Fixed errors in atomic convolutions (ad154738) · Commits · 钟慕尧 / deepchem

deepchem/feat/graph_features.py

+13 −21

Original line number	Diff line number	Diff line
		@@ -604,18 +604,11 @@ class AtomicConvFeaturizer(ComplexNeighborListFragmentAtomicCoordinates):
		self.labels = labels

		def featurize_complexes(self, mol_files, protein_files):
		pool = multiprocessing.Pool()
		results = []
		for i, (mol_file, protein_pdb) in enumerate(zip(mol_files, protein_files)):
		log_message = "Featurizing %d / %d" % (i, len(mol_files))
		results.append(
		pool.apply_async(_featurize_complex,
		(self, mol_file, protein_pdb, log_message)))
		pool.close()
		features = []
		failures = []
		for ind, result in enumerate(results):
		new_features = result.get()
		for i, (mol_file, protein_pdb) in enumerate(zip(mol_files, protein_files)):
		logging.info("Featurizing %d / %d" % (i, len(mol_files)))
		new_features = self._featurize_complex(mol_file, protein_pdb)
		# Handle loading failures which return None
		if new_features is not None:
		features.append(new_features)
		@@ -630,20 +623,19 @@ class AtomicConvFeaturizer(ComplexNeighborListFragmentAtomicCoordinates):
		self.atomic_conv_model.fit(dataset, nb_epoch=self.epochs)

		# Add the Atomic Convolution layers to fetches
		layers_to_fetch = list()
		for layer in self.atomic_conv_model.layers.values():
		if isinstance(layer, dc.models.atomic_conv.AtomicConvolution):
		layers_to_fetch.append(layer)
		layers_to_fetch = [
		self.atomic_conv_model._frag1_conv, self.atomic_conv_model._frag2_conv,
		self.atomic_conv_model._complex_conv
		]

		# Extract the atomic convolution features
		atomic_conv_features = list()
		feed_dict_generator = self.atomic_conv_model.default_generator(
		batch_generator = self.atomic_conv_model.default_generator(
		dataset=dataset, epochs=1)

		for feed_dict in self.atomic_conv_model._create_feed_dicts(
		feed_dict_generator, training=False):
		frag1_conv, frag2_conv, complex_conv = self.atomic_conv_model._run_graph(
		outputs=layers_to_fetch, feed_dict=feed_dict, training=False)
		for X, y, w in batch_generator:
		frag1_conv, frag2_conv, complex_conv = self.atomic_conv_model.predict_on_batch(
		X, outputs=layers_to_fetch)
		concatenated = np.concatenate(
		[frag1_conv, frag2_conv, complex_conv], axis=1)
		atomic_conv_features.append(concatenated)

deepchem/models/atomic_conv.py

+8 −6

Original line number	Diff line number	Diff line
		@@ -221,20 +221,22 @@ class AtomicConvModel(KerasModel):
		complex_nbrs_z = Input(shape=(complex_num_atoms, max_num_neighbors))
		complex_z = Input(shape=(complex_num_atoms,))

		frag1_conv = AtomicConvolution(
		self._frag1_conv = AtomicConvolution(
		atom_types=self.atom_types, radial_params=rp,
		boxsize=None)([frag1_X, frag1_nbrs, frag1_nbrs_z])

		frag2_conv = AtomicConvolution(
		self._frag2_conv = AtomicConvolution(
		atom_types=self.atom_types, radial_params=rp,
		boxsize=None)([frag2_X, frag2_nbrs, frag2_nbrs_z])

		complex_conv = AtomicConvolution(
		self._complex_conv = AtomicConvolution(
		atom_types=self.atom_types, radial_params=rp,
		boxsize=None)([complex_X, complex_nbrs, complex_nbrs_z])

		score = AtomicConvScore(self.atom_types, layer_sizes)(
		[frag1_conv, frag2_conv, complex_conv, frag1_z, frag2_z, complex_z])
		score = AtomicConvScore(self.atom_types, layer_sizes)([
		self._frag1_conv, self._frag2_conv, self._complex_conv, frag1_z,
		frag2_z, complex_z
		])

		model = tf.keras.Model(
		inputs=[

deepchem/models/keras_model.py

+2 −3

Original line number	Diff line number	Diff line
		@@ -618,8 +618,7 @@ class KerasModel(Model):
		a NumPy array of the model produces a single output, or a list of arrays
		if it produces multiple outputs
		"""
		dataset = NumpyDataset(X=X, y=None)
		return self.predict(dataset, transformers, outputs)
		return self.predict_on_generator([(X, None, None)], transformers, outputs)

		def predict_uncertainty_on_batch(self, X, masks=50):
		"""

deepchem/models/tests/test_atomic_conv.py

+2 −2

Original line number	Diff line number	Diff line
		@@ -107,9 +107,9 @@ class TestAtomicConv(unittest.TestCase):
		"""A simple test for running an atomic convolution on featurized data."""
		dir_path = os.path.dirname(os.path.realpath(__file__))
		ligand_file = os.path.join(dir_path,
		"../../../feat/tests/data/3zso_ligand_hyd.pdb")
		"../../feat/tests/data/3zso_ligand_hyd.pdb")
		protein_file = os.path.join(dir_path,
		"../../../feat/tests/data/3zso_protein.pdb")
		"../../feat/tests/data/3zso_protein.pdb")
		# Pulled from PDB files. For larger datasets with more PDBs, would use
		# max num atoms instead of exact.
		frag1_num_atoms = 44 # for ligand atoms

Admin message