Merge pull request #1295 from peastman/tensorboard (c03dfa29) · Commits · 钟慕尧 / deepchem

deepchem/models/tensorgraph/layers.py

+6 −4

Original line number	Diff line number	Diff line
		@@ -230,7 +230,7 @@ class Layer(object):
		self.collections = collections
		self.tensorboard = True

		def add_summary_to_tg(self, layer_vars):
		def add_summary_to_tg(self, layer_output, layer_vars):
		"""
		Create the summary operation for this layer, if set_summary() has been called on it.

		@@ -238,6 +238,8 @@ class Layer(object):

		Parameters
		----------
		layer_output: tensor
		the output tensor to log to Tensorboard
		layer_vars: list of variables
		the list of variables to log to Tensorboard
		"""
		@@ -245,20 +247,20 @@ class Layer(object):
		return

		if self.summary_op == "tensor_summary":
		tf.summary.tensor_summary(self.name, self.out_tensor,
		tf.summary.tensor_summary(self.name, layer_output,
		self.summary_description, self.collections)
		if self.include_variables:
		for var in layer_vars:
		tf.summary.tensor_summary(var.name, var, self.summary_description,
		self.collections)
		elif self.summary_op == 'scalar':
		tf.summary.scalar(self.name, self.out_tensor, self.collections)
		tf.summary.scalar(self.name, layer_output, self.collections)
		if self.include_variables:
		for var in layer_vars:
		tf.summary.tensor_summary(var.name, var, self.collections,
		self.collections)
		elif self.summary_op == 'histogram':
		tf.summary.histogram(self.name, self.out_tensor, self.collections)
		tf.summary.histogram(self.name, layer_output, self.collections)
		if self.include_variables:
		for var in layer_vars:
		tf.summary.histogram(var.name, var, self.collections)

deepchem/models/tensorgraph/tensor_graph.py

+6 −3

Original line number	Diff line number	Diff line
		@@ -668,7 +668,8 @@ class TensorGraph(Model):
		self.rnn_initial_states += layer.rnn_initial_states
		self.rnn_final_states += layer.rnn_final_states
		self.rnn_zero_states += layer.rnn_zero_states
		layer.add_summary_to_tg(self.get_layer_variables(layer))
		layer.add_summary_to_tg(layer.out_tensor,
		self.get_layer_variables(layer))
		self.session = tf.Session(config=self.configproto)

		# Ensure all training operators have been created.
		@@ -1184,7 +1185,9 @@ class TensorGraph(Model):
		tensor = layer.create_tensor(
		in_layers=inputs, set_tensors=False, training=training)
		tensors[layer] = tensor
		layer.add_summary_to_tg(tensor)
		vars = tf.get_collection(
		tf.GraphKeys.TRAINABLE_VARIABLES, scope=layer.name)
		layer.add_summary_to_tg(tensor, vars)
		return tensor

		# Define the model function.
		@@ -1195,7 +1198,7 @@ class TensorGraph(Model):
		tensors = self.create_estimator_inputs(feature_columns, weight_column,
		features, labels, mode)
		for layer, tensor in tensors.items():
		layer.add_summary_to_tg(tensor)
		layer.add_summary_to_tg(tensor, [])

		# Create the correct outputs, based on the mode.

deepchem/models/tensorgraph/tests/test_estimators.py

+40 −2

Original line number	Diff line number	Diff line
		@@ -2,8 +2,8 @@ import unittest
		import numpy as np
		import tensorflow as tf
		import deepchem as dc
		import deepchem.models.tensorgraph.layers as layers
		from deepchem.data import NumpyDataset
		from deepchem.models.tensorgraph.layers import Dense


		class TestEstimators(unittest.TestCase):
		@@ -212,7 +212,7 @@ class TestEstimators(unittest.TestCase):
		# Create the model.

		model = dc.models.Sequential(loss="mse", learning_rate=0.01)
		model.add(Dense(out_channels=1))
		model.add(layers.Dense(out_channels=1))

		# Create an estimator from it.

		@@ -329,3 +329,41 @@ class TestEstimators(unittest.TestCase):
		results = estimator.evaluate(input_fn=lambda: input_fn(1))
		assert results['loss'] < 0.5
		assert results['accuracy'] > 0.6

		def test_tensorboard(self):
		"""Test creating an Estimator from a TensorGraph that logs information to TensorBoard."""
		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.TensorGraph()
		features = layers.Feature(shape=(None, n_features))
		dense = layers.Dense(out_channels=n_tasks, in_layers=features)
		dense.set_summary('histogram')
		model.add_output(dense)
		labels = layers.Label(shape=(None, n_tasks))
		loss = layers.ReduceMean(layers.L2Loss(in_layers=[labels, dense]))
		model.set_loss(loss)

		# Create an estimator from it.

		x_col = tf.feature_column.numeric_column('x', shape=(n_features,))
		estimator = model.make_estimator(feature_columns=[x_col])

		# Train the model.

		estimator.train(input_fn=lambda: input_fn(100))

Admin message