matgl package

MatGL (Materials Graph Library) is a graph deep learning library for materials science.

Subpackages

  • matgl.apps package
    • matgl.apps.pes module
      • Potential
        • Potential.forward()
  • matgl.data package
    • matgl.data.transformer module
      • LogTransformer
        • LogTransformer.inverse_transform()
        • LogTransformer.transform()
      • Normalizer
        • Normalizer.from_data()
        • Normalizer.inverse_transform()
        • Normalizer.transform()
      • Transformer
        • Transformer.inverse_transform()
        • Transformer.transform()
  • matgl.ext package
    • matgl.ext.ase module
      • Atoms2Graph
        • Atoms2Graph.get_graph()
      • M3GNetCalculator
        • M3GNetCalculator.calculate()
        • M3GNetCalculator.implemented_properties
      • MolecularDynamics
        • MolecularDynamics.run()
        • MolecularDynamics.set_atoms()
      • OPTIMIZERS
        • OPTIMIZERS.bfgs
        • OPTIMIZERS.bfgslinesearch
        • OPTIMIZERS.fire
        • OPTIMIZERS.lbfgs
        • OPTIMIZERS.lbfgslinesearch
        • OPTIMIZERS.mdmin
        • OPTIMIZERS.scipyfminbfgs
        • OPTIMIZERS.scipyfmincg
      • Relaxer
        • Relaxer.relax()
      • TrajectoryObserver
        • TrajectoryObserver.as_pandas()
        • TrajectoryObserver.save()
    • matgl.ext.pymatgen module
      • Molecule2Graph
        • Molecule2Graph.get_graph()
      • Structure2Graph
        • Structure2Graph.get_graph()
      • get_element_list()
  • matgl.graph package
    • matgl.graph.compute module
      • compute_3body()
      • compute_pair_vector_and_distance()
      • compute_theta()
      • compute_theta_and_phi()
      • create_line_graph()
    • matgl.graph.converters module
      • GraphConverter
        • GraphConverter.get_graph()
        • GraphConverter.get_graph_from_processed_structure()
    • matgl.graph.data module
      • MGLDataset
        • MGLDataset.has_cache()
        • MGLDataset.load()
        • MGLDataset.process()
        • MGLDataset.save()
      • MEGNetDataset
        • MEGNetDataset.has_cache()
        • MEGNetDataset.load()
        • MEGNetDataset.process()
        • MEGNetDataset.save()
      • MGLDataLoader()
      • collate_fn()
      • collate_fn_efs()
  • matgl.layers package
    • matgl.layers._activations module
      • ActivationFunction
        • ActivationFunction.sigmoid
        • ActivationFunction.softexp
        • ActivationFunction.softplus
        • ActivationFunction.softplus2
        • ActivationFunction.swish
        • ActivationFunction.tanh
      • SoftExponential
        • SoftExponential.forward()
      • SoftPlus2
        • SoftPlus2.forward()
    • matgl.layers._atom_ref module
      • AtomRef
        • AtomRef.fit()
        • AtomRef.forward()
        • AtomRef.get_feature_matrix()
    • matgl.layers._basis module
      • FourierExpansion
        • FourierExpansion.forward()
      • GaussianExpansion
        • GaussianExpansion.forward()
        • GaussianExpansion.reset_parameters()
      • RadialBesselFunction
        • RadialBesselFunction.forward()
      • SphericalBesselFunction
        • SphericalBesselFunction.rbf_j0()
      • SphericalBesselWithHarmonics
        • SphericalBesselWithHarmonics.forward()
      • SphericalHarmonicsFunction
      • spherical_bessel_smooth()
    • matgl.layers._bond module
      • BondExpansion
        • BondExpansion.forward()
    • matgl.layers._core module
      • EdgeSet2Set
        • EdgeSet2Set.forward()
        • EdgeSet2Set.reset_parameters()
      • GatedMLP
        • GatedMLP.forward()
      • MLP
        • MLP.depth
        • MLP.forward()
        • MLP.in_features
        • MLP.last_linear
        • MLP.out_features
    • matgl.layers._embedding module
      • EmbeddingBlock
        • EmbeddingBlock.forward()
    • matgl.layers._graph_convolution module
      • M3GNetBlock
        • M3GNetBlock.forward()
      • M3GNetGraphConv
        • M3GNetGraphConv.edge_update_()
        • M3GNetGraphConv.forward()
        • M3GNetGraphConv.from_dims()
        • M3GNetGraphConv.node_update_()
        • M3GNetGraphConv.state_update_()
      • MEGNetBlock
        • MEGNetBlock.forward()
      • MEGNetGraphConv
        • MEGNetGraphConv.edge_update_()
        • MEGNetGraphConv.forward()
        • MEGNetGraphConv.from_dims()
        • MEGNetGraphConv.node_update_()
        • MEGNetGraphConv.state_update_()
    • matgl.layers._readout module
      • ReduceReadOut
        • ReduceReadOut.forward()
      • Set2SetReadOut
        • Set2SetReadOut.forward()
      • WeightedReadOut
        • WeightedReadOut.forward()
      • WeightedReadOutPair
        • WeightedReadOutPair.forward()
    • matgl.layers._three_body module
      • ThreeBodyInteractions
        • ThreeBodyInteractions.forward()
      • combine_sbf_shf()
  • matgl.models package
    • matgl.models._m3gnet module
      • M3GNet
        • M3GNet.forward()
        • M3GNet.predict_structure()
    • matgl.models._megnet module
      • MEGNet
        • MEGNet.forward()
        • MEGNet.predict_structure()
    • matgl.models._wrappers module
      • TransformedTargetModel
        • TransformedTargetModel.forward()
        • TransformedTargetModel.predict_structure()
  • matgl.utils package
    • matgl.utils.cutoff module
      • cosine_cutoff()
      • polynomial_cutoff()
    • matgl.utils.io module
      • IOMixIn
        • IOMixIn.load()
        • IOMixIn.save()
        • IOMixIn.save_args()
      • RemoteFile
      • get_available_pretrained_models()
      • load_model()
    • matgl.utils.maths module
      • broadcast()
      • broadcast_states_to_atoms()
      • broadcast_states_to_bonds()
      • get_range_indices_from_n()
      • get_segment_indices_from_n()
      • repeat_with_n()
      • scatter_sum()
      • spherical_bessel_roots()
      • unsorted_segment_fraction()
    • matgl.utils.training module
      • MatglLightningModuleMixin
        • MatglLightningModuleMixin.configure_optimizers()
        • MatglLightningModuleMixin.on_test_model_eval()
        • MatglLightningModuleMixin.on_train_epoch_end()
        • MatglLightningModuleMixin.predict_step()
        • MatglLightningModuleMixin.test_step()
        • MatglLightningModuleMixin.training_step()
        • MatglLightningModuleMixin.validation_step()
      • ModelLightningModule
        • ModelLightningModule.forward()
        • ModelLightningModule.loss_fn()
        • ModelLightningModule.step()
      • PotentialLightningModule
        • PotentialLightningModule.forward()
        • PotentialLightningModule.loss_fn()
        • PotentialLightningModule.step()
      • xavier_init()

matgl.cli module

Command line interface for matgl.

matgl.cli.clear_cache(args)

Clear cache command.

  • Parameters: args – Args from CLI.

matgl.cli.main()

Handle main.

matgl.cli.predict_structure(args)

Use MatGL models to perform predictions on structures.

  • Parameters: args – Args from CLI.

matgl.cli.relax_structure(args)

Relax crystals.

  • Parameters: args – Args from CLI.

matgl.config module

Global configuration variables for matgl.

matgl.config.clear_cache(confirm: bool = True)

Deletes all files in the matgl.cache. This is used to clean out downloaded models.

  • Parameters: confirm – Whether to ask for confirmation. Default is True.
© Copyright 2022, Materials Virtual Lab