Loading deepchem/feat/materials_featurizers.py +102 −30 Original line number Diff line number Diff line Loading @@ -7,24 +7,44 @@ import numpy as np from deepchem.feat import Featurizer from deepchem.utils import pad_array from matminer.featurizers.composition import ElementProperty from pymatgen import Composition, Structure from pymatgen.analysis.graphs import StructureGraph from pymatgen.analysis.local_env import MinimumDistanceNN class ChemicalFingerprint(Featurizer): """ Chemical fingerprint of elemental properties from composition. Based on the data source chosen, returns properties and statistics (min, max, range, mean, standard deviation, mode) for a compound based on elemental stoichiometry. E.g., the average electronegativity of atoms in a crystal structure. The chemical fingerprint is a vector of these statistics. For a full list of properties and statistics, see ElementProperty(data_source).feature_labels(). This featurizer requires the optional dependencies pymatgen and matminer. It may be useful when only crystal compositions are available (and not 3D coordinates). References are given for each data source: MagPie data: Ward, L. et al. npj Comput Mater 2, 16028 (2016). https://doi.org/10.1038/npjcompumats.2016.28 Deml data: Deml, A. et al. Physical Review B 93, 085142 (2016). 10.1103/PhysRevB.93.085142 Matminer: Ward, L. et al. Comput. Mater. Sci. 152, 60-69 (2018). Pymatgen: Ong, S.P. et al. Comput. Mater. Sci. 68, 314-319 (2013). """ def __init__(self, data_source='matminer'): """ Parameters ---------- data_source : str, optional (default "matminer") Source for element property data ("matminer", "magpie", "deml") data_source : {"matminer", "magpie", "deml"} Source for element property data. """ def __init__(self, data_source='matminer'): self.data_source = data_source def _featurize(self, comp): Loading @@ -33,9 +53,14 @@ class ChemicalFingerprint(Featurizer): Parameters ---------- comp : Reduced formula of crystal. comp : str Reduced formula of crystal. """ from pymatgen import Composition from matminer.featurizers.composition import ElementProperty # Get pymatgen Composition object c = Composition(comp) Loading @@ -53,21 +78,36 @@ class SineCoulombMatrix(Featurizer): """ Calculate sine Coulomb matrix for crystals. Variant of Coulomb matrix for periodic crystals Faber et al. (Inter. J. Quantum Chem. 115, 16, 2015). A variant of Coulomb matrix for periodic crystals, based on Faber et al. Inter. J. Quantum Chem. 115, 16, (2015). The sine Coulomb matrix is identical to the Coulomb matrix, except that the inverse distance function is replaced by the inverse of sin**2 of the vector between sites which are periodic in the dimensions of the crystal lattice. Features are flattened into a vector of matrix eigenvalues by default for ML-readiness. To ensure that all feature vectors are equal length, the maximum number of atoms (eigenvalues) in the input dataset must be specified. This featurizer requires the optional dependency pymatgen. It may be useful when crystal structures with 3D coordinates are available. """ def __init__(self, max_atoms, eig=True): """ Parameters ---------- max_atoms : int Maximum number of atoms for any crystal in the dataset. Used to pad the Coulomb matrix. eig : bool (default True) Return flattened vector of matrix eigenvalues Return flattened vector of matrix eigenvalues. """ def __init__(self, max_atoms, eig=True): self.max_atoms = int(max_atoms) self.eig = eig Loading @@ -77,9 +117,13 @@ class SineCoulombMatrix(Featurizer): Parameters ---------- struct : pymatgen structure dictionary struct : dict pymatgen structure dictionary """ from pymatgen import Structure s = Structure.from_dict(struct) features = self.sine_coulomb_matrix(s) features = np.asarray(features) Loading @@ -93,6 +137,7 @@ class SineCoulombMatrix(Featurizer): Parameters ---------- s : pymatgen structure """ sites = s.sites Loading Loading @@ -132,15 +177,36 @@ class StructureGraphFeaturizer(Featurizer): """ Calculate structure graph for crystals. Create a graph representation of a crystal structure where atoms are nodes and connections between atoms (bonds) are edges. Bonds are determined by choosing a strategy for finding nearest neighbors from pymatgen.analysis.local_env. For periodic graphs, each edge belongs to a lattice image. The NetworkX package is used for graph representations. Hagberg, A. et al. SciPy2008, 11-15 (2008). This featurizer requires the optional dependency pymatgen. It may be useful when using graph network models and crystal graph convolutional networks. #TODO (@ncfrey) process graph features for models """ def __init__(self, strategy=None): """ Parameters ---------- strategy : pymatgen.analysis.local_env.NearNeighbors An instance of NearNeighbors that determines how graph is constructed. #TODO (@ncfrey) process graph features for models """ def __init__(self, strategy=MinimumDistanceNN()): if not strategy: from pymatgen.analysis.local_env import MinimumDistanceNN strategy = MinimumDistanceNN() self.strategy = strategy def _featurize(self, struct): Loading @@ -149,9 +215,13 @@ class StructureGraphFeaturizer(Featurizer): Parameters ---------- struct : pymatgen structure dictionary. struct : dict pymatgen structure dictionary. """ from pymatgen import Structure # Get pymatgen structure object s = Structure.from_dict(struct) Loading @@ -166,10 +236,12 @@ class StructureGraphFeaturizer(Featurizer): Parameters ---------- struct : pymatgen structure """ atom_features = np.array([site.specie.Z for site in struct], dtype='int32') from pymatgen.analysis.graphs import StructureGraph atom_features = np.array([site.specie.Z for site in struct], dtype='int32') sg = StructureGraph.with_local_env_strategy(struct, self.strategy) nodes = np.asarray(sg.graph.nodes) Loading deepchem/feat/tests/test_materials_featurizers.py +5 −1 Original line number Diff line number Diff line Loading @@ -52,8 +52,11 @@ class TestMaterialFeaturizers(unittest.TestCase): """ featurizer = ChemicalFingerprint(data_source='matminer') features = featurizer.featurize([self.formula]) assert isinstance(featurizer, ChemicalFingerprint) assert len(features[0]) == 65 assert np.allclose( features[0][:5], [2.16, 2.58, 0.42, 2.44, 0.29698485], atol=0.1) def testSCM(self): """ Loading @@ -63,4 +66,5 @@ class TestMaterialFeaturizers(unittest.TestCase): featurizer = SineCoulombMatrix(1) features = featurizer.featurize([self.struct_dict]) assert len(features) == 1 assert np.isclose(features[0], 1244, atol=.5) docs/featurizers.rst +17 −0 Original line number Diff line number Diff line Loading @@ -116,6 +116,23 @@ AtomConvFeaturizer .. autoclass:: deepchem.feat.NeighborListComplexAtomicCoordinates :members: ChemicalFingerprint ^^^^^^^^^^^^^^^^^^^ .. autoclass:: deepchem.feat.ChemicalFingerprint :members: SineCoulombMatrix ^^^^^^^^^^^^^^^^^ .. autoclass:: deepchem.feat.SineCoulombMatrix :members: StructureGraphFeaturizer ^^^^^^^^^^^^^^^^^^^^^^^^ .. autoclass:: deepchem.feat.StructureGraphFeaturizer :members: BindingPocketFeaturizer ----------------------- Loading Loading
deepchem/feat/materials_featurizers.py +102 −30 Original line number Diff line number Diff line Loading @@ -7,24 +7,44 @@ import numpy as np from deepchem.feat import Featurizer from deepchem.utils import pad_array from matminer.featurizers.composition import ElementProperty from pymatgen import Composition, Structure from pymatgen.analysis.graphs import StructureGraph from pymatgen.analysis.local_env import MinimumDistanceNN class ChemicalFingerprint(Featurizer): """ Chemical fingerprint of elemental properties from composition. Based on the data source chosen, returns properties and statistics (min, max, range, mean, standard deviation, mode) for a compound based on elemental stoichiometry. E.g., the average electronegativity of atoms in a crystal structure. The chemical fingerprint is a vector of these statistics. For a full list of properties and statistics, see ElementProperty(data_source).feature_labels(). This featurizer requires the optional dependencies pymatgen and matminer. It may be useful when only crystal compositions are available (and not 3D coordinates). References are given for each data source: MagPie data: Ward, L. et al. npj Comput Mater 2, 16028 (2016). https://doi.org/10.1038/npjcompumats.2016.28 Deml data: Deml, A. et al. Physical Review B 93, 085142 (2016). 10.1103/PhysRevB.93.085142 Matminer: Ward, L. et al. Comput. Mater. Sci. 152, 60-69 (2018). Pymatgen: Ong, S.P. et al. Comput. Mater. Sci. 68, 314-319 (2013). """ def __init__(self, data_source='matminer'): """ Parameters ---------- data_source : str, optional (default "matminer") Source for element property data ("matminer", "magpie", "deml") data_source : {"matminer", "magpie", "deml"} Source for element property data. """ def __init__(self, data_source='matminer'): self.data_source = data_source def _featurize(self, comp): Loading @@ -33,9 +53,14 @@ class ChemicalFingerprint(Featurizer): Parameters ---------- comp : Reduced formula of crystal. comp : str Reduced formula of crystal. """ from pymatgen import Composition from matminer.featurizers.composition import ElementProperty # Get pymatgen Composition object c = Composition(comp) Loading @@ -53,21 +78,36 @@ class SineCoulombMatrix(Featurizer): """ Calculate sine Coulomb matrix for crystals. Variant of Coulomb matrix for periodic crystals Faber et al. (Inter. J. Quantum Chem. 115, 16, 2015). A variant of Coulomb matrix for periodic crystals, based on Faber et al. Inter. J. Quantum Chem. 115, 16, (2015). The sine Coulomb matrix is identical to the Coulomb matrix, except that the inverse distance function is replaced by the inverse of sin**2 of the vector between sites which are periodic in the dimensions of the crystal lattice. Features are flattened into a vector of matrix eigenvalues by default for ML-readiness. To ensure that all feature vectors are equal length, the maximum number of atoms (eigenvalues) in the input dataset must be specified. This featurizer requires the optional dependency pymatgen. It may be useful when crystal structures with 3D coordinates are available. """ def __init__(self, max_atoms, eig=True): """ Parameters ---------- max_atoms : int Maximum number of atoms for any crystal in the dataset. Used to pad the Coulomb matrix. eig : bool (default True) Return flattened vector of matrix eigenvalues Return flattened vector of matrix eigenvalues. """ def __init__(self, max_atoms, eig=True): self.max_atoms = int(max_atoms) self.eig = eig Loading @@ -77,9 +117,13 @@ class SineCoulombMatrix(Featurizer): Parameters ---------- struct : pymatgen structure dictionary struct : dict pymatgen structure dictionary """ from pymatgen import Structure s = Structure.from_dict(struct) features = self.sine_coulomb_matrix(s) features = np.asarray(features) Loading @@ -93,6 +137,7 @@ class SineCoulombMatrix(Featurizer): Parameters ---------- s : pymatgen structure """ sites = s.sites Loading Loading @@ -132,15 +177,36 @@ class StructureGraphFeaturizer(Featurizer): """ Calculate structure graph for crystals. Create a graph representation of a crystal structure where atoms are nodes and connections between atoms (bonds) are edges. Bonds are determined by choosing a strategy for finding nearest neighbors from pymatgen.analysis.local_env. For periodic graphs, each edge belongs to a lattice image. The NetworkX package is used for graph representations. Hagberg, A. et al. SciPy2008, 11-15 (2008). This featurizer requires the optional dependency pymatgen. It may be useful when using graph network models and crystal graph convolutional networks. #TODO (@ncfrey) process graph features for models """ def __init__(self, strategy=None): """ Parameters ---------- strategy : pymatgen.analysis.local_env.NearNeighbors An instance of NearNeighbors that determines how graph is constructed. #TODO (@ncfrey) process graph features for models """ def __init__(self, strategy=MinimumDistanceNN()): if not strategy: from pymatgen.analysis.local_env import MinimumDistanceNN strategy = MinimumDistanceNN() self.strategy = strategy def _featurize(self, struct): Loading @@ -149,9 +215,13 @@ class StructureGraphFeaturizer(Featurizer): Parameters ---------- struct : pymatgen structure dictionary. struct : dict pymatgen structure dictionary. """ from pymatgen import Structure # Get pymatgen structure object s = Structure.from_dict(struct) Loading @@ -166,10 +236,12 @@ class StructureGraphFeaturizer(Featurizer): Parameters ---------- struct : pymatgen structure """ atom_features = np.array([site.specie.Z for site in struct], dtype='int32') from pymatgen.analysis.graphs import StructureGraph atom_features = np.array([site.specie.Z for site in struct], dtype='int32') sg = StructureGraph.with_local_env_strategy(struct, self.strategy) nodes = np.asarray(sg.graph.nodes) Loading
deepchem/feat/tests/test_materials_featurizers.py +5 −1 Original line number Diff line number Diff line Loading @@ -52,8 +52,11 @@ class TestMaterialFeaturizers(unittest.TestCase): """ featurizer = ChemicalFingerprint(data_source='matminer') features = featurizer.featurize([self.formula]) assert isinstance(featurizer, ChemicalFingerprint) assert len(features[0]) == 65 assert np.allclose( features[0][:5], [2.16, 2.58, 0.42, 2.44, 0.29698485], atol=0.1) def testSCM(self): """ Loading @@ -63,4 +66,5 @@ class TestMaterialFeaturizers(unittest.TestCase): featurizer = SineCoulombMatrix(1) features = featurizer.featurize([self.struct_dict]) assert len(features) == 1 assert np.isclose(features[0], 1244, atol=.5)
docs/featurizers.rst +17 −0 Original line number Diff line number Diff line Loading @@ -116,6 +116,23 @@ AtomConvFeaturizer .. autoclass:: deepchem.feat.NeighborListComplexAtomicCoordinates :members: ChemicalFingerprint ^^^^^^^^^^^^^^^^^^^ .. autoclass:: deepchem.feat.ChemicalFingerprint :members: SineCoulombMatrix ^^^^^^^^^^^^^^^^^ .. autoclass:: deepchem.feat.SineCoulombMatrix :members: StructureGraphFeaturizer ^^^^^^^^^^^^^^^^^^^^^^^^ .. autoclass:: deepchem.feat.StructureGraphFeaturizer :members: BindingPocketFeaturizer ----------------------- Loading
