Ran yapf to update formatting (f7a0d2b0) · Commits · 钟慕尧 / deepchem

deepchem/models/tensorflow_models/fcnet.py

+59 −27

Original line number	Diff line number	Diff line
		@@ -18,11 +18,12 @@ from deepchem.models.tensorflow_models import TensorflowClassifier
		from deepchem.models.tensorflow_models import TensorflowRegressor
		from deepchem.metrics import to_one_hot


		from deepchem.models.tensorgraph.tensor_graph import TensorGraph, TFWrapper
		from deepchem.models.tensorgraph.layers import Feature, Label, Weights, WeightedError, Dense, Dropout, WeightDecay, Reshape, SoftMaxCrossEntropy, L2Loss


		class TensorGraphMultiTaskClassifier(TensorGraph):

		def __init__(self,
		n_tasks,
		n_features,
		@@ -73,27 +74,41 @@ class TensorGraphMultiTaskClassifier(TensorGraph):

		# Add the dense layers

		for size, weight_stddev, bias_const, dropout in zip(layer_sizes, weight_init_stddevs, bias_init_consts, dropouts):
		layer = Dense(in_layers=[prev_layer], out_channels=size, activation_fn=tf.nn.relu,
		weights_initializer=TFWrapper(tf.truncated_normal_initializer, stddev=weight_stddev),
		biases_initializer=TFWrapper(tf.constant_initializer, value=bias_const))
		for size, weight_stddev, bias_const, dropout in zip(
		layer_sizes, weight_init_stddevs, bias_init_consts, dropouts):
		layer = Dense(
		in_layers=[prev_layer],
		out_channels=size,
		activation_fn=tf.nn.relu,
		weights_initializer=TFWrapper(
		tf.truncated_normal_initializer, stddev=weight_stddev),
		biases_initializer=TFWrapper(
		tf.constant_initializer, value=bias_const))
		if dropout > 0.0:
		layer = Dropout(dropout, in_layers=[layer])
		prev_layer = layer

		# Compute the loss function for each label.

		output = Reshape(shape=(-1, n_tasks, n_classes), in_layers=[Dense(in_layers=[prev_layer], out_channels=n_tasks*n_classes)])
		output = Reshape(
		shape=(-1, n_tasks, n_classes),
		in_layers=[
		Dense(in_layers=[prev_layer], out_channels=n_tasks * n_classes)
		])
		self.add_output(output)
		labels = Label(shape=(None, n_tasks, n_classes))
		weights = Weights(shape=(None, n_tasks))
		loss = Reshape(shape=(-1, n_tasks), in_layers=[SoftMaxCrossEntropy(in_layers=[labels, output])])
		loss = Reshape(
		shape=(-1, n_tasks),
		in_layers=[SoftMaxCrossEntropy(in_layers=[labels, output])])
		weighted_loss = WeightedError(in_layers=[loss, weights])
		if weight_decay_penalty != 0.0:
		weighted_loss = WeightDecay(weight_decay_penalty, weight_decay_penalty_type, in_layers=[weighted_loss])
		weighted_loss = WeightDecay(
		weight_decay_penalty,
		weight_decay_penalty_type,
		in_layers=[weighted_loss])
		self.set_loss(weighted_loss)


		def default_generator(self,
		dataset,
		epochs=1,
		@@ -106,7 +121,9 @@ class TensorGraphMultiTaskClassifier(TensorGraph):
		pad_batches=pad_batches):
		feed_dict = dict()
		if y_b is not None and not predict:
		feed_dict[self.labels[0]] = to_one_hot(y_b.flatten(), self.n_classes).reshape(-1, self.n_tasks, self.n_classes)
		feed_dict[self.labels[0]] = to_one_hot(
		y_b.flatten(), self.n_classes).reshape(-1, self.n_tasks,
		self.n_classes)
		if X_b is not None:
		feed_dict[self.features[0]] = X_b
		if w_b is not None and not predict:
		@@ -114,9 +131,8 @@ class TensorGraphMultiTaskClassifier(TensorGraph):
		yield feed_dict




		class TensorGraphMultiTaskRegressor(TensorGraph):

		def __init__(self,
		n_tasks,
		n_features,
		@@ -164,29 +180,47 @@ class TensorGraphMultiTaskRegressor(TensorGraph):

		# Add the dense layers

		for size, weight_stddev, bias_const, dropout in zip(layer_sizes, weight_init_stddevs, bias_init_consts, dropouts):
		layer = Dense(in_layers=[prev_layer], out_channels=size, activation_fn=tf.nn.relu,
		weights_initializer=TFWrapper(tf.truncated_normal_initializer, stddev=weight_stddev),
		biases_initializer=TFWrapper(tf.constant_initializer, value=bias_const))
		for size, weight_stddev, bias_const, dropout in zip(
		layer_sizes, weight_init_stddevs, bias_init_consts, dropouts):
		layer = Dense(
		in_layers=[prev_layer],
		out_channels=size,
		activation_fn=tf.nn.relu,
		weights_initializer=TFWrapper(
		tf.truncated_normal_initializer, stddev=weight_stddev),
		biases_initializer=TFWrapper(
		tf.constant_initializer, value=bias_const))
		if dropout > 0.0:
		layer = Dropout(dropout, in_layers=[layer])
		prev_layer = layer

		# Compute the loss function for each label.

		output = Reshape(shape=(-1, n_tasks, 1), in_layers=[Dense(in_layers=[prev_layer], out_channels=n_tasks,
		weights_initializer=TFWrapper(tf.truncated_normal_initializer, stddev=weight_init_stddevs[-1]),
		biases_initializer=TFWrapper(tf.constant_initializer, value=bias_init_consts[-1]))])
		output = Reshape(
		shape=(-1, n_tasks, 1),
		in_layers=[
		Dense(
		in_layers=[prev_layer],
		out_channels=n_tasks,
		weights_initializer=TFWrapper(
		tf.truncated_normal_initializer,
		stddev=weight_init_stddevs[-1]),
		biases_initializer=TFWrapper(
		tf.constant_initializer, value=bias_init_consts[-1]))
		])
		self.add_output(output)
		labels = Label(shape=(None, n_tasks, 1))
		weights = Weights(shape=(None, n_tasks))
		loss = Reshape(shape=(-1, n_tasks), in_layers=[L2Loss(in_layers=[labels, output])])
		loss = Reshape(
		shape=(-1, n_tasks), in_layers=[L2Loss(in_layers=[labels, output])])
		weighted_loss = WeightedError(in_layers=[loss, weights])
		if weight_decay_penalty != 0.0:
		weighted_loss = WeightDecay(weight_decay_penalty, weight_decay_penalty_type, in_layers=[weighted_loss])
		weighted_loss = WeightDecay(
		weight_decay_penalty,
		weight_decay_penalty_type,
		in_layers=[weighted_loss])
		self.set_loss(weighted_loss)


		def default_generator(self,
		dataset,
		epochs=1,
		@@ -207,8 +241,6 @@ class TensorGraphMultiTaskRegressor(TensorGraph):
		yield feed_dict




		class TensorGraphMultiTaskFitRegressor(TensorGraphMultiTaskRegressor):
		"""Implements a TensorGraphMultiTaskRegressor that performs on-the-fly transformation during fit/predict.

		@@ -269,7 +301,6 @@ class TensorGraphMultiTaskFitRegressor(TensorGraphMultiTaskRegressor):
		print("n_features after fit_transform: %d" % int(n_features))
		super().__init__(n_tasks, n_features, batch_size=batch_size, **kwargs)


		def default_generator(self,
		dataset,
		epochs=1,
		@@ -292,8 +323,8 @@ class TensorGraphMultiTaskFitRegressor(TensorGraphMultiTaskRegressor):
		feed_dict[self.task_weights[0]] = w_b
		yield feed_dict


		def predict_proba_on_generator(self, generator, transformers=[]):

		def transform_generator():
		for feed_dict in generator:
		X = feed_dict[self.features[0]]
		@@ -303,8 +334,9 @@ class TensorGraphMultiTaskFitRegressor(TensorGraphMultiTaskRegressor):
		X_t = transformer.X_transform(X_t)
		feed_dict[self.features[0]] = X_t
		yield feed_dict
		return super().predict_proba_on_generator(transform_generator(), transformers)

		return super().predict_proba_on_generator(transform_generator(),
		transformers)


		class TensorflowMultiTaskClassifier(TensorflowClassifier):

deepchem/models/tensorgraph/layers.py

+4 −2

Original line number	Diff line number	Diff line
		@@ -1287,6 +1287,7 @@ class NeighborList(Layer):
		tf.transpose(tf.stack(tf.meshgrid(*mesh_args))), (self.n_cells,
		self.ndim)))


		class Dropout(Layer):

		def __init__(self, dropout_prob, **kwargs):
		@@ -1330,7 +1331,8 @@ class WeightDecay(Layer):
		in_layers = self.in_layers
		in_layers = convert_to_layers(in_layers)
		parent_tensor = in_layers[0].out_tensor
		self.out_tensor = parent_tensor+model_ops.weight_decay(self.penalty_type, self.penalty)
		self.out_tensor = parent_tensor + model_ops.weight_decay(self.penalty_type,
		self.penalty)
		return self.out_tensor

deepchem/models/tensorgraph/tensor_graph.py

+4 −2

Original line number	Diff line number	Diff line
		@@ -60,7 +60,8 @@ class TensorGraph(Model):
		self.loss = None
		self.built = False
		self.queue_installed = False
		self.optimizer = TFWrapper(tf.train.AdamOptimizer, learning_rate=0.001, beta1=0.9, beta2=0.999)
		self.optimizer = TFWrapper(
		tf.train.AdamOptimizer, learning_rate=0.001, beta1=0.9, beta2=0.999)

		# Singular place to hold Tensor objects which don't serialize
		# These have to be reconstructed on restoring from pickle
		@@ -448,7 +449,8 @@ class TensorGraph(Model):
		elif obj == "FileWriter":
		self.tensor_objects['FileWriter'] = tf.summary.FileWriter(self.model_dir)
		elif obj == 'train_op':
		self.tensor_objects['train_op'] = self.optimizer().minimize(self.loss.out_tensor)
		self.tensor_objects['train_op'] = self.optimizer().minimize(
		self.loss.out_tensor)
		elif obj == 'summary_op':
		self.tensor_objects['summary_op'] = tf.summary.merge_all(
		key=tf.GraphKeys.SUMMARIES)

Admin message