nvalchemi.models.base.ModelConfig

class nvalchemi.models.base.ModelConfig(*, compute_forces=True, compute_stresses=False, compute_hessians=False, compute_dipoles=False, compute_charges=False, compute_embeddings=False, compute_energies=True, gradient_keys=<factory>)

Configuration structure for a given model.

All models that inherit from BaseModelMixin should have a model_config attribute that is an instance of this class, which can be used to change the behavior of the model.

Parameters:

• compute_forces (bool)

• compute_stresses (bool)

• compute_hessians (bool)

• compute_dipoles (bool)

• compute_charges (bool)

• compute_embeddings (bool)

• compute_energies (bool)

• gradient_keys (set[str])

compute_forces

Set to enable or disable force computation.

Type:

bool, default True

compute_stresses

Set to enable or disable stress computation.

Type:

bool, default False

compute_hessians

Set to enable or disable Hessian computation.

Type:

bool, default False

compute_dipoles

Set to enable or disable dipole computation.

Type:

bool, default False

gradient_keys

Set of keys to enable gradients for in the Batch of AtomicData structure.

Type:

set[str], default set()

model_config = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].