mxtaltools.models.utils

mxtaltools.models.utils.clean_cell_params(samples, sg_inds, lattice_means, lattice_stds, symmetries_dict, asym_unit_dict, rescale_asymmetric_unit=True, destandardize=False, mode='soft', fractional_basis='asymmetric_unit', skip_angular_dof=False)[source]

# todo deprecate An important function for enforcing physical limits on cell parameterization with randomly generated samples of different sources.

Parameters:
mxtaltools.models.utils.clean_generator_output(samples=None, lattice_lengths=None, lattice_angles=None, mol_positions=None, mol_orientations=None, lattice_means=None, lattice_stds=None, destandardize=True, mode='soft', skip_angular_dof=False)[source]

# TODO rewrite - this is a very important function but it’s currently a disaster convert from raw model output to the actual cell parameters with appropriate bounds considering raw outputs to be in the standardized basis, we destandardize, then enforce bounds

mxtaltools.models.utils.compute_prior_loss(norm_factors: Tensor, sg_inds: LongTensor, generator_raw_samples: Tensor, prior: Tensor, variation_factor: Tensor) → tuple[Tensor, Tensor][source]

Take the norm of the scaled distances between prior and generated samples, and apply a quadratic penalty when it is larger than variation_factor :param data: :param generator_raw_samples: :param prior: :param variation_factor:

mxtaltools.models.utils.compute_reduced_volume_fraction(cell_lengths: tensor, cell_angles: tensor, atom_radii: tensor, batch: tensor, crystal_multiplicity: tensor)[source]

# TODO DEPRECATE IN FAVOUR OF PACKING COEFFICIENT

Parameters:

  • cell_lengths

  • cell_angles

  • atom_radii

  • crystal_multiplicity

Returns: asymmetric unit volume / sum of vdw volumes - so-called ‘reduced volume fraction’

mxtaltools.models.utils.decode_to_sph_rotvec(mol_orientations)[source]

each angle is predicted with 2 params we bound the encodings for theta on 0-1 to restrict the range of theta to [0,pi/2]

mxtaltools.models.utils.decode_to_sph_rotvec2(mol_orientation_components)[source]

# todo decide whether to use/keep or deprecate this each angle is predicted with 2 params we bound the encodings for theta on 0-1 to restrict the range of theta to [0,pi/2]

identical to the above, but considering theta as a simple scalar [n, 5] input to [n, 3] output

mxtaltools.models.utils.denormalize_generated_cell_params(normed_cell_samples: FloatTensor, mol_data, asym_unit_dict: dict)[source]
mxtaltools.models.utils.dict_of_tensors_to_cpu_numpy(stats)[source]
mxtaltools.models.utils.embed_crystal_list(batch_size: int, crystal_list: list, embedding_type: str, encoder_checkpoint_path: Optional = None, device: str | None = 'cpu', redo_crystal_analysis: bool | None = False) → list[source]
mxtaltools.models.utils.enforce_1d_bound(x: tensor, x_span, x_center, mode='soft')[source]

constrains function to range x_center plus/minus x_span :param x: :param x_span: :param x_center: :param mode:

mxtaltools.models.utils.get_model_sizes(models_dict: dict)[source]
mxtaltools.models.utils.get_mol_embedding_for_proxy(crystal_batch, embedding_type, encoder: Optional = None)[source]
mxtaltools.models.utils.get_regression_loss(regressor, data, targets, mean, std)[source]
mxtaltools.models.utils.increment_value(value, increment, maxval, minval=0)[source]
mxtaltools.models.utils.load_encoder(checkpoint_path)[source]
mxtaltools.models.utils.norm_scores(score, tracking_features, dataDims)[source]

norm the incoming score according to some feature of the molecule (generally size)

mxtaltools.models.utils.renormalize_generated_cell_params(generator_raw_samples, mol_data, asym_unit_dict)[source]
mxtaltools.models.utils.softmax_and_score(raw_classwise_output, temperature=1, old_method=False, correct_discontinuity=True) → Tensor | ndarray[source]
Parameters:

  • raw_classwise_output (numpy array or torch tensor with dimension [n,2], representing the non-normalized [false,true] probabilities)

  • temperature (softmax temperature)

  • old_method (use more complicated method from first paper)

  • correct_discontinuity (correct discontinuity at 0 only in the old method)

Returns:

score

Return type:

linearizes the input probabilities from (0,1) to [-inf, inf] for easier visualization

mxtaltools.models.utils.test_gradient_flow(grad_params, operation)[source]
mxtaltools.models.utils.undo_1d_bound(x: tensor, x_span, x_center, mode='soft')[source]

undo / rescale an enforced 1d bound only setup for soft rescaling