Loading deepchem/models/tensorgraph/fcnet.py +17 −23 Original line number Diff line number Diff line Loading @@ -18,7 +18,7 @@ from deepchem.metrics import to_one_hot, from_one_hot 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, ReduceSum from deepchem.models.tensorgraph.layers import Feature, Label, Weights, WeightedError, Dense, Dropout, WeightDecay, Reshape, SoftMax, SoftMaxCrossEntropy, L2Loss, ReduceSum logger = logging.getLogger(__name__) Loading Loading @@ -114,15 +114,16 @@ class MultiTaskClassifier(TensorGraph): # Compute the loss function for each label. output = Reshape( logits = Reshape( shape=(-1, n_tasks, n_classes), in_layers=[ Dense(in_layers=[prev_layer], out_channels=n_tasks * n_classes) ]) output = SoftMax(logits) self.add_output(output) labels = Label(shape=(None, n_tasks, n_classes)) weights = Weights(shape=(None, n_tasks)) loss = SoftMaxCrossEntropy(in_layers=[labels, output]) loss = SoftMaxCrossEntropy(in_layers=[labels, logits]) weighted_loss = WeightedError(in_layers=[loss, weights]) if weight_decay_penalty != 0.0: weighted_loss = WeightDecay( Loading Loading @@ -154,6 +155,19 @@ class MultiTaskClassifier(TensorGraph): feed_dict[self.task_weights[0]] = w_b yield feed_dict def create_estimator_inputs(self, feature_columns, weight_column, features, labels, mode): tensors = {} for layer, column in zip(self.features, feature_columns): tensors[layer] = tf.feature_column.input_layer(features, [column]) if weight_column is not None: tensors[self.task_weights[0]] = tf.feature_column.input_layer( features, [weight_column]) if labels is not None: tensors[self.labels[0]] = tf.one_hot( tf.cast(labels, tf.int32), self.n_classes) return tensors def predict_proba(self, dataset, transformers=[], outputs=None): return super(MultiTaskClassifier, self).predict(dataset, transformers, outputs) Loading Loading @@ -291,26 +305,6 @@ class MultiTaskRegressor(TensorGraph): in_layers=[weighted_loss]) self.set_loss(weighted_loss) def default_generator(self, dataset, epochs=1, predict=False, deterministic=True, pad_batches=True): for epoch in range(epochs): for (X_b, y_b, w_b, ids_b) in dataset.iterbatches( batch_size=self.batch_size, deterministic=deterministic, pad_batches=pad_batches): feed_dict = dict() if y_b is not None and not predict: feed_dict[self.labels[0]] = y_b.reshape(-1, self.n_tasks, 1) if X_b is not None: feed_dict[self.features[0]] = X_b if w_b is not None and not predict: feed_dict[self.task_weights[0]] = w_b yield feed_dict class MultiTaskFitTransformRegressor(MultiTaskRegressor): """Implements a MultiTaskRegressor that performs on-the-fly transformation during fit/predict. Loading deepchem/models/tensorgraph/layers.py +16 −10 Original line number Diff line number Diff line Loading @@ -152,9 +152,8 @@ class Layer(object): def set_summary(self, summary_op, summary_description=None, collections=None): """Annotates a tensor with a tf.summary operation Collects data from self.out_tensor by default but can be changed by setting self.tb_input to another tensor in create_tensor This causes self.out_tensor to be logged to Tensorboard. Parameters ---------- Loading @@ -175,21 +174,28 @@ class Layer(object): self.collections = collections self.tensorboard = True def add_summary_to_tg(self): def add_summary_to_tg(self, tb_input=None): """ Can only be called after self.create_layer to gaurentee that name is not none Create the summary operation for this layer, if set_summary() has been called on it. Can only be called after self.create_layer to guarantee that name is not None. Parameters ---------- tb_input: tensor the tensor to log to Tensorboard. If None, self.out_tensor is used. """ if self.tensorboard == False: return if self.tb_input == None: self.tb_input = self.out_tensor if tb_input == None: tb_input = self.out_tensor if self.summary_op == "tensor_summary": tf.summary.tensor_summary(self.name, self.tb_input, self.summary_description, self.collections) tf.summary.tensor_summary(self.name, tb_input, self.summary_description, self.collections) elif self.summary_op == 'scalar': tf.summary.scalar(self.name, self.tb_input, self.collections) tf.summary.scalar(self.name, tb_input, self.collections) elif self.summary_op == 'histogram': tf.summary.histogram(self.name, self.tb_input, self.collections) tf.summary.histogram(self.name, tb_input, self.collections) def copy(self, replacements={}, variables_graph=None, shared=False): """Duplicate this Layer and all its inputs. Loading deepchem/models/tensorgraph/tensor_graph.py +150 −0 Original line number Diff line number Diff line Loading @@ -842,6 +842,146 @@ class TensorGraph(Model): feed_dict[self._training_placeholder] = train_value yield feed_dict def make_estimator(self, feature_columns, weight_column=None, model_dir=None, metrics={}): """Construct a Tensorflow Estimator from this model. tf.estimator.Estimator is the standard Tensorflow API for representing models. This method provides interoperability between DeepChem and other Tensorflow based tools by allowing any model to be used an Estimator. Once this method returns, the Estimator it created is independent of the model it was created from. They do not share tensors, variables, save files, or any other resources. The Estimator is a self contained object with its own methods for training, evaluation, prediction, checkpointing, etc. Parameters ---------- feature_columns: list of tf.feature_column objects this describes the input features to the models. There must be one entry for each Feature layer in this model's features field. weight_column: tf.feature_column or None if this model includes a Weights layer, this describes the input weights. Otherwise, this should be None. metrics: map metrics that should be computed in calls to evaluate(). For each entry, the key is the name to report for the metric, and the value is a function of the form f(labels, predictions, weights) that returns the tensors for computing the metric. Any of the functions in tf.metrics can be used, as can other functions that satisfy the same interface. model_dir: str the directory in which the Estimator should save files. If None, this defaults to the model's model_dir. """ # Check the inputs. if len(feature_columns) != len(self.features): raise ValueError( 'This model requires %d feature column(s)' % len(self.features)) if len(self.labels) != 1: raise ValueError( 'Can only create an Estimator from a model with exactly one Label input' ) if len(self.task_weights) > 1: raise ValueError( 'Cannot create an Estimator from a model with multiple Weight inputs') if weight_column is None: if len(self.task_weights) > 0: raise ValueError('This model requires a weight column') else: if len(self.task_weights) == 0: raise ValueError( 'Cannot specify weight_column for a model with no Weight inputs') if model_dir is None: model_dir = self.model_dir # Define a function that recursively creates tensors from layers. def create_tensors(layer, tensors, training): if layer in tensors: return tensors[layer] inputs = [ create_tensors(in_layer, tensors, training) for in_layer in layer.in_layers ] tensor = layer.create_tensor( in_layers=inputs, set_tensors=False, training=training) tensors[layer] = tensor layer.add_summary_to_tg(tensor) return tensor # Define the model function. def model_fn(features, labels, mode): # Define the inputs. tensors = self.create_estimator_inputs(feature_columns, weight_column, features, labels, mode) for layer, tensor in tensors.items(): layer.add_summary_to_tg(tensor) # Create the correct outputs, based on the mode. if mode == tf.estimator.ModeKeys.PREDICT: predictions = {} for i, output in enumerate(self.outputs): predictions[i] = create_tensors(output, tensors, 0) return tf.estimator.EstimatorSpec(mode, predictions=predictions) if mode == tf.estimator.ModeKeys.EVAL: loss = create_tensors(self.loss, tensors, 0) predictions = create_tensors(self.outputs[0], tensors, 0) if len(self.task_weights) == 0: weights = None else: weights = tensors[self.task_weights[0]] eval_metric_ops = {} for name, function in metrics.items(): eval_metric_ops[name] = function(tensors[self.labels[0]], predictions, weights) return tf.estimator.EstimatorSpec( mode, loss=loss, eval_metric_ops=eval_metric_ops) if mode == tf.estimator.ModeKeys.TRAIN: loss = create_tensors(self.loss, tensors, 1) global_step = tf.train.get_global_step() optimizer = self.optimizer._create_optimizer(global_step) train_op = optimizer.minimize(loss, global_step=global_step) return tf.estimator.EstimatorSpec(mode, loss=loss, train_op=train_op) raise ValueError('Unknown mode') # Create the Estimator. return tf.estimator.Estimator(model_fn=model_fn, model_dir=model_dir) def create_estimator_inputs(self, feature_columns, weight_column, features, labels, mode): """This is called by make_estimator() to create tensors for the inputs. feature_columns and weight_column are the arguments passed to make_estimator(). features, labels, and mode are the arguments passed to the estimator's model function. This method creates and returns a dict with one entry for every Feature, Label, or Weights layer in the graph. The keys are the layers, and the values are the tensors that correspond to them. Any subclass that overrides default_generator() must also override this method. """ if self.__class__.default_generator != TensorGraph.default_generator: raise ValueError( "Class overrides default_generator() but not create_estimator_inputs()" ) tensors = {} for layer, column in zip(self.features, feature_columns): tensors[layer] = tf.feature_column.input_layer(features, [column]) if weight_column is not None: tensors[self.task_weights[0]] = tf.feature_column.input_layer( features, [weight_column]) if labels is not None: tensors[self.labels[0]] = tf.cast(labels, self.labels[0].dtype) return tensors def _enqueue_batch(tg, generator, graph, sess, n_enqueued, final_sample): """ Loading @@ -865,6 +1005,16 @@ def _enqueue_batch(tg, generator, graph, sess, n_enqueued, final_sample): for layer in tg.features + tg.labels + tg.task_weights: if layer in feed_dict: value = feed_dict[layer] # Add or remove dimensions of size 1 to match the shape of the layer. value_dims = len(value.shape) layer_dims = len(layer.shape) if value_dims < layer_dims: if all(i == 1 for i in layer.shape[value_dims:]): value = value.reshape( list(value.shape) + [1] * (layer_dims - value_dims)) if value_dims > layer_dims: if all(i == 1 for i in value.shape[layer_dims:]): value = value.reshape(value.shape[:layer_dims]) else: value = np.zeros( [0] + list(layer.shape[1:]), dtype=layer.dtype.as_numpy_dtype) Loading deepchem/models/tensorgraph/tests/test_estimators.py 0 → 100644 +95 −0 Original line number Diff line number Diff line import unittest import numpy as np import tensorflow as tf import deepchem as dc from deepchem.data import NumpyDataset class TestEstimators(unittest.TestCase): """ Test converting TensorGraphs to Estimators. """ def test_multi_task_classifier(self): """Test creating an Estimator from a MultiTaskClassifier.""" n_samples = 10 n_features = 3 n_tasks = 2 # Create a dataset and an input function for processing it. np.random.seed(123) X = np.random.rand(n_samples, n_features) y = np.zeros((n_samples, n_tasks)) dataset = dc.data.NumpyDataset(X, y) def input_fn(epochs): x, y, weights = dataset.make_iterator( batch_size=n_samples, epochs=epochs).get_next() return {'x': x, 'weights': weights}, y # Create a TensorGraph model. model = dc.models.MultiTaskClassifier(n_tasks, n_features, dropouts=0) # Create an estimator from it. x_col = tf.feature_column.numeric_column('x', shape=(n_features,)) weight_col = tf.feature_column.numeric_column('weights', shape=(n_tasks,)) def accuracy(labels, predictions, weights): return tf.metrics.accuracy(labels, tf.round(predictions), weights) metrics = {'accuracy': accuracy} estimator = model.make_estimator( feature_columns=[x_col], weight_column=weight_col, metrics=metrics) # Train the model. estimator.train(input_fn=lambda: input_fn(100)) # Evaluate the model. results = estimator.evaluate(input_fn=lambda: input_fn(1)) assert results['loss'] < 1e-4 assert results['accuracy'] > 0.9 def test_multi_task_regressor(self): """Test creating an Estimator from a MultiTaskRegressor.""" n_samples = 10 n_features = 3 n_tasks = 2 # Create a dataset and an input function for processing it. np.random.seed(123) X = np.random.rand(n_samples, n_features) y = np.zeros((n_samples, n_tasks)) dataset = dc.data.NumpyDataset(X, y) def input_fn(epochs): x, y, weights = dataset.make_iterator( batch_size=n_samples, epochs=epochs).get_next() return {'x': x, 'weights': weights}, y # Create a TensorGraph model. model = dc.models.MultiTaskRegressor(n_tasks, n_features, dropouts=0) # Create an estimator from it. x_col = tf.feature_column.numeric_column('x', shape=(n_features,)) weight_col = tf.feature_column.numeric_column('weights', shape=(n_tasks,)) metrics = {'error': tf.metrics.mean_absolute_error} estimator = model.make_estimator( feature_columns=[x_col], weight_column=weight_col, metrics=metrics) # Train the model. estimator.train(input_fn=lambda: input_fn(100)) # Evaluate the model. results = estimator.evaluate(input_fn=lambda: input_fn(1)) assert results['loss'] < 1e-3 assert results['error'] < 0.1 Loading
deepchem/models/tensorgraph/fcnet.py +17 −23 Original line number Diff line number Diff line Loading @@ -18,7 +18,7 @@ from deepchem.metrics import to_one_hot, from_one_hot 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, ReduceSum from deepchem.models.tensorgraph.layers import Feature, Label, Weights, WeightedError, Dense, Dropout, WeightDecay, Reshape, SoftMax, SoftMaxCrossEntropy, L2Loss, ReduceSum logger = logging.getLogger(__name__) Loading Loading @@ -114,15 +114,16 @@ class MultiTaskClassifier(TensorGraph): # Compute the loss function for each label. output = Reshape( logits = Reshape( shape=(-1, n_tasks, n_classes), in_layers=[ Dense(in_layers=[prev_layer], out_channels=n_tasks * n_classes) ]) output = SoftMax(logits) self.add_output(output) labels = Label(shape=(None, n_tasks, n_classes)) weights = Weights(shape=(None, n_tasks)) loss = SoftMaxCrossEntropy(in_layers=[labels, output]) loss = SoftMaxCrossEntropy(in_layers=[labels, logits]) weighted_loss = WeightedError(in_layers=[loss, weights]) if weight_decay_penalty != 0.0: weighted_loss = WeightDecay( Loading Loading @@ -154,6 +155,19 @@ class MultiTaskClassifier(TensorGraph): feed_dict[self.task_weights[0]] = w_b yield feed_dict def create_estimator_inputs(self, feature_columns, weight_column, features, labels, mode): tensors = {} for layer, column in zip(self.features, feature_columns): tensors[layer] = tf.feature_column.input_layer(features, [column]) if weight_column is not None: tensors[self.task_weights[0]] = tf.feature_column.input_layer( features, [weight_column]) if labels is not None: tensors[self.labels[0]] = tf.one_hot( tf.cast(labels, tf.int32), self.n_classes) return tensors def predict_proba(self, dataset, transformers=[], outputs=None): return super(MultiTaskClassifier, self).predict(dataset, transformers, outputs) Loading Loading @@ -291,26 +305,6 @@ class MultiTaskRegressor(TensorGraph): in_layers=[weighted_loss]) self.set_loss(weighted_loss) def default_generator(self, dataset, epochs=1, predict=False, deterministic=True, pad_batches=True): for epoch in range(epochs): for (X_b, y_b, w_b, ids_b) in dataset.iterbatches( batch_size=self.batch_size, deterministic=deterministic, pad_batches=pad_batches): feed_dict = dict() if y_b is not None and not predict: feed_dict[self.labels[0]] = y_b.reshape(-1, self.n_tasks, 1) if X_b is not None: feed_dict[self.features[0]] = X_b if w_b is not None and not predict: feed_dict[self.task_weights[0]] = w_b yield feed_dict class MultiTaskFitTransformRegressor(MultiTaskRegressor): """Implements a MultiTaskRegressor that performs on-the-fly transformation during fit/predict. Loading
deepchem/models/tensorgraph/layers.py +16 −10 Original line number Diff line number Diff line Loading @@ -152,9 +152,8 @@ class Layer(object): def set_summary(self, summary_op, summary_description=None, collections=None): """Annotates a tensor with a tf.summary operation Collects data from self.out_tensor by default but can be changed by setting self.tb_input to another tensor in create_tensor This causes self.out_tensor to be logged to Tensorboard. Parameters ---------- Loading @@ -175,21 +174,28 @@ class Layer(object): self.collections = collections self.tensorboard = True def add_summary_to_tg(self): def add_summary_to_tg(self, tb_input=None): """ Can only be called after self.create_layer to gaurentee that name is not none Create the summary operation for this layer, if set_summary() has been called on it. Can only be called after self.create_layer to guarantee that name is not None. Parameters ---------- tb_input: tensor the tensor to log to Tensorboard. If None, self.out_tensor is used. """ if self.tensorboard == False: return if self.tb_input == None: self.tb_input = self.out_tensor if tb_input == None: tb_input = self.out_tensor if self.summary_op == "tensor_summary": tf.summary.tensor_summary(self.name, self.tb_input, self.summary_description, self.collections) tf.summary.tensor_summary(self.name, tb_input, self.summary_description, self.collections) elif self.summary_op == 'scalar': tf.summary.scalar(self.name, self.tb_input, self.collections) tf.summary.scalar(self.name, tb_input, self.collections) elif self.summary_op == 'histogram': tf.summary.histogram(self.name, self.tb_input, self.collections) tf.summary.histogram(self.name, tb_input, self.collections) def copy(self, replacements={}, variables_graph=None, shared=False): """Duplicate this Layer and all its inputs. Loading
deepchem/models/tensorgraph/tensor_graph.py +150 −0 Original line number Diff line number Diff line Loading @@ -842,6 +842,146 @@ class TensorGraph(Model): feed_dict[self._training_placeholder] = train_value yield feed_dict def make_estimator(self, feature_columns, weight_column=None, model_dir=None, metrics={}): """Construct a Tensorflow Estimator from this model. tf.estimator.Estimator is the standard Tensorflow API for representing models. This method provides interoperability between DeepChem and other Tensorflow based tools by allowing any model to be used an Estimator. Once this method returns, the Estimator it created is independent of the model it was created from. They do not share tensors, variables, save files, or any other resources. The Estimator is a self contained object with its own methods for training, evaluation, prediction, checkpointing, etc. Parameters ---------- feature_columns: list of tf.feature_column objects this describes the input features to the models. There must be one entry for each Feature layer in this model's features field. weight_column: tf.feature_column or None if this model includes a Weights layer, this describes the input weights. Otherwise, this should be None. metrics: map metrics that should be computed in calls to evaluate(). For each entry, the key is the name to report for the metric, and the value is a function of the form f(labels, predictions, weights) that returns the tensors for computing the metric. Any of the functions in tf.metrics can be used, as can other functions that satisfy the same interface. model_dir: str the directory in which the Estimator should save files. If None, this defaults to the model's model_dir. """ # Check the inputs. if len(feature_columns) != len(self.features): raise ValueError( 'This model requires %d feature column(s)' % len(self.features)) if len(self.labels) != 1: raise ValueError( 'Can only create an Estimator from a model with exactly one Label input' ) if len(self.task_weights) > 1: raise ValueError( 'Cannot create an Estimator from a model with multiple Weight inputs') if weight_column is None: if len(self.task_weights) > 0: raise ValueError('This model requires a weight column') else: if len(self.task_weights) == 0: raise ValueError( 'Cannot specify weight_column for a model with no Weight inputs') if model_dir is None: model_dir = self.model_dir # Define a function that recursively creates tensors from layers. def create_tensors(layer, tensors, training): if layer in tensors: return tensors[layer] inputs = [ create_tensors(in_layer, tensors, training) for in_layer in layer.in_layers ] tensor = layer.create_tensor( in_layers=inputs, set_tensors=False, training=training) tensors[layer] = tensor layer.add_summary_to_tg(tensor) return tensor # Define the model function. def model_fn(features, labels, mode): # Define the inputs. tensors = self.create_estimator_inputs(feature_columns, weight_column, features, labels, mode) for layer, tensor in tensors.items(): layer.add_summary_to_tg(tensor) # Create the correct outputs, based on the mode. if mode == tf.estimator.ModeKeys.PREDICT: predictions = {} for i, output in enumerate(self.outputs): predictions[i] = create_tensors(output, tensors, 0) return tf.estimator.EstimatorSpec(mode, predictions=predictions) if mode == tf.estimator.ModeKeys.EVAL: loss = create_tensors(self.loss, tensors, 0) predictions = create_tensors(self.outputs[0], tensors, 0) if len(self.task_weights) == 0: weights = None else: weights = tensors[self.task_weights[0]] eval_metric_ops = {} for name, function in metrics.items(): eval_metric_ops[name] = function(tensors[self.labels[0]], predictions, weights) return tf.estimator.EstimatorSpec( mode, loss=loss, eval_metric_ops=eval_metric_ops) if mode == tf.estimator.ModeKeys.TRAIN: loss = create_tensors(self.loss, tensors, 1) global_step = tf.train.get_global_step() optimizer = self.optimizer._create_optimizer(global_step) train_op = optimizer.minimize(loss, global_step=global_step) return tf.estimator.EstimatorSpec(mode, loss=loss, train_op=train_op) raise ValueError('Unknown mode') # Create the Estimator. return tf.estimator.Estimator(model_fn=model_fn, model_dir=model_dir) def create_estimator_inputs(self, feature_columns, weight_column, features, labels, mode): """This is called by make_estimator() to create tensors for the inputs. feature_columns and weight_column are the arguments passed to make_estimator(). features, labels, and mode are the arguments passed to the estimator's model function. This method creates and returns a dict with one entry for every Feature, Label, or Weights layer in the graph. The keys are the layers, and the values are the tensors that correspond to them. Any subclass that overrides default_generator() must also override this method. """ if self.__class__.default_generator != TensorGraph.default_generator: raise ValueError( "Class overrides default_generator() but not create_estimator_inputs()" ) tensors = {} for layer, column in zip(self.features, feature_columns): tensors[layer] = tf.feature_column.input_layer(features, [column]) if weight_column is not None: tensors[self.task_weights[0]] = tf.feature_column.input_layer( features, [weight_column]) if labels is not None: tensors[self.labels[0]] = tf.cast(labels, self.labels[0].dtype) return tensors def _enqueue_batch(tg, generator, graph, sess, n_enqueued, final_sample): """ Loading @@ -865,6 +1005,16 @@ def _enqueue_batch(tg, generator, graph, sess, n_enqueued, final_sample): for layer in tg.features + tg.labels + tg.task_weights: if layer in feed_dict: value = feed_dict[layer] # Add or remove dimensions of size 1 to match the shape of the layer. value_dims = len(value.shape) layer_dims = len(layer.shape) if value_dims < layer_dims: if all(i == 1 for i in layer.shape[value_dims:]): value = value.reshape( list(value.shape) + [1] * (layer_dims - value_dims)) if value_dims > layer_dims: if all(i == 1 for i in value.shape[layer_dims:]): value = value.reshape(value.shape[:layer_dims]) else: value = np.zeros( [0] + list(layer.shape[1:]), dtype=layer.dtype.as_numpy_dtype) Loading
deepchem/models/tensorgraph/tests/test_estimators.py 0 → 100644 +95 −0 Original line number Diff line number Diff line import unittest import numpy as np import tensorflow as tf import deepchem as dc from deepchem.data import NumpyDataset class TestEstimators(unittest.TestCase): """ Test converting TensorGraphs to Estimators. """ def test_multi_task_classifier(self): """Test creating an Estimator from a MultiTaskClassifier.""" n_samples = 10 n_features = 3 n_tasks = 2 # Create a dataset and an input function for processing it. np.random.seed(123) X = np.random.rand(n_samples, n_features) y = np.zeros((n_samples, n_tasks)) dataset = dc.data.NumpyDataset(X, y) def input_fn(epochs): x, y, weights = dataset.make_iterator( batch_size=n_samples, epochs=epochs).get_next() return {'x': x, 'weights': weights}, y # Create a TensorGraph model. model = dc.models.MultiTaskClassifier(n_tasks, n_features, dropouts=0) # Create an estimator from it. x_col = tf.feature_column.numeric_column('x', shape=(n_features,)) weight_col = tf.feature_column.numeric_column('weights', shape=(n_tasks,)) def accuracy(labels, predictions, weights): return tf.metrics.accuracy(labels, tf.round(predictions), weights) metrics = {'accuracy': accuracy} estimator = model.make_estimator( feature_columns=[x_col], weight_column=weight_col, metrics=metrics) # Train the model. estimator.train(input_fn=lambda: input_fn(100)) # Evaluate the model. results = estimator.evaluate(input_fn=lambda: input_fn(1)) assert results['loss'] < 1e-4 assert results['accuracy'] > 0.9 def test_multi_task_regressor(self): """Test creating an Estimator from a MultiTaskRegressor.""" n_samples = 10 n_features = 3 n_tasks = 2 # Create a dataset and an input function for processing it. np.random.seed(123) X = np.random.rand(n_samples, n_features) y = np.zeros((n_samples, n_tasks)) dataset = dc.data.NumpyDataset(X, y) def input_fn(epochs): x, y, weights = dataset.make_iterator( batch_size=n_samples, epochs=epochs).get_next() return {'x': x, 'weights': weights}, y # Create a TensorGraph model. model = dc.models.MultiTaskRegressor(n_tasks, n_features, dropouts=0) # Create an estimator from it. x_col = tf.feature_column.numeric_column('x', shape=(n_features,)) weight_col = tf.feature_column.numeric_column('weights', shape=(n_tasks,)) metrics = {'error': tf.metrics.mean_absolute_error} estimator = model.make_estimator( feature_columns=[x_col], weight_column=weight_col, metrics=metrics) # Train the model. estimator.train(input_fn=lambda: input_fn(100)) # Evaluate the model. results = estimator.evaluate(input_fn=lambda: input_fn(1)) assert results['loss'] < 1e-3 assert results['error'] < 0.1
