. Within these directories, the trajectories are named to identify the contents of the file. Here is an example and the anatomy of the name:
desorption_id_83_2409_9_111-4_neb1.0.traj
desorptionindicates the reaction type (dissociation and transfer are the other possibilities)ididentifies that the material belongs to the validation in domain split (ood - out of domain is th e other possibility)83is the task id. This does not provide relevant information2409is the bulk index of the bulk used in the ocdata bulk pickle file9is the reaction index. for each reaction type there is a reaction pickle file in the repository. In this case it is the 9th entry to that pickle file111-4the first 3 numbers are the miller indices (i.e. the (1,1,1) surface), and the last number corresponds to the shift value. In this case the 4th shift enumerated was the one used.neb1.0the number here indicates the k value used. For the full dataset, 1.0 was used so this does not distinguish any of the trajectories from one another.
The content of these trajectory files is the repeating frame sets. Despite the initial and final frames not being optimized during the NEB, the initial and final frames are saved for every iteration in the trajectory. For the dataset, 10 frames were used - 8 which were optimized over the neb. So the length of the trajectory is the number of iterations (N) * 10. If you wanted to look at the frame set prior to optimization and the optimized frame set, you could get them like this:
from __future__ import annotations
from pathlib import Path
from urllib.request import urlretrieve
trajectory_path = Path("desorption_id_83_2409_9_111-4_neb1.0.traj")
if not trajectory_path.exists():
urlretrieve(
"https://dl.fbaipublicfiles.com/opencatalystproject/data/large_files/"
"desorption_id_83_2409_9_111-4_neb1.0.traj",
trajectory_path,
)
from ase.io import read
traj = read(trajectory_path, ":")
unrelaxed_frames = traj[0:10]
relaxed_frames = traj[-10:]Download¶
| Splits | Size of compressed version (in bytes) | Size of uncompressed version (in bytes) | MD5 checksum (download link) |
|---|---|---|---|
| ASE Trajectories | 1.5G | 6.3G | 52af34a93758c82fae951e52af445089 |
Use¶
One more note: We have not prepared an lmdb for this dataset. This is because it is NEB calculations are not supported directly in ocp. You must use the ase native OCP class along with ase infrastructure to run NEB calculations. Here is an example of a use:
import os
from ase.io import read
from ase.mep import DyNEB
from ase.optimize import BFGS
from fairchem.core import FAIRChemCalculator, pretrained_mlip
traj = read("desorption_id_83_2409_9_111-4_neb1.0.traj", ":")
images = traj[0:10]
predictor = pretrained_mlip.get_predict_unit("uma-s-1p2p1")
neb = DyNEB(images, k=1)
for image in images:
image.calc = FAIRChemCalculator(predictor, task_name="oc20")
optimizer = BFGS(
neb,
trajectory="neb.traj",
)
# Use a small number of steps here to keep the docs fast during CI, but otherwise do quite reasonable settings.
fast_docs = os.environ.get("FAST_DOCS", "false").lower() == "true"
if fast_docs:
optimization_steps = 20
else:
optimization_steps = 300
conv = optimizer.run(fmax=0.45, steps=optimization_steps)
if conv:
neb.climb = True
conv = optimizer.run(fmax=0.05, steps=optimization_steps)Warp 1.18.0 initialized:
CUDA Toolkit 13.4, Driver 13.1
Devices:
"cpu" : "x86_64"
"cuda:0" : "Tesla T4" (16 GiB, sm_75, mempool enabled)
Kernel cache:
/home/runner/.cache/warp/1.18.0
WARNING:root:device was not explicitly set, using device='cuda'.
WARNING:root:UMA Nitro cannot serve this request: nitro mode supports only CUDA compute capability 8.x-12.x, not 7.5.
WARNING:root:Now running with nitro=False, execution_mode='umas_fast_gpu' (TF32).
WARNING:root:Running in a reduced-precision mode: some values, such as higher-order derivatives and phonon calculations, will be less accurate. For standard FP32 precision, use inference_settings='high-precision'.
/home/runner/work/_tool/Python/3.12.16/x64/lib/python3.12/site-packages/ase/mep/neb.py:329: UserWarning: The default method has changed from 'aseneb' to 'improvedtangent'. The 'aseneb' method is an unpublished, custom implementation that is not recommended as it frequently results in very poor bands. Please explicitly set method='improvedtangent' to silence this warning, or set method='aseneb' if you strictly require the old behavior (results may vary). See: https://gitlab.com/ase/ase/-/merge_requests/3952
warnings.warn(
WARNING:root:Model is being compiled. This might take a while the first time.
WARNING:root:Running (unknown commit) with inference settings: multigpu
W1027 21:59:00.933000 8869 site-packages/torch/_logging/_internal.py:1345] [0/0] Profiler record function <class 'torch.autograd.profiler.record_function'> will be ignored
W1027 21:59:39.367000 8869 site-packages/torch/_inductor/utils.py:1953] [6/0] Not enough SMs to use max_autotune_gemm mode
/home/runner/work/_tool/Python/3.12.16/x64/lib/python3.12/site-packages/torch/_inductor/lowering.py:2352: FutureWarning: `torch._prims_common.check` is deprecated and will be removed in the future. Please use `torch._check*` functions instead.
check(
Step Time Energy fmax
BFGS: 0 22:00:33 -305.668762 5.272435
BFGS: 1 22:00:41 -305.597497 11.665584
BFGS: 2 22:00:43 -305.823063 1.877241
BFGS: 3 22:00:46 -305.840877 2.466166
BFGS: 4 22:00:50 -305.935035 2.365342
BFGS: 5 22:00:56 -305.942108 7.320115
BFGS: 6 22:01:02 -306.228828 8.432068
BFGS: 7 22:01:09 -306.223760 2.478627
BFGS: 8 22:01:14 -306.273578 4.309764
BFGS: 9 22:01:20 -306.300465 0.770947
BFGS: 10 22:01:30 -306.314716 0.581313
BFGS: 11 22:01:35 -306.349713 1.184270
BFGS: 12 22:01:40 -306.408279 1.993297
BFGS: 13 22:01:49 -306.427398 0.445973
BFGS: 14 22:01:53 -306.423898 0.895116
BFGS: 15 22:01:57 -306.406793 2.155161
BFGS: 16 22:01:58 -306.344966 2.776459
BFGS: 17 22:02:03 -306.328394 1.021517
BFGS: 18 22:02:07 -306.226966 1.874940
BFGS: 19 22:02:16 -306.129472 3.274678
BFGS: 20 22:02:19 -306.147870 2.499411
BFGS: 21 22:02:25 -306.144297 0.834558
BFGS: 22 22:02:30 -306.177485 0.913793
BFGS: 23 22:02:35 -306.179304 0.410390
BFGS: 24 22:02:36 -306.226625 1.331984
BFGS: 25 22:02:42 -306.249267 1.249043
BFGS: 26 22:02:46 -306.234369 0.639631
BFGS: 27 22:02:54 -306.223025 0.480729
BFGS: 28 22:02:56 -306.226945 0.285392
BFGS: 29 22:03:02 -306.233365 0.352942
BFGS: 30 22:03:14 -306.248053 0.361709
BFGS: 31 22:03:20 -306.251789 0.433618
BFGS: 32 22:03:23 -306.247831 0.332604
BFGS: 33 22:03:32 -306.241332 0.308962