Loading deepchem/models/__init__.py +5 −3 Original line number Diff line number Diff line Loading @@ -17,15 +17,17 @@ from deepchem.models.tensorgraph.IRV import TensorflowMultiTaskIRVClassifier from deepchem.models.tensorgraph.robust_multitask import RobustMultitaskClassifier from deepchem.models.tensorgraph.robust_multitask import RobustMultitaskRegressor from deepchem.models.tensorgraph.progressive_multitask import ProgressiveMultitaskRegressor, ProgressiveMultitaskClassifier from deepchem.models.tensorgraph.models.graph_models import WeaveModel, DTNNTensorGraph, DAGTensorGraph, GraphConvModel, MPNNTensorGraph from deepchem.models.tensorgraph.models.graph_models import WeaveModel, DTNNModel, DAGModel, GraphConvModel, MPNNModel from deepchem.models.tensorgraph.models.symmetry_function_regression import BPSymmetryFunctionRegression, ANIRegression from deepchem.models.tensorgraph.models.scscore import ScScoreModel from deepchem.models.tensorgraph.models.seqtoseq import SeqToSeq from deepchem.models.tensorgraph.models.gan import GAN, WGAN from deepchem.models.tensorgraph.models.text_cnn import TextCNNTensorGraph from deepchem.models.tensorgraph.models.text_cnn import TextCNNModel from deepchem.models.tensorgraph.sequential import Sequential from deepchem.models.tensorgraph.models.sequence_dnn import SequenceDNN #################### Compatibility imports for renamed TensorGraph models. Remove below with DeepChem 3.0. #################### from deepchem.models.tensorgraph.models.graph_models import WeaveTensorGraph, GraphConvTensorGraph No newline at end of file from deepchem.models.tensorgraph.models.text_cnn import TextCNNTensorGraph from deepchem.models.tensorgraph.models.graph_models import WeaveTensorGraph, DTNNTensorGraph, DAGTensorGraph, GraphConvTensorGraph, MPNNTensorGraph deepchem/models/tensorgraph/initializations.py +5 −6 Original line number Diff line number Diff line Loading @@ -4,8 +4,7 @@ from __future__ import unicode_literals import numpy as np import tensorflow as tf from deepchem.models.tensorgraph.model_ops import random_uniform_variable from deepchem.models.tensorgraph.model_ops import random_normal_variable from deepchem.models.tensorgraph.model_ops import random_uniform_variable, random_normal_variable, create_variable from deepchem.models.tensorgraph.activations import get_from_module Loading Loading @@ -94,7 +93,7 @@ def orthogonal(shape, scale=1.1, name=None): # Pick the one with the correct shape. q = u if u.shape == flat_shape else v q = q.reshape(shape) return tf.Variable( return create_variable( scale * q[:shape[0], :shape[1]], dtype=tf.float32, name=name) Loading @@ -103,16 +102,16 @@ def identity(shape, scale=1, name=None): raise ValueError('Identity matrix initialization can only be used ' 'for 2D square matrices.') else: return tf.Variable( return create_variable( scale * np.identity(shape[0]), dtype=tf.float32, name=name) def zero(shape, name=None): return tf.Variable(tf.zeros(shape), dtype=tf.float32, name=name) return create_variable(tf.zeros(shape), dtype=tf.float32, name=name) def one(shape, name=None): return tf.Variable(tf.ones(shape), dtype=tf.float32, name=name) return create_variable(tf.ones(shape), dtype=tf.float32, name=name) def get(identifier, **kwargs): Loading deepchem/models/tensorgraph/layers.py +504 −191 File changed.Preview size limit exceeded, changes collapsed. Show changes deepchem/models/tensorgraph/model_ops.py +31 −22 Original line number Diff line number Diff line Loading @@ -11,6 +11,7 @@ import sys import traceback import numpy as np import tensorflow as tf import tensorflow.contrib.eager as tfe from tensorflow.python.training import moving_averages from collections import defaultdict Loading @@ -22,28 +23,36 @@ def _to_tensor(x, dtype): return x def create_variable(value, dtype=None, name=None): """Create a tf.Variable or tfe.Variable, depending on the current mode.""" if tfe.in_eager_mode(): return tfe.Variable(value, dtype=dtype, name=name) else: return tf.Variable(value, dtype=dtype, name=name) def ones(shape, dtype=None, name=None): """Instantiates an all-ones tensor variable and returns it. Parameters ---------- shape: Tuple of integers, shape of returned Keras variable. shape: Tuple of integers, shape of returned Tensorflow variable. dtype: Tensorflow dtype name: String, name of returned Keras variable. name: String, name of returned Tensorflow variable. Returns ------- A Keras variable, filled with `1.0`. A Tensorflow variable, filled with `1.0`. """ if dtype is None: dtype = tf.float32 shape = tuple(map(int, shape)) return tf.Variable( return create_variable( tf.constant_initializer(1., dtype=dtype)(shape), dtype, name) def cast_to_floatx(x): """Cast a Numpy array to the default Keras float type. """Cast a Numpy array to the default Tensorflow float type. Parameters ---------- Loading @@ -65,7 +74,7 @@ def moving_average_update(variable, value, momentum): def int_shape(x): """Returns the shape of a Keras tensor or a Keras variable as a tuple of """Returns the shape of a Tensorflow tensor or a Tensorflow variable as a tuple of integers or None entries. Arguments Loading Loading @@ -193,7 +202,7 @@ def get_ndim(x): def get_dtype(x): """Returns the dtype of a Keras tensor or variable, as a string. """Returns the dtype of a Tensorflow tensor or variable, as a string. Parameters ---------- Loading Loading @@ -259,7 +268,7 @@ def random_uniform_variable(shape, seed = np.random.randint(10e8) value = tf.random_uniform_initializer( low, high, dtype=dtype, seed=seed)(shape) return tf.Variable(value, dtype=dtype, name=name) return create_variable(value, dtype=dtype, name=name) def random_normal_variable(shape, Loading @@ -268,16 +277,16 @@ def random_normal_variable(shape, dtype=tf.float32, name=None, seed=None): """Instantiates an Keras variable filled with """Instantiates an Tensorflow variable filled with samples drawn from a normal distribution and returns it. Parameters ---------- shape: Tuple of integers, shape of returned Keras variable. shape: Tuple of integers, shape of returned Tensorflow variable. mean: Float, mean of the normal distribution. scale: Float, standard deviation of the normal distribution. dtype: Tensorflow dtype name: String, name of returned Keras variable. name: String, name of returned Tensorflow variable. seed: Integer, random seed. Returns Loading @@ -290,7 +299,7 @@ def random_normal_variable(shape, seed = np.random.randint(10e8) value = tf.random_normal_initializer( mean, scale, dtype=dtype, seed=seed)(shape) return tf.Variable(value, dtype=dtype, name=name) return create_variable(value, dtype=dtype, name=name) def max(x, axis=None, keepdims=False): Loading Loading @@ -338,7 +347,7 @@ def categorical_crossentropy(output, target, from_logits=False): # TODO(rbharath): Should probably swap this over to tf mode. """ # Note: tf.nn.softmax_cross_entropy_with_logits # expects logits, Keras expects probabilities. # expects logits, Tensorflow expects probabilities. if not from_logits: # scale preds so that the class probas of each sample sum to 1 output /= tf.reduce_sum( Loading @@ -362,7 +371,7 @@ def sparse_categorical_crossentropy(output, target, from_logits=False): and a target tensor, where the target is an integer tensor. """ # Note: tf.nn.softmax_cross_entropy_with_logits # expects logits, Keras expects probabilities. # expects logits, Tensorflow expects probabilities. if not from_logits: epsilon = _to_tensor(_EPSILON, output.dtype.base_dtype) output = tf.clip_by_value(output, epsilon, 1 - epsilon) Loading Loading @@ -398,7 +407,7 @@ def binary_crossentropy(output, target, from_logits=False): A tensor. """ # Note: tf.nn.softmax_cross_entropy_with_logits # expects logits, Keras expects probabilities. # expects logits, Tensorflow expects probabilities. if not from_logits: # transform back to logits epsilon = _to_tensor(_EPSILON, output.dtype.base_dtype) Loading Loading @@ -437,16 +446,16 @@ def zeros(shape, dtype=tf.float32, name=None): Parameters ---------- shape: Tuple of integers, shape of returned Keras variable shape: Tuple of integers, shape of returned Tensorflow variable dtype: Tensorflow dtype name: String, name of returned Keras variable name: String, name of returned Tensorflow variable Returns ------- A variable (including Keras metadata), filled with `0.0`. A variable (including Tensorflow metadata), filled with `0.0`. """ shape = tuple(map(int, shape)) return tf.Variable( return create_variable( tf.constant_initializer(0., dtype=dtype)(shape), dtype, name) Loading Loading @@ -674,7 +683,7 @@ def add_bias(tensor, init=None, name=None): if init is None: init = tf.zeros([tensor.get_shape()[-1].value]) with tf.name_scope(name, tensor.op.name, [tensor]): b = tf.Variable(init, name='b') b = create_variable(init, name='b') return tf.nn.bias_add(tensor, b) Loading Loading @@ -752,8 +761,8 @@ def fully_connected_layer(tensor, bias_init = tf.zeros([size]) with tf.name_scope(name, 'fully_connected', [tensor]): w = tf.Variable(weight_init, name='w', dtype=tf.float32) b = tf.Variable(bias_init, name='b', dtype=tf.float32) w = create_variable(weight_init, name='w', dtype=tf.float32) b = create_variable(bias_init, name='b', dtype=tf.float32) return tf.nn.xw_plus_b(tensor, w, b) Loading deepchem/models/tensorgraph/models/graph_models.py +57 −13 Original line number Diff line number Diff line Loading @@ -14,7 +14,7 @@ from deepchem.models.tensorgraph.graph_layers import WeaveGather, \ from deepchem.models.tensorgraph.graph_layers import WeaveLayerFactory from deepchem.models.tensorgraph.layers import Dense, SoftMax, \ SoftMaxCrossEntropy, GraphConv, BatchNorm, \ GraphPool, GraphGather, WeightedError, Dropout, BatchNormalization, Stack, Flatten, GraphCNN, GraphCNNPool GraphPool, GraphGather, WeightedError, Dropout, BatchNorm, Stack, Flatten, GraphCNN, GraphCNNPool from deepchem.models.tensorgraph.layers import L2Loss, Label, Weights, Feature from deepchem.models.tensorgraph.tensor_graph import TensorGraph from deepchem.trans import undo_transforms Loading Loading @@ -86,7 +86,7 @@ class WeaveModel(TensorGraph): out_channels=self.n_graph_feat, activation_fn=tf.nn.tanh, in_layers=weave_layer2A) batch_norm1 = BatchNormalization(epsilon=1e-5, mode=1, in_layers=[dense1]) batch_norm1 = BatchNorm(epsilon=1e-5, in_layers=[dense1]) weave_gather = WeaveGather( self.batch_size, n_input=self.n_graph_feat, Loading Loading @@ -194,7 +194,7 @@ class WeaveModel(TensorGraph): return out class DTNNTensorGraph(TensorGraph): class DTNNModel(TensorGraph): def __init__(self, n_tasks, Loading Loading @@ -237,7 +237,7 @@ class DTNNTensorGraph(TensorGraph): self.steps = np.expand_dims(self.steps, 0) self.output_activation = output_activation self.mode = mode super(DTNNTensorGraph, self).__init__(**kwargs) super(DTNNModel, self).__init__(**kwargs) assert self.mode == "regression" self.build_graph() Loading Loading @@ -351,14 +351,14 @@ class DTNNTensorGraph(TensorGraph): if transformers != [] and not isinstance(outputs, collections.Sequence): raise ValueError( "DTNN does not support single tensor output with transformers") retval = super(DTNNTensorGraph, self).predict(dataset, outputs=outputs) retval = super(DTNNModel, self).predict(dataset, outputs=outputs) if not isinstance(outputs, collections.Sequence): return retval retval = np.concatenate(retval, axis=-1) return undo_transforms(retval, transformers) class DAGTensorGraph(TensorGraph): class DAGModel(TensorGraph): def __init__(self, n_tasks, Loading Loading @@ -390,7 +390,7 @@ class DAGTensorGraph(TensorGraph): self.n_graph_feat = n_graph_feat self.n_outputs = n_outputs self.mode = mode super(DAGTensorGraph, self).__init__(**kwargs) super(DAGModel, self).__init__(**kwargs) self.build_graph() def build_graph(self): Loading Loading @@ -508,7 +508,7 @@ class DAGTensorGraph(TensorGraph): yield feed_dict def predict_on_generator(self, generator, transformers=[], outputs=None): out = super(DAGTensorGraph, self).predict_on_generator( out = super(DAGModel, self).predict_on_generator( generator, transformers=[], outputs=outputs) if outputs is None: outputs = self.outputs Loading @@ -519,7 +519,7 @@ class DAGTensorGraph(TensorGraph): return out class PetroskiSuchTensorGraph(TensorGraph): class PetroskiSuchModel(TensorGraph): """ Model from Robust Spatial Filtering with Graph Convolutional Neural Networks https://arxiv.org/abs/1703.00792 Loading @@ -545,7 +545,7 @@ class PetroskiSuchTensorGraph(TensorGraph): self.error_bars = True if 'error_bars' in kwargs and kwargs['error_bars'] else False self.dropout = dropout kwargs['use_queue'] = False super(PetroskiSuchTensorGraph, self).__init__(**kwargs) super(PetroskiSuchModel, self).__init__(**kwargs) self.build_graph() def build_graph(self): Loading Loading @@ -952,7 +952,7 @@ class GraphConvModel(TensorGraph): return y_ class MPNNTensorGraph(TensorGraph): class MPNNModel(TensorGraph): """ Message Passing Neural Network, default structures built according to https://arxiv.org/abs/1511.06391 """ Loading Loading @@ -987,7 +987,7 @@ class MPNNTensorGraph(TensorGraph): self.T = T self.M = M self.mode = mode super(MPNNTensorGraph, self).__init__(**kwargs) super(MPNNModel, self).__init__(**kwargs) self.build_graph() def build_graph(self): Loading Loading @@ -1172,3 +1172,47 @@ class WeaveTensorGraph(WeaveModel): def __init__(self, *args, **kwargs): super(WeaveModel, self).__init__(*args, **kwargs) class DTNNTensorGraph(DTNNModel): warnings.warn( TENSORGRAPH_DEPRECATION.format("DTNNTensorGraph", "DTNNModel"), FutureWarning) def __init__(self, *args, **kwargs): super(DTNNModel, self).__init__(*args, **kwargs) class DAGTensorGraph(DAGModel): warnings.warn( TENSORGRAPH_DEPRECATION.format("DAGTensorGraph", "DAGModel"), FutureWarning) def __init__(self, *args, **kwargs): super(DAGModel, self).__init__(*args, **kwargs) class PetroskiSuchTensorGraph(PetroskiSuchModel): warnings.warn( TENSORGRAPH_DEPRECATION.format("PetroskiSuchTensorGraph", "PetroskiSuchModel"), FutureWarning) def __init__(self, *args, **kwargs): super(PetroskiSuchModel, self).__init__(*args, **kwargs) class MPNNTensorGraph(MPNNModel): warnings.warn( TENSORGRAPH_DEPRECATION.format("MPNNTensorGraph", "MPNNModel"), FutureWarning) def __init__(self, *args, **kwargs): super(MPNNModel, self).__init__(*args, **kwargs) Loading
deepchem/models/__init__.py +5 −3 Original line number Diff line number Diff line Loading @@ -17,15 +17,17 @@ from deepchem.models.tensorgraph.IRV import TensorflowMultiTaskIRVClassifier from deepchem.models.tensorgraph.robust_multitask import RobustMultitaskClassifier from deepchem.models.tensorgraph.robust_multitask import RobustMultitaskRegressor from deepchem.models.tensorgraph.progressive_multitask import ProgressiveMultitaskRegressor, ProgressiveMultitaskClassifier from deepchem.models.tensorgraph.models.graph_models import WeaveModel, DTNNTensorGraph, DAGTensorGraph, GraphConvModel, MPNNTensorGraph from deepchem.models.tensorgraph.models.graph_models import WeaveModel, DTNNModel, DAGModel, GraphConvModel, MPNNModel from deepchem.models.tensorgraph.models.symmetry_function_regression import BPSymmetryFunctionRegression, ANIRegression from deepchem.models.tensorgraph.models.scscore import ScScoreModel from deepchem.models.tensorgraph.models.seqtoseq import SeqToSeq from deepchem.models.tensorgraph.models.gan import GAN, WGAN from deepchem.models.tensorgraph.models.text_cnn import TextCNNTensorGraph from deepchem.models.tensorgraph.models.text_cnn import TextCNNModel from deepchem.models.tensorgraph.sequential import Sequential from deepchem.models.tensorgraph.models.sequence_dnn import SequenceDNN #################### Compatibility imports for renamed TensorGraph models. Remove below with DeepChem 3.0. #################### from deepchem.models.tensorgraph.models.graph_models import WeaveTensorGraph, GraphConvTensorGraph No newline at end of file from deepchem.models.tensorgraph.models.text_cnn import TextCNNTensorGraph from deepchem.models.tensorgraph.models.graph_models import WeaveTensorGraph, DTNNTensorGraph, DAGTensorGraph, GraphConvTensorGraph, MPNNTensorGraph
deepchem/models/tensorgraph/initializations.py +5 −6 Original line number Diff line number Diff line Loading @@ -4,8 +4,7 @@ from __future__ import unicode_literals import numpy as np import tensorflow as tf from deepchem.models.tensorgraph.model_ops import random_uniform_variable from deepchem.models.tensorgraph.model_ops import random_normal_variable from deepchem.models.tensorgraph.model_ops import random_uniform_variable, random_normal_variable, create_variable from deepchem.models.tensorgraph.activations import get_from_module Loading Loading @@ -94,7 +93,7 @@ def orthogonal(shape, scale=1.1, name=None): # Pick the one with the correct shape. q = u if u.shape == flat_shape else v q = q.reshape(shape) return tf.Variable( return create_variable( scale * q[:shape[0], :shape[1]], dtype=tf.float32, name=name) Loading @@ -103,16 +102,16 @@ def identity(shape, scale=1, name=None): raise ValueError('Identity matrix initialization can only be used ' 'for 2D square matrices.') else: return tf.Variable( return create_variable( scale * np.identity(shape[0]), dtype=tf.float32, name=name) def zero(shape, name=None): return tf.Variable(tf.zeros(shape), dtype=tf.float32, name=name) return create_variable(tf.zeros(shape), dtype=tf.float32, name=name) def one(shape, name=None): return tf.Variable(tf.ones(shape), dtype=tf.float32, name=name) return create_variable(tf.ones(shape), dtype=tf.float32, name=name) def get(identifier, **kwargs): Loading
deepchem/models/tensorgraph/layers.py +504 −191 File changed.Preview size limit exceeded, changes collapsed. Show changes
deepchem/models/tensorgraph/model_ops.py +31 −22 Original line number Diff line number Diff line Loading @@ -11,6 +11,7 @@ import sys import traceback import numpy as np import tensorflow as tf import tensorflow.contrib.eager as tfe from tensorflow.python.training import moving_averages from collections import defaultdict Loading @@ -22,28 +23,36 @@ def _to_tensor(x, dtype): return x def create_variable(value, dtype=None, name=None): """Create a tf.Variable or tfe.Variable, depending on the current mode.""" if tfe.in_eager_mode(): return tfe.Variable(value, dtype=dtype, name=name) else: return tf.Variable(value, dtype=dtype, name=name) def ones(shape, dtype=None, name=None): """Instantiates an all-ones tensor variable and returns it. Parameters ---------- shape: Tuple of integers, shape of returned Keras variable. shape: Tuple of integers, shape of returned Tensorflow variable. dtype: Tensorflow dtype name: String, name of returned Keras variable. name: String, name of returned Tensorflow variable. Returns ------- A Keras variable, filled with `1.0`. A Tensorflow variable, filled with `1.0`. """ if dtype is None: dtype = tf.float32 shape = tuple(map(int, shape)) return tf.Variable( return create_variable( tf.constant_initializer(1., dtype=dtype)(shape), dtype, name) def cast_to_floatx(x): """Cast a Numpy array to the default Keras float type. """Cast a Numpy array to the default Tensorflow float type. Parameters ---------- Loading @@ -65,7 +74,7 @@ def moving_average_update(variable, value, momentum): def int_shape(x): """Returns the shape of a Keras tensor or a Keras variable as a tuple of """Returns the shape of a Tensorflow tensor or a Tensorflow variable as a tuple of integers or None entries. Arguments Loading Loading @@ -193,7 +202,7 @@ def get_ndim(x): def get_dtype(x): """Returns the dtype of a Keras tensor or variable, as a string. """Returns the dtype of a Tensorflow tensor or variable, as a string. Parameters ---------- Loading Loading @@ -259,7 +268,7 @@ def random_uniform_variable(shape, seed = np.random.randint(10e8) value = tf.random_uniform_initializer( low, high, dtype=dtype, seed=seed)(shape) return tf.Variable(value, dtype=dtype, name=name) return create_variable(value, dtype=dtype, name=name) def random_normal_variable(shape, Loading @@ -268,16 +277,16 @@ def random_normal_variable(shape, dtype=tf.float32, name=None, seed=None): """Instantiates an Keras variable filled with """Instantiates an Tensorflow variable filled with samples drawn from a normal distribution and returns it. Parameters ---------- shape: Tuple of integers, shape of returned Keras variable. shape: Tuple of integers, shape of returned Tensorflow variable. mean: Float, mean of the normal distribution. scale: Float, standard deviation of the normal distribution. dtype: Tensorflow dtype name: String, name of returned Keras variable. name: String, name of returned Tensorflow variable. seed: Integer, random seed. Returns Loading @@ -290,7 +299,7 @@ def random_normal_variable(shape, seed = np.random.randint(10e8) value = tf.random_normal_initializer( mean, scale, dtype=dtype, seed=seed)(shape) return tf.Variable(value, dtype=dtype, name=name) return create_variable(value, dtype=dtype, name=name) def max(x, axis=None, keepdims=False): Loading Loading @@ -338,7 +347,7 @@ def categorical_crossentropy(output, target, from_logits=False): # TODO(rbharath): Should probably swap this over to tf mode. """ # Note: tf.nn.softmax_cross_entropy_with_logits # expects logits, Keras expects probabilities. # expects logits, Tensorflow expects probabilities. if not from_logits: # scale preds so that the class probas of each sample sum to 1 output /= tf.reduce_sum( Loading @@ -362,7 +371,7 @@ def sparse_categorical_crossentropy(output, target, from_logits=False): and a target tensor, where the target is an integer tensor. """ # Note: tf.nn.softmax_cross_entropy_with_logits # expects logits, Keras expects probabilities. # expects logits, Tensorflow expects probabilities. if not from_logits: epsilon = _to_tensor(_EPSILON, output.dtype.base_dtype) output = tf.clip_by_value(output, epsilon, 1 - epsilon) Loading Loading @@ -398,7 +407,7 @@ def binary_crossentropy(output, target, from_logits=False): A tensor. """ # Note: tf.nn.softmax_cross_entropy_with_logits # expects logits, Keras expects probabilities. # expects logits, Tensorflow expects probabilities. if not from_logits: # transform back to logits epsilon = _to_tensor(_EPSILON, output.dtype.base_dtype) Loading Loading @@ -437,16 +446,16 @@ def zeros(shape, dtype=tf.float32, name=None): Parameters ---------- shape: Tuple of integers, shape of returned Keras variable shape: Tuple of integers, shape of returned Tensorflow variable dtype: Tensorflow dtype name: String, name of returned Keras variable name: String, name of returned Tensorflow variable Returns ------- A variable (including Keras metadata), filled with `0.0`. A variable (including Tensorflow metadata), filled with `0.0`. """ shape = tuple(map(int, shape)) return tf.Variable( return create_variable( tf.constant_initializer(0., dtype=dtype)(shape), dtype, name) Loading Loading @@ -674,7 +683,7 @@ def add_bias(tensor, init=None, name=None): if init is None: init = tf.zeros([tensor.get_shape()[-1].value]) with tf.name_scope(name, tensor.op.name, [tensor]): b = tf.Variable(init, name='b') b = create_variable(init, name='b') return tf.nn.bias_add(tensor, b) Loading Loading @@ -752,8 +761,8 @@ def fully_connected_layer(tensor, bias_init = tf.zeros([size]) with tf.name_scope(name, 'fully_connected', [tensor]): w = tf.Variable(weight_init, name='w', dtype=tf.float32) b = tf.Variable(bias_init, name='b', dtype=tf.float32) w = create_variable(weight_init, name='w', dtype=tf.float32) b = create_variable(bias_init, name='b', dtype=tf.float32) return tf.nn.xw_plus_b(tensor, w, b) Loading
deepchem/models/tensorgraph/models/graph_models.py +57 −13 Original line number Diff line number Diff line Loading @@ -14,7 +14,7 @@ from deepchem.models.tensorgraph.graph_layers import WeaveGather, \ from deepchem.models.tensorgraph.graph_layers import WeaveLayerFactory from deepchem.models.tensorgraph.layers import Dense, SoftMax, \ SoftMaxCrossEntropy, GraphConv, BatchNorm, \ GraphPool, GraphGather, WeightedError, Dropout, BatchNormalization, Stack, Flatten, GraphCNN, GraphCNNPool GraphPool, GraphGather, WeightedError, Dropout, BatchNorm, Stack, Flatten, GraphCNN, GraphCNNPool from deepchem.models.tensorgraph.layers import L2Loss, Label, Weights, Feature from deepchem.models.tensorgraph.tensor_graph import TensorGraph from deepchem.trans import undo_transforms Loading Loading @@ -86,7 +86,7 @@ class WeaveModel(TensorGraph): out_channels=self.n_graph_feat, activation_fn=tf.nn.tanh, in_layers=weave_layer2A) batch_norm1 = BatchNormalization(epsilon=1e-5, mode=1, in_layers=[dense1]) batch_norm1 = BatchNorm(epsilon=1e-5, in_layers=[dense1]) weave_gather = WeaveGather( self.batch_size, n_input=self.n_graph_feat, Loading Loading @@ -194,7 +194,7 @@ class WeaveModel(TensorGraph): return out class DTNNTensorGraph(TensorGraph): class DTNNModel(TensorGraph): def __init__(self, n_tasks, Loading Loading @@ -237,7 +237,7 @@ class DTNNTensorGraph(TensorGraph): self.steps = np.expand_dims(self.steps, 0) self.output_activation = output_activation self.mode = mode super(DTNNTensorGraph, self).__init__(**kwargs) super(DTNNModel, self).__init__(**kwargs) assert self.mode == "regression" self.build_graph() Loading Loading @@ -351,14 +351,14 @@ class DTNNTensorGraph(TensorGraph): if transformers != [] and not isinstance(outputs, collections.Sequence): raise ValueError( "DTNN does not support single tensor output with transformers") retval = super(DTNNTensorGraph, self).predict(dataset, outputs=outputs) retval = super(DTNNModel, self).predict(dataset, outputs=outputs) if not isinstance(outputs, collections.Sequence): return retval retval = np.concatenate(retval, axis=-1) return undo_transforms(retval, transformers) class DAGTensorGraph(TensorGraph): class DAGModel(TensorGraph): def __init__(self, n_tasks, Loading Loading @@ -390,7 +390,7 @@ class DAGTensorGraph(TensorGraph): self.n_graph_feat = n_graph_feat self.n_outputs = n_outputs self.mode = mode super(DAGTensorGraph, self).__init__(**kwargs) super(DAGModel, self).__init__(**kwargs) self.build_graph() def build_graph(self): Loading Loading @@ -508,7 +508,7 @@ class DAGTensorGraph(TensorGraph): yield feed_dict def predict_on_generator(self, generator, transformers=[], outputs=None): out = super(DAGTensorGraph, self).predict_on_generator( out = super(DAGModel, self).predict_on_generator( generator, transformers=[], outputs=outputs) if outputs is None: outputs = self.outputs Loading @@ -519,7 +519,7 @@ class DAGTensorGraph(TensorGraph): return out class PetroskiSuchTensorGraph(TensorGraph): class PetroskiSuchModel(TensorGraph): """ Model from Robust Spatial Filtering with Graph Convolutional Neural Networks https://arxiv.org/abs/1703.00792 Loading @@ -545,7 +545,7 @@ class PetroskiSuchTensorGraph(TensorGraph): self.error_bars = True if 'error_bars' in kwargs and kwargs['error_bars'] else False self.dropout = dropout kwargs['use_queue'] = False super(PetroskiSuchTensorGraph, self).__init__(**kwargs) super(PetroskiSuchModel, self).__init__(**kwargs) self.build_graph() def build_graph(self): Loading Loading @@ -952,7 +952,7 @@ class GraphConvModel(TensorGraph): return y_ class MPNNTensorGraph(TensorGraph): class MPNNModel(TensorGraph): """ Message Passing Neural Network, default structures built according to https://arxiv.org/abs/1511.06391 """ Loading Loading @@ -987,7 +987,7 @@ class MPNNTensorGraph(TensorGraph): self.T = T self.M = M self.mode = mode super(MPNNTensorGraph, self).__init__(**kwargs) super(MPNNModel, self).__init__(**kwargs) self.build_graph() def build_graph(self): Loading Loading @@ -1172,3 +1172,47 @@ class WeaveTensorGraph(WeaveModel): def __init__(self, *args, **kwargs): super(WeaveModel, self).__init__(*args, **kwargs) class DTNNTensorGraph(DTNNModel): warnings.warn( TENSORGRAPH_DEPRECATION.format("DTNNTensorGraph", "DTNNModel"), FutureWarning) def __init__(self, *args, **kwargs): super(DTNNModel, self).__init__(*args, **kwargs) class DAGTensorGraph(DAGModel): warnings.warn( TENSORGRAPH_DEPRECATION.format("DAGTensorGraph", "DAGModel"), FutureWarning) def __init__(self, *args, **kwargs): super(DAGModel, self).__init__(*args, **kwargs) class PetroskiSuchTensorGraph(PetroskiSuchModel): warnings.warn( TENSORGRAPH_DEPRECATION.format("PetroskiSuchTensorGraph", "PetroskiSuchModel"), FutureWarning) def __init__(self, *args, **kwargs): super(PetroskiSuchModel, self).__init__(*args, **kwargs) class MPNNTensorGraph(MPNNModel): warnings.warn( TENSORGRAPH_DEPRECATION.format("MPNNTensorGraph", "MPNNModel"), FutureWarning) def __init__(self, *args, **kwargs): super(MPNNModel, self).__init__(*args, **kwargs)
