Loading deepchem/models/tensorgraph/symmetry_functions.py +30 −21 Original line number Diff line number Diff line Loading @@ -403,8 +403,8 @@ class AtomicDifferentiatedDense(Layer): init='glorot_uniform', activation='relu', **kwargs): self.init = initializations.get(init) # Set weight initialization self.activation = activations.get(activation) # Get activations self.init = init # Set weight initialization self.activation = activation # Get activations self.max_atoms = max_atoms self.out_channels = out_channels self.atom_number_cases = atom_number_cases Loading @@ -413,6 +413,8 @@ class AtomicDifferentiatedDense(Layer): def create_tensor(self, in_layers=None, set_tensors=True, **kwargs): """ Generate Radial Symmetry Function """ init_fn = initializations.get(self.init) # Set weight initialization activation_fn = activations.get(self.activation) if in_layers is None: in_layers = self.in_layers in_layers = convert_to_layers(in_layers) Loading @@ -420,13 +422,12 @@ class AtomicDifferentiatedDense(Layer): inputs = in_layers[0].out_tensor atom_numbers = in_layers[1].out_tensor in_channels = inputs.get_shape().as_list()[-1] self.W = self.init( self.W = init_fn( [len(self.atom_number_cases), in_channels, self.out_channels]) self.b = model_ops.zeros((len(self.atom_number_cases), self.out_channels)) outputs = [] for i, atom_case in enumerate(self.atom_number_cases): # optimization to allow for tensorcontraction/broadcasted mmul # using a reshape trick. Note that the np and tf matmul behavior # differs when dealing with broadcasts Loading @@ -439,7 +440,7 @@ class AtomicDifferentiatedDense(Layer): ak = tf.shape(a)[2] bl = tf.shape(b)[1] output = self.activation( output = activation_fn( tf.reshape(tf.matmul(tf.reshape(a, [ai * aj, ak]), b), [ai, aj, bl]) + self.b[i, :]) Loading @@ -448,3 +449,11 @@ class AtomicDifferentiatedDense(Layer): self.out_channels)) outputs.append(output) self.out_tensor = tf.add_n(outputs) def none_tensors(self): w, b, out_tensor = self.W, self.b, self.out_tensor self.W, self.b, self.out_tensor = None, None, None return w, b, out_tensor def set_tensors(self, tensor): self.W, self.b, self.out_tensor = tensor deepchem/models/tensorgraph/tests/test_layers_pickle.py +25 −8 Original line number Diff line number Diff line import numpy as np import tensorflow as tf from deepchem.models import TensorGraph from deepchem.models.tensorgraph.graph_layers import Combine_AP, Separate_AP, \ WeaveLayer, WeaveGather, DTNNEmbedding, DTNNGather, DTNNStep, \ DTNNExtract, DAGLayer, DAGGather, MessagePassing, SetGather from deepchem.models.tensorgraph.layers import Feature, Conv1D, Dense, Flatten, Reshape, Squeeze, Transpose, \ CombineMeanStd, Repeat, Gather, GRU, L2Loss, Concat, SoftMax, Constant, Variable, Add, Multiply, Log, InteratomicL2Distances, \ CombineMeanStd, Repeat, Gather, GRU, L2Loss, Concat, SoftMax, \ Constant, Variable, Add, Multiply, Log, InteratomicL2Distances, \ SoftMaxCrossEntropy, ReduceMean, ToFloat, ReduceSquareDifference, Conv2D, MaxPool2D, ReduceSum, GraphConv, GraphPool, \ GraphGather, BatchNorm, WeightedError, \ Conv3D, MaxPool3D, \ LSTMStep, AttnLSTMEmbedding, IterRefLSTMEmbedding from deepchem.models.tensorgraph.graph_layers import Combine_AP, Separate_AP, \ WeaveLayer, WeaveGather, DTNNEmbedding, DTNNGather, DTNNStep, \ DTNNExtract, DAGLayer, DAGGather, MessagePassing, SetGather from deepchem.models.tensorgraph.symmetry_functions import AtomicDifferentiatedDense def test_Conv1D_pickle(): Loading Loading @@ -541,3 +544,17 @@ def test_SetGather_pickle(): tg.set_loss(Gather) tg.build() tg.save() def test_AtomicDifferentialDense_pickle(): max_atoms = 23 atom_features = 100 tg = TensorGraph() atom_feature = Feature(shape=(None, max_atoms, atom_features)) atom_numbers = Feature(shape=(None, max_atoms)) atomic_differential_dense = AtomicDifferentiatedDense( max_atoms=23, out_channels=5, in_layers=[atom_feature, atom_numbers]) tg.add_output(atomic_differential_dense) tg.set_loss(atomic_differential_dense) tg.build() tg.save() Loading
deepchem/models/tensorgraph/symmetry_functions.py +30 −21 Original line number Diff line number Diff line Loading @@ -403,8 +403,8 @@ class AtomicDifferentiatedDense(Layer): init='glorot_uniform', activation='relu', **kwargs): self.init = initializations.get(init) # Set weight initialization self.activation = activations.get(activation) # Get activations self.init = init # Set weight initialization self.activation = activation # Get activations self.max_atoms = max_atoms self.out_channels = out_channels self.atom_number_cases = atom_number_cases Loading @@ -413,6 +413,8 @@ class AtomicDifferentiatedDense(Layer): def create_tensor(self, in_layers=None, set_tensors=True, **kwargs): """ Generate Radial Symmetry Function """ init_fn = initializations.get(self.init) # Set weight initialization activation_fn = activations.get(self.activation) if in_layers is None: in_layers = self.in_layers in_layers = convert_to_layers(in_layers) Loading @@ -420,13 +422,12 @@ class AtomicDifferentiatedDense(Layer): inputs = in_layers[0].out_tensor atom_numbers = in_layers[1].out_tensor in_channels = inputs.get_shape().as_list()[-1] self.W = self.init( self.W = init_fn( [len(self.atom_number_cases), in_channels, self.out_channels]) self.b = model_ops.zeros((len(self.atom_number_cases), self.out_channels)) outputs = [] for i, atom_case in enumerate(self.atom_number_cases): # optimization to allow for tensorcontraction/broadcasted mmul # using a reshape trick. Note that the np and tf matmul behavior # differs when dealing with broadcasts Loading @@ -439,7 +440,7 @@ class AtomicDifferentiatedDense(Layer): ak = tf.shape(a)[2] bl = tf.shape(b)[1] output = self.activation( output = activation_fn( tf.reshape(tf.matmul(tf.reshape(a, [ai * aj, ak]), b), [ai, aj, bl]) + self.b[i, :]) Loading @@ -448,3 +449,11 @@ class AtomicDifferentiatedDense(Layer): self.out_channels)) outputs.append(output) self.out_tensor = tf.add_n(outputs) def none_tensors(self): w, b, out_tensor = self.W, self.b, self.out_tensor self.W, self.b, self.out_tensor = None, None, None return w, b, out_tensor def set_tensors(self, tensor): self.W, self.b, self.out_tensor = tensor
deepchem/models/tensorgraph/tests/test_layers_pickle.py +25 −8 Original line number Diff line number Diff line import numpy as np import tensorflow as tf from deepchem.models import TensorGraph from deepchem.models.tensorgraph.graph_layers import Combine_AP, Separate_AP, \ WeaveLayer, WeaveGather, DTNNEmbedding, DTNNGather, DTNNStep, \ DTNNExtract, DAGLayer, DAGGather, MessagePassing, SetGather from deepchem.models.tensorgraph.layers import Feature, Conv1D, Dense, Flatten, Reshape, Squeeze, Transpose, \ CombineMeanStd, Repeat, Gather, GRU, L2Loss, Concat, SoftMax, Constant, Variable, Add, Multiply, Log, InteratomicL2Distances, \ CombineMeanStd, Repeat, Gather, GRU, L2Loss, Concat, SoftMax, \ Constant, Variable, Add, Multiply, Log, InteratomicL2Distances, \ SoftMaxCrossEntropy, ReduceMean, ToFloat, ReduceSquareDifference, Conv2D, MaxPool2D, ReduceSum, GraphConv, GraphPool, \ GraphGather, BatchNorm, WeightedError, \ Conv3D, MaxPool3D, \ LSTMStep, AttnLSTMEmbedding, IterRefLSTMEmbedding from deepchem.models.tensorgraph.graph_layers import Combine_AP, Separate_AP, \ WeaveLayer, WeaveGather, DTNNEmbedding, DTNNGather, DTNNStep, \ DTNNExtract, DAGLayer, DAGGather, MessagePassing, SetGather from deepchem.models.tensorgraph.symmetry_functions import AtomicDifferentiatedDense def test_Conv1D_pickle(): Loading Loading @@ -541,3 +544,17 @@ def test_SetGather_pickle(): tg.set_loss(Gather) tg.build() tg.save() def test_AtomicDifferentialDense_pickle(): max_atoms = 23 atom_features = 100 tg = TensorGraph() atom_feature = Feature(shape=(None, max_atoms, atom_features)) atom_numbers = Feature(shape=(None, max_atoms)) atomic_differential_dense = AtomicDifferentiatedDense( max_atoms=23, out_channels=5, in_layers=[atom_feature, atom_numbers]) tg.add_output(atomic_differential_dense) tg.set_loss(atomic_differential_dense) tg.build() tg.save()
