What does atype stand for in the section of inference?

[ShangChien]

What does atype stand for in the section of inference?

from deepmd.infer import DeepPot
import numpy as np

dp = DeepPot("graph.pb")
coord = np.array([[1, 0, 0], [0, 0, 1.5], [1, 0, 3]]).reshape([1, -1])
cell = np.diag(10 * np.ones(3)).reshape([1, -1])
atype = [1, 0, 1]
e, f, v = dp.eval(coord, cell, atype)

but, in another section of ASE usage, using H₂O as an example：

from ase import Atoms
from deepmd.calculator import DP

water = Atoms(
    "H2O",
    positions=[(0.7601, 1.9270, 1), (1.9575, 1, 1), (1.0, 1.0, 1.0)],
    cell=[100, 100, 100],
    calculator=DP(model="frozen_model.pb"),
)
print(water.get_potential_energy())

Intuitively, atype = [1, 0, 1] means [He, H, He], which doesn't make sense.
So, is atype = [1, 0, 1] refer to the Periodic Table or type_map in the model of graph.pb? It's probably the latter.

Prefer to use a global Periodic Table index, or specify element symbols list like ASE usage, which will reduce mental burden for coupling parameter with the setting of model training.
This style may facilitate widespread use of subsequent distilled DPA models in other extensions.

Thank you 🍰

[wanghan-iapcm]

atom type, specified by the "type_map" section in your training script.