refinement and changes (96de0a46) · Commits · 钟慕尧 / deepchem

deepchem/data/datasets.py

+0 −1

Original line number	Diff line number	Diff line
		@@ -80,7 +80,6 @@ def pad_batch(batch_size, X_b, y_b, w_b, ids_b):
		num_samples = len(X_b)
		if num_samples == batch_size:
		return (X_b, y_b, w_b, ids_b)
		else:
		# By invariant of when this is called, can assume num_samples > 0
		# and num_samples < batch_size
		if len(X_b.shape) > 1:

deepchem/feat/graph_features.py

+7 −26

Original line number	Diff line number	Diff line
		@@ -163,19 +163,7 @@ def atom_features(atom, bool_id_feat=False, explicit_H=False):
		'Pb',
		'Unknown'
		]) + one_of_k_encoding(atom.GetDegree(),
		[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10])
		# In case of explicit hydrogen(QM8, QM9), avoid calling `GetTotalNumHs`
		if explicit_H:
		results = results + \
		one_of_k_encoding_unk(atom.GetImplicitValence(), [0, 1, 2, 3, 4, 5, 6]) + \
		[atom.GetFormalCharge(), atom.GetNumRadicalElectrons()] + \
		one_of_k_encoding_unk(atom.GetHybridization(), [
		Chem.rdchem.HybridizationType.SP, Chem.rdchem.HybridizationType.SP2,
		Chem.rdchem.HybridizationType.SP3, Chem.rdchem.HybridizationType.
		SP3D, Chem.rdchem.HybridizationType.SP3D2
		]) + [atom.GetIsAromatic()]
		else:
		results = results + one_of_k_encoding_unk(atom.GetTotalNumHs(), [0, 1, 2, 3, 4]) + \
		[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10]) + \
		one_of_k_encoding_unk(atom.GetImplicitValence(), [0, 1, 2, 3, 4, 5, 6]) + \
		[atom.GetFormalCharge(), atom.GetNumRadicalElectrons()] + \
		one_of_k_encoding_unk(atom.GetHybridization(), [
		@@ -183,6 +171,9 @@ def atom_features(atom, bool_id_feat=False, explicit_H=False):
		Chem.rdchem.HybridizationType.SP3, Chem.rdchem.HybridizationType.
		SP3D, Chem.rdchem.HybridizationType.SP3D2
		]) + [atom.GetIsAromatic()]
		# In case of explicit hydrogen(QM8, QM9), avoid calling `GetTotalNumHs`
		if not explicit_H:
		results = results + one_of_k_encoding_unk(atom.GetTotalNumHs(), [0, 1, 2, 3, 4])

		return np.array(results)

		@@ -290,27 +281,17 @@ class WeaveFeaturizer(Featurizer):

		name = ['weave_mol']

		def __init__(self, graph_distance=True, explicit_H=None):
		def __init__(self, graph_distance=True, explicit_H=False):
		# Distance is either graph distance(True) or Euclidean distance(False,
		# only support datasets providing Cartesian coordinates)
		self.graph_distance = graph_distance
		# Set dtype
		self.dtype = object
		# Check if there are explicit hydrogens, default to be False
		self.check_H = False
		if explicit_H is None:
		self.explicit_H = False
		# Set to True if explicit hydrogen is not specified
		self.check_H = True
		# If includes explicit hydrogens
		self.explicit_H = explicit_H

		def _featurize(self, mol):
		"""Encodes mol as a WeaveMol object."""
		# Check hydrogen in the molecule
		if self.check_H and not self.explicit_H:
		for a in mol.GetAtoms():
		if a.GetSymbol() == 'H':
		self.explicit_H = True
		break
		# Atom features
		idx_nodes = [(a.GetIdx(), atom_features(a, explicit_H=self.explicit_H))
		for a in mol.GetAtoms()]

deepchem/models/tensorgraph/graph_layers.py

+16 −6

Original line number	Diff line number	Diff line
		@@ -873,7 +873,6 @@ class MessagePassing(Layer):
		self.out_tensor = out_tensor
		return out_tensor


		class EdgeNetwork(object):
		""" Submodule for Message Passing """

		@@ -971,11 +970,11 @@ class SetGather(Layer):
		atom_features = in_layers[0].out_tensor
		atom_split = in_layers[1].out_tensor

		c = tf.zeros((self.batch_size, self.n_hidden))
		h = tf.zeros((self.batch_size, self.n_hidden))
		self.c = tf.zeros((self.batch_size, self.n_hidden))
		self.h = tf.zeros((self.batch_size, self.n_hidden))

		for i in range(self.M):
		q_expanded = tf.gather(h, atom_split)
		q_expanded = tf.gather(self.h, atom_split)
		e = tf.reduce_sum(atom_features * q_expanded, 1)
		e_mols = tf.dynamic_partition(e, atom_split, self.batch_size)
		# Add another value(~-Inf) to prevent error in softmax
		@@ -985,8 +984,8 @@ class SetGather(Layer):
		a = tf.concat([tf.nn.softmax(e_mol)[:-1] for e_mol in e_mols], 0)
		r = tf.segment_sum(tf.reshape(a, [-1, 1]) * atom_features, atom_split)
		# Model using this layer must set pad_batches=True
		q_star = tf.concat([h, r], axis=1)
		h, c = self.LSTMStep(q_star, c)
		q_star = tf.concat([self.h, r], axis=1)
		self.h, self.c = self.LSTMStep(q_star, self.c)

		out_tensor = q_star
		if set_tensors:
		@@ -1005,3 +1004,14 @@ class SetGather(Layer):
		h_out = o * tf.nn.tanh(c_out)

		return h_out, c_out


		def none_tensors(self):
		self.out_tensor = None
		self.h = None
		self.c = None
		saved_tensors = [self.out_tensor, self.h, self.c]
		return saved_tensors

		def set_tensors(self, tensors):
		self.out_tensor, self.h, self.c = tensors

examples/results.csv

deleted100644 → 0

+0 −5

Original line number	Diff line number	Diff line
		qm7b,random,regression,dtnn,mean-pearson_r2_score,train,0.93149678885630138,valid,0.9070831996898977,time_for_running,628.2365071773529
		sampl,random,regression,krr,mean-pearson_r2_score,train,0.99985273293498711,valid,0.81413385193207155,time_for_running,0.484935998916626
		ppb,random,regression,krr,mean-pearson_r2_score,train,0.99436827538907779,valid,0.36149235755484083,time_for_running,1.822425127029419
		qm7,random,regression,ani,mean-pearson_r2_score,train,0.90124742726047558,valid,0.86827080462582273,time_for_running,1619.0067789554596
		qm7,random,regression,ani,mean-pearson_r2_score,train,0.96049723833152711,valid,0.9318116152192033,time_for_running,1144.9879438877106

Admin message