Merge pull request #1229 from miaecle/DAG (01430dd7) · Commits · 钟慕尧 / deepchem

deepchem/models/tensorgraph/graph_layers.py

+25 −13

Original line number	Diff line number	Diff line
		@@ -581,16 +581,18 @@ class DAGLayer(Layer):
		Number of features listed per atom.
		max_atoms: int, optional
		Maximum number of atoms in molecules.
		layer_sizes: list of int, optional(default=[1000])
		Structure of hidden layer(s)
		layer_sizes: list of int, optional(default=[100])
		List of hidden layer size(s):
		length of this list represents the number of hidden layers,
		and each element is the width of corresponding hidden layer.
		init: str, optional
		Weight initialization for filters.
		activation: str, optional
		Activation function applied
		Activation function applied.
		dropout: float, optional
		Dropout probability, not supported here
		Dropout probability in hidden layer(s).
		batch_size: int, optional
		number of molecules in a batch
		number of molecules in a batch.
		"""
		super(DAGLayer, self).__init__(**kwargs)

		@@ -683,7 +685,8 @@ class DAGLayer(Layer):
		# DAGgraph_step maps from batch_inputs to a batch of graph_features
		# of shape: (batch_sizemax_atoms) n_graph_features
		# representing the graph features of target atoms in each graph
		batch_outputs = self.DAGgraph_step(batch_inputs, self.W_list, self.b_list)
		batch_outputs = self.DAGgraph_step(batch_inputs, self.W_list, self.b_list,
		**kwargs)

		# index for targe atoms
		target_index = tf.stack([tf.range(n_atoms), parents[:, count, 0]], axis=1)
		@@ -698,11 +701,14 @@ class DAGLayer(Layer):
		self.out_tensor = out_tensor
		return out_tensor

		def DAGgraph_step(self, batch_inputs, W_list, b_list):
		def DAGgraph_step(self, batch_inputs, W_list, b_list, **kwargs):
		outputs = batch_inputs
		for idw, W in enumerate(W_list):
		outputs = tf.nn.xw_plus_b(outputs, W, b_list[idw])
		outputs = self.activation(outputs)
		training = kwargs['training'] if 'training' in kwargs else 1.0
		if not self.dropout is None:
		outputs = tf.nn.dropout(outputs, 1.0 - self.dropout * training)
		return outputs

		def none_tensors(self):
		@@ -733,19 +739,21 @@ class DAGGather(Layer):
		Parameters
		----------
		n_graph_feat: int, optional
		Number of features for each atom
		Number of features for each atom.
		n_outputs: int, optional
		Number of features for each molecule.
		max_atoms: int, optional
		Maximum number of atoms in molecules.
		layer_sizes: list of int, optional
		Structure of hidden layer(s)
		List of hidden layer size(s):
		length of this list represents the number of hidden layers,
		and each element is the width of corresponding hidden layer.
		init: str, optional
		Weight initialization for filters.
		activation: str, optional
		Activation function applied
		Activation function applied.
		dropout: float, optional
		Dropout probability, not supported
		Dropout probability in the hidden layer(s).
		"""
		super(DAGGather, self).__init__(**kwargs)

		@@ -794,18 +802,22 @@ class DAGGather(Layer):
		# Extract atom_features
		graph_features = tf.segment_sum(atom_features, membership)
		# sum all graph outputs
		outputs = self.DAGgraph_step(graph_features, self.W_list, self.b_list)
		outputs = self.DAGgraph_step(graph_features, self.W_list, self.b_list,
		**kwargs)
		out_tensor = outputs
		if set_tensors:
		self.variables = self.trainable_weights
		self.out_tensor = out_tensor
		return out_tensor

		def DAGgraph_step(self, batch_inputs, W_list, b_list):
		def DAGgraph_step(self, batch_inputs, W_list, b_list, **kwargs):
		outputs = batch_inputs
		for idw, W in enumerate(W_list):
		outputs = tf.nn.xw_plus_b(outputs, W, b_list[idw])
		outputs = self.activation(outputs)
		training = kwargs['training'] if 'training' in kwargs else 1.0
		if not self.dropout is None:
		outputs = tf.nn.dropout(outputs, 1.0 - self.dropout * training)
		return outputs

		def none_tensors(self):

deepchem/models/tensorgraph/models/graph_models.py

+23 −5

Original line number	Diff line number	Diff line
		@@ -366,22 +366,33 @@ class DAGModel(TensorGraph):
		n_atom_feat=75,
		n_graph_feat=30,
		n_outputs=30,
		layer_sizes=[100],
		layer_sizes_gather=[100],
		dropout=None,
		mode="classification",
		**kwargs):
		"""
		Parameters
		----------
		n_tasks: int
		Number of tasks
		Number of tasks.
		max_atoms: int, optional
		Maximum number of atoms in a molecule, should be defined based on dataset
		Maximum number of atoms in a molecule, should be defined based on dataset.
		n_atom_feat: int, optional
		Number of features per atom.
		n_graph_feat: int, optional
		Number of features for atom in the graph
		Number of features for atom in the graph.
		n_outputs: int, optional
		Number of features for each molecule
		mode: str
		Number of features for each molecule.
		layer_sizes: list of int, optional
		List of hidden layer size(s) in the propagation step:
		length of this list represents the number of hidden layers,
		and each element is the width of corresponding hidden layer.
		layer_sizes_gather: list of int, optional
		List of hidden layer size(s) in the gather step.
		dropout: None or float, optional
		Dropout probability, applied after each propagation step and gather step.
		mode: str, optional
		Either "classification" or "regression" for type of model.
		"""
		self.n_tasks = n_tasks
		@@ -389,6 +400,9 @@ class DAGModel(TensorGraph):
		self.n_atom_feat = n_atom_feat
		self.n_graph_feat = n_graph_feat
		self.n_outputs = n_outputs
		self.layer_sizes = layer_sizes
		self.layer_sizes_gather = layer_sizes_gather
		self.dropout = dropout
		self.mode = mode
		super(DAGModel, self).__init__(**kwargs)
		self.build_graph()
		@@ -410,6 +424,8 @@ class DAGModel(TensorGraph):
		n_graph_feat=self.n_graph_feat,
		n_atom_feat=self.n_atom_feat,
		max_atoms=self.max_atoms,
		layer_sizes=self.layer_sizes,
		dropout=self.dropout,
		batch_size=self.batch_size,
		in_layers=[
		self.atom_features, self.parents, self.calculation_orders,
		@@ -419,6 +435,8 @@ class DAGModel(TensorGraph):
		n_graph_feat=self.n_graph_feat,
		n_outputs=self.n_outputs,
		max_atoms=self.max_atoms,
		layer_sizes=self.layer_sizes_gather,
		dropout=self.dropout,
		in_layers=[dag_layer1, self.membership])

		costs = []

Admin message