yapfed (9327bc7c) · Commits · 钟慕尧 / deepchem

deepchem/data/datasets.py

+4 −8

Original line number	Diff line number	Diff line
		@@ -469,8 +469,7 @@ class NumpyDataset(Dataset):
		w = np.concatenate([w, dataset.w], axis=0)
		ids = np.concatenate(
		[ids, dataset.ids],
		axis=0,
		)
		axis=0,)

		return NumpyDataset(X, y, w, ids, n_tasks=y.shape[1])

		@@ -648,8 +647,7 @@ class DiskDataset(Dataset):
		if not len(self.metadata_df):
		raise ValueError("No data in dataset.")
		sample_X = load_from_disk(
		os.path.join(self.data_dir,
		next(self.metadata_df.iterrows())[1]['X']))
		os.path.join(self.data_dir, next(self.metadata_df.iterrows())[1]['X']))
		return np.shape(sample_X)[1:]

		def get_shard_size(self):
		@@ -657,8 +655,7 @@ class DiskDataset(Dataset):
		if not len(self.metadata_df):
		raise ValueError("No data in dataset.")
		sample_y = load_from_disk(
		os.path.join(self.data_dir,
		next(self.metadata_df.iterrows())[1]['y']))
		os.path.join(self.data_dir, next(self.metadata_df.iterrows())[1]['y']))
		return len(sample_y)

		def _get_metadata_filename(self):
		@@ -789,7 +786,6 @@ class DiskDataset(Dataset):
		else:
		shard_batch_size = batch_size


		if n_shard_samples == 0:
		cur_shard += 1
		if batch_size is None:

deepchem/models/tensorgraph/layers.py

+21 −18

Original line number	Diff line number	Diff line
		@@ -735,8 +735,7 @@ class Squeeze(Layer):
		self._shape = [i for i in parent_shape if i != 1]
		else:
		self._shape = [
		parent_shape[i]
		for i in range(len(parent_shape))
		parent_shape[i] for i in range(len(parent_shape))
		if i not in squeeze_dims
		]
		except:
		@@ -1561,6 +1560,7 @@ class ReduceMean(Layer):
		self.out_tensor = out_tensor
		return out_tensor


		class ReduceMax(Layer):

		def __init__(self, in_layers=None, axis=None, **kwargs):
		@@ -1591,6 +1591,7 @@ class ReduceMax(Layer):
		self.out_tensor = out_tensor
		return out_tensor


		class ToFloat(Layer):

		def __init__(self, in_layers=None, **kwargs):
		@@ -2093,8 +2094,8 @@ class MaxPool1D(Layer):
		super(MaxPool1D, self).__init__(**kwargs)
		try:
		parent_shape = self.in_layers[0].shape
		self._shape = tuple(
		None if p is None else p // s for p, s in zip(parent_shape, strides))
		self._shape = tuple(None if p is None else p // s
		for p, s in zip(parent_shape, strides))
		except:
		pass

		@@ -2125,8 +2126,8 @@ class MaxPool2D(Layer):
		super(MaxPool2D, self).__init__(**kwargs)
		try:
		parent_shape = self.in_layers[0].shape
		self._shape = tuple(
		None if p is None else p // s for p, s in zip(parent_shape, strides))
		self._shape = tuple(None if p is None else p // s
		for p, s in zip(parent_shape, strides))
		except:
		pass

		@@ -2172,8 +2173,8 @@ class MaxPool3D(Layer):
		super(MaxPool3D, self).__init__(**kwargs)
		try:
		parent_shape = self.in_layers[0].shape
		self._shape = tuple(
		None if p is None else p // s for p, s in zip(parent_shape, strides))
		self._shape = tuple(None if p is None else p // s
		for p, s in zip(parent_shape, strides))
		except:
		pass

		@@ -2938,13 +2939,15 @@ class VinaFreeEnergy(Layer):

		def hydrophobic(self, d):
		"""Computes Autodock Vina's hydrophobic interaction term."""
		out_tensor = tf.where(d < 0.5, tf.ones_like(d),
		out_tensor = tf.where(d < 0.5,
		tf.ones_like(d),
		tf.where(d < 1.5, 1.5 - d, tf.zeros_like(d)))
		return out_tensor

		def hydrogen_bond(self, d):
		"""Computes Autodock Vina's hydrogen bond interaction term."""
		out_tensor = tf.where(d < -0.7, tf.ones_like(d),
		out_tensor = tf.where(d < -0.7,
		tf.ones_like(d),
		tf.where(d < 0, (1.0 / 0.7) * (0 - d),
		tf.zeros_like(d)))
		return out_tensor
		@@ -3321,8 +3324,8 @@ class NeighborList(Layer):
		mesh_args = [tf.range(start, stop, nbr_cutoff) for _ in range(self.ndim)]
		return tf.to_float(
		tf.reshape(
		tf.transpose(tf.stack(tf.meshgrid(*mesh_args))),
		(self.n_cells, self.ndim)))
		tf.transpose(tf.stack(tf.meshgrid(*mesh_args))), (self.n_cells,
		self.ndim)))


		class Dropout(Layer):
		@@ -3609,8 +3612,8 @@ class AtomicConvolution(Layer):
		example_tensors = tf.unstack(X, axis=0)
		example_nbrs = tf.unstack(nbr_indices, axis=0)
		all_nbr_coords = []
		for example, (example_tensor, example_nbr) in enumerate(
		zip(example_tensors, example_nbrs)):
		for example, (example_tensor,
		example_nbr) in enumerate(zip(example_tensors, example_nbrs)):
		nbr_coords = tf.gather(example_tensor, example_nbr)
		all_nbr_coords.append(nbr_coords)
		neighbors = tf.stack(all_nbr_coords)
		@@ -4176,13 +4179,13 @@ class GraphCNN(Layer):
		no_features = V.get_shape()[2].value
		W = tf.get_variable(
		'%s_weights' % self.name, [no_features * no_A, self.num_filters],
		initializer=tf.truncated_normal_initializer(
		stddev=math.sqrt(1.0 / (no_features * (no_A + 1) * 1.0))),
		initializer=tf.truncated_normal_initializer(stddev=math.sqrt(
		1.0 / (no_features * (no_A + 1) * 1.0))),
		dtype=tf.float32)
		W_I = tf.get_variable(
		'%s_weights_I' % self.name, [no_features, self.num_filters],
		initializer=tf.truncated_normal_initializer(
		stddev=math.sqrt(1.0 / (no_features * (no_A + 1) * 1.0))),
		initializer=tf.truncated_normal_initializer(stddev=math.sqrt(
		1.0 / (no_features * (no_A + 1) * 1.0))),
		dtype=tf.float32)

		b = tf.get_variable(

deepchem/models/tensorgraph/models/graph_models.py

+27 −30

Original line number	Diff line number	Diff line
		@@ -172,8 +172,8 @@ class WeaveTensorGraph(TensorGraph):
		atom_feat.append(mol.get_atom_features())
		# pair features
		pair_feat.append(
		np.reshape(mol.get_pair_features(),
		(n_atoms * n_atoms, self.n_pair_feat)))
		np.reshape(mol.get_pair_features(), (n_atoms * n_atoms,
		self.n_pair_feat)))

		feed_dict[self.atom_features] = np.concatenate(atom_feat, axis=0)
		feed_dict[self.pair_features] = np.concatenate(pair_feat, axis=0)
		@@ -184,9 +184,7 @@ class WeaveTensorGraph(TensorGraph):

		def predict_on_generator(self, generator, transformers=[], outputs=None):
		out = super(WeaveTensorGraph, self).predict_on_generator(
		generator,
		transformers=[],
		outputs=outputs)
		generator, transformers=[], outputs=outputs)
		if outputs is None:
		outputs = self.outputs
		if len(outputs) > 1:
		@@ -196,7 +194,6 @@ class WeaveTensorGraph(TensorGraph):
		return out



		class DTNNTensorGraph(TensorGraph):

		def __init__(self,
		@@ -320,8 +317,8 @@ class DTNNTensorGraph(TensorGraph):
		num_atoms = list(map(sum, X_b.astype(bool)[:, :, 0]))
		atom_number = [
		np.round(
		np.power(2 * np.diag(X_b[i, :num_atoms[i], :num_atoms[i]]),
		1 / 2.4)).astype(int) for i in range(len(num_atoms))
		np.power(2 * np.diag(X_b[i, :num_atoms[i], :num_atoms[i]]), 1 /
		2.4)).astype(int) for i in range(len(num_atoms))
		]
		start = 0
		for im, molecule in enumerate(atom_number):
		@@ -360,6 +357,7 @@ class DTNNTensorGraph(TensorGraph):
		retval = np.concatenate(retval, axis=-1)
		return undo_transforms(retval, transformers)


		class DAGTensorGraph(TensorGraph):

		def __init__(self,
		@@ -511,9 +509,7 @@ class DAGTensorGraph(TensorGraph):

		def predict_on_generator(self, generator, transformers=[], outputs=None):
		out = super(DAGTensorGraph, self).predict_on_generator(
		generator,
		transformers=[],
		outputs=outputs)
		generator, transformers=[], outputs=outputs)
		if outputs is None:
		outputs = self.outputs
		if len(outputs) > 1:
		@@ -522,6 +518,7 @@ class DAGTensorGraph(TensorGraph):
		out = undo_transforms(out, transformers)
		return out


		class PetroskiSuchTensorGraph(TensorGraph):
		"""
		Model from Robust Spatial Filtering with Graph Convolutional Neural Networks
		@@ -1093,8 +1090,8 @@ class MPNNTensorGraph(TensorGraph):
		atom_feat.append(mol.get_atom_features())
		# pair features
		pair_feat.append(
		np.reshape(mol.get_pair_features(),
		(n_atoms * n_atoms, self.n_pair_feat)))
		np.reshape(mol.get_pair_features(), (n_atoms * n_atoms,
		self.n_pair_feat)))

		feed_dict[self.atom_features] = np.concatenate(atom_feat, axis=0)
		feed_dict[self.pair_features] = np.concatenate(pair_feat, axis=0)

deepchem/models/tensorgraph/models/text_cnn.py

+11 −14

Original line number	Diff line number	Diff line
		@@ -175,9 +175,8 @@ class TextCNNTensorGraph(TensorGraph):
		padding='valid',
		in_layers=[self.Embedding]))
		# Max-over-time pooling
		self.pooled_outputs.append(ReduceMax(
		axis=1,
		in_layers=[self.conv_layers[-1]]))
		self.pooled_outputs.append(
		ReduceMax(axis=1, in_layers=[self.conv_layers[-1]]))
		# Concat features from all filters(one feature per filter)
		concat_outputs = Concat(axis=1, in_layers=self.pooled_outputs)
		dropout = Dropout(dropout_prob=self.dropout, in_layers=[concat_outputs])
		@@ -273,9 +272,7 @@ class TextCNNTensorGraph(TensorGraph):

		def predict_on_generator(self, generator, transformers=[], outputs=None):
		out = super(TextCNNTensorGraph, self).predict_on_generator(
		generator,
		transformers=[],
		outputs=outputs)
		generator, transformers=[], outputs=outputs)
		if outputs is None:
		outputs = self.outputs
		if len(outputs) > 1:

deepchem/molnet/run_benchmark_models.py

+14 −12

Original line number	Diff line number	Diff line
		@@ -262,8 +262,10 @@ def benchmark_classification(train_dataset,
		filter_sizes = hyper_parameters['filter_sizes']
		num_filters = hyper_parameters['num_filters']

		all_data = deepchem.data.DiskDataset.merge([train_dataset, valid_dataset, test_dataset])
		char_dict, length = deepchem.models.TextCNNTensorGraph.build_char_dict(all_data)
		all_data = deepchem.data.DiskDataset.merge(
		[train_dataset, valid_dataset, test_dataset])
		char_dict, length = deepchem.models.TextCNNTensorGraph.build_char_dict(
		all_data)

		model = deepchem.models.TextCNNTensorGraph(
		len(tasks),
		@@ -505,7 +507,6 @@ def benchmark_regression(train_dataset,
		n_filters = hyper_parameters['n_filters']
		n_fully_connected_nodes = hyper_parameters['n_fully_connected_nodes']


		model = deepchem.models.GraphConvTensorGraph(
		len(tasks),
		graph_conv_layers=[n_filters] * 2,
		@@ -596,7 +597,8 @@ def benchmark_regression(train_dataset,
		filter_sizes = hyper_parameters['filter_sizes']
		num_filters = hyper_parameters['num_filters']

		char_dict, length = deepchem.models.TextCNNTensorGraph.build_char_dict(train_dataset)
		char_dict, length = deepchem.models.TextCNNTensorGraph.build_char_dict(
		train_dataset)

		model = deepchem.models.TextCNNTensorGraph(
		len(tasks),

Admin message