Merge Conflict (3a5d413c) · Commits · 钟慕尧 / deepchem

deepchem/feat/init.py

+1 −0

Original line number	Diff line number	Diff line
		@@ -42,6 +42,7 @@ from deepchem.feat.complex_featurizers import ComplexNeighborListFragmentAtomicC
		from deepchem.feat.material_featurizers import ElementPropertyFingerprint
		from deepchem.feat.material_featurizers import SineCoulombMatrix
		from deepchem.feat.material_featurizers import CGCNNFeaturizer
		from deepchem.feat.material_featurizers import ElemNetFeaturizer
		from deepchem.feat.material_featurizers import LCNNFeaturizer

		try:

deepchem/feat/graph_data.py

+5 −3

Original line number	Diff line number	Diff line
		@@ -146,9 +146,6 @@ class GraphData:

		src = self.edge_index[0]
		dst = self.edge_index[1]
		if self_loop:
		src = np.concatenate([src, np.arange(self.num_nodes)])
		dst = np.concatenate([dst, np.arange(self.num_nodes)])

		g = dgl.graph(
		(torch.from_numpy(src).long(), torch.from_numpy(dst).long()),
		@@ -161,6 +158,11 @@ class GraphData:
		if self.edge_features is not None:
		g.edata['edge_attr'] = torch.from_numpy(self.edge_features).float()

		if self_loop:
		# This assumes that the edge features for self loops are full-zero tensors
		# In the future we may want to support featurization for self loops
		g.add_edges(np.arange(self.num_nodes), np.arange(self.num_nodes))

		return g

deepchem/feat/material_featurizers/init.py

+1 −0

Original line number	Diff line number	Diff line
		@@ -5,4 +5,5 @@ Featurizers for inorganic crystals.
		from deepchem.feat.material_featurizers.element_property_fingerprint import ElementPropertyFingerprint
		from deepchem.feat.material_featurizers.sine_coulomb_matrix import SineCoulombMatrix
		from deepchem.feat.material_featurizers.cgcnn_featurizer import CGCNNFeaturizer
		from deepchem.feat.material_featurizers.elemnet_featurizer import ElemNetFeaturizer
		from deepchem.feat.material_featurizers.lcnn_featurizer import LCNNFeaturizer

deepchem/feat/material_featurizers/elemnet_featurizer.py

0 → 100644

+83 −0

Original line number	Diff line number	Diff line
		import numpy as np
		from typing import DefaultDict, Union

		from deepchem.utils.typing import PymatgenComposition
		from deepchem.feat import MaterialCompositionFeaturizer

		elements_tl = [
		'H', 'Li', 'Be', 'B', 'C', 'N', 'O', 'F', 'Na', 'Mg', 'Al', 'Si', 'P', 'S',
		'Cl', 'K', 'Ca', 'Sc', 'Ti', 'V', 'Cr', 'Mn', 'Fe', 'Co', 'Ni', 'Cu', 'Zn',
		'Ga', 'Ge', 'As', 'Se', 'Br', 'Kr', 'Rb', 'Sr', 'Y', 'Zr', 'Nb', 'Mo', 'Tc',
		'Ru', 'Rh', 'Pd', 'Ag', 'Cd', 'In', 'Sn', 'Sb', 'Te', 'I', 'Xe', 'Cs', 'Ba',
		'La', 'Ce', 'Pr', 'Nd', 'Pm', 'Sm', 'Eu', 'Gd', 'Tb', 'Dy', 'Ho', 'Er',
		'Tm', 'Yb', 'Lu', 'Hf', 'Ta', 'W', 'Re', 'Os', 'Ir', 'Pt', 'Au', 'Hg', 'Tl',
		'Pb', 'Bi', 'Ac', 'Th', 'Pa', 'U', 'Np', 'Pu'
		]


		class ElemNetFeaturizer(MaterialCompositionFeaturizer):
		"""
		Fixed size vector of length 86 containing raw fractional elemental
		compositions in the compound. The 86 chosen elements are based on the
		original implementation at https://github.com/NU-CUCIS/ElemNet.

		Returns a vector containing fractional compositions of each element
		in the compound.

		References
		----------
		.. [1] Jha, D., Ward, L., Paul, A. et al. Sci Rep 8, 17593 (2018).
		https://doi.org/10.1038/s41598-018-35934-y

		Examples
		--------
		>>> import pymatgen as mg
		>>> comp = "Fe2O3"
		>>> featurizer = ElemNetFeaturizer()
		>>> features = featurizer.featurize([comp])

		Notes
		-----
		This class requires Pymatgen to be installed.
		"""

		def get_vector(self, comp: DefaultDict) -> Union[np.ndarray, None]:
		"""
		Converts a dictionary containing element names and corresponding
		compositional fractions into a vector of fractions.

		Parameters
		----------
		comp: collections.defaultdict object
		Dictionary mapping element names to fractional compositions.

		Returns
		-------
		fractions: np.ndarray
		Vector of fractional compositions of each element.
		"""
		if all(e in elements_tl for e in comp):
		fractions = np.array([comp[e] if e in comp else 0 for e in elements_tl],
		np.float32)
		else:
		fractions = None
		return fractions

		def _featurize(self, composition: PymatgenComposition) -> np.ndarray:
		"""
		Calculate 86 dimensional vector containing fractional compositions of
		each element in the compound.

		Parameters
		----------
		composition: pymatgen.Composition object
		Composition object.

		Returns
		-------
		feats: np.ndarray
		86 dimensional vector containing fractional compositions of elements.
		"""
		fractions = composition.fractional_composition.get_el_amt_dict()
		feat = self.get_vector(fractions)
		return feat

deepchem/feat/tests/test_materials_featurizers.py

+14 −1

Original line number	Diff line number	Diff line
		@@ -4,7 +4,7 @@ Test featurizers for inorganic crystals.
		import unittest
		import numpy as np

		from deepchem.feat import ElementPropertyFingerprint, SineCoulombMatrix, CGCNNFeaturizer
		from deepchem.feat import ElementPropertyFingerprint, SineCoulombMatrix, CGCNNFeaturizer, ElemNetFeaturizer


		class TestMaterialFeaturizers(unittest.TestCase):
		@@ -83,3 +83,16 @@ class TestMaterialFeaturizers(unittest.TestCase):
		assert graph_features[0].node_features.shape == (1, 92)
		assert graph_features[0].edge_index.shape == (2, 6)
		assert graph_features[0].edge_features.shape == (6, 11)

		def test_elemnet_featurizer(self):
		"""
		Test ElemNetFeaturizer.
		"""

		featurizer = ElemNetFeaturizer()
		features = featurizer.featurize([self.formula])

		assert features.shape[1] == 86
		assert np.isclose(features[0][13], 0.6666667, atol=0.01)
		assert np.isclose(features[0][38], 0.33333334, atol=0.01)
		assert np.isclose(features.sum(), 1.0, atol=0.01)

Admin message