Replace graph_def with graph when passed to the SummaryWriter and update tutorials to reflect the new API.

Change: 118033430

12 files changed, 24 insertions, 37 deletions

diff --git a/tensorflow/examples/how_tos/reading_data/fully_connected_preloaded.py b/tensorflow/examples/how_tos/reading_data/fully_connected_preloaded.py
index ef46329cd5..d08a15433a 100644
--- a/tensorflow/examples/how_tos/reading_data/fully_connected_preloaded.py
+++ b/tensorflow/examples/how_tos/reading_data/fully_connected_preloaded.py
@@ -103,8 +103,7 @@ def run_training():
     sess.run(init_op)
 
     # Instantiate a SummaryWriter to output summaries and the Graph.
-    summary_writer = tf.train.SummaryWriter(FLAGS.train_dir,
-                                            graph_def=sess.graph_def)
+    summary_writer = tf.train.SummaryWriter(FLAGS.train_dir, sess.graph)
 
     # Start input enqueue threads.
     coord = tf.train.Coordinator()
diff --git a/tensorflow/examples/how_tos/reading_data/fully_connected_preloaded_var.py b/tensorflow/examples/how_tos/reading_data/fully_connected_preloaded_var.py
index b8bd143ec8..3b43abc81c 100644
--- a/tensorflow/examples/how_tos/reading_data/fully_connected_preloaded_var.py
+++ b/tensorflow/examples/how_tos/reading_data/fully_connected_preloaded_var.py
@@ -113,8 +113,7 @@ def run_training():
              feed_dict={labels_initializer: data_sets.train.labels})
 
     # Instantiate a SummaryWriter to output summaries and the Graph.
-    summary_writer = tf.train.SummaryWriter(FLAGS.train_dir,
-                                            graph_def=sess.graph_def)
+    summary_writer = tf.train.SummaryWriter(FLAGS.train_dir, sess.graph)
 
     # Start input enqueue threads.
     coord = tf.train.Coordinator()
diff --git a/tensorflow/examples/tutorials/mnist/fully_connected_feed.py b/tensorflow/examples/tutorials/mnist/fully_connected_feed.py
index 8a9884a508..eda1ac5b59 100644
--- a/tensorflow/examples/tutorials/mnist/fully_connected_feed.py
+++ b/tensorflow/examples/tutorials/mnist/fully_connected_feed.py
@@ -163,8 +163,7 @@ def run_training():
     sess.run(init)
 
     # Instantiate a SummaryWriter to output summaries and the Graph.
-    summary_writer = tf.train.SummaryWriter(FLAGS.train_dir,
-                                            graph_def=sess.graph_def)
+    summary_writer = tf.train.SummaryWriter(FLAGS.train_dir, sess.graph)
 
     # And then after everything is built, start the training loop.
     for step in xrange(FLAGS.max_steps):
diff --git a/tensorflow/examples/tutorials/mnist/mnist_with_summaries.py b/tensorflow/examples/tutorials/mnist/mnist_with_summaries.py
index 33dc13c813..e637032b2e 100644
--- a/tensorflow/examples/tutorials/mnist/mnist_with_summaries.py
+++ b/tensorflow/examples/tutorials/mnist/mnist_with_summaries.py
@@ -82,8 +82,7 @@ def main(_):
 
   # Merge all the summaries and write them out to /tmp/mnist_logs (by default)
   merged = tf.merge_all_summaries()
-  writer = tf.train.SummaryWriter(FLAGS.summaries_dir,
-                                  sess.graph.as_graph_def(add_shapes=True))
+  writer = tf.train.SummaryWriter(FLAGS.summaries_dir, sess.graph)
   tf.initialize_all_variables().run()
 
   # Train the model, and feed in test data and record summaries every 10 steps
diff --git a/tensorflow/g3doc/how_tos/graph_viz/index.md b/tensorflow/g3doc/how_tos/graph_viz/index.md
index 77c9332790..a4689c492b 100644
--- a/tensorflow/g3doc/how_tos/graph_viz/index.md
+++ b/tensorflow/g3doc/how_tos/graph_viz/index.md
@@ -233,10 +233,9 @@ The images below give an illustration for a piece of a real-life graph.
 
 When the serialized `GraphDef` includes tensor shapes, the graph visualizer
 labels edges with tensor dimensions, and edge thickness reflects total tensor
-size. To include tensor shapes in the `GraphDef` pass
-`sess.graph.as_graph_def(add_shapes=True)` to the `SummaryWriter` when
-serializing the graph. The images below show the CIFAR-10 model with tensor
-shape information:
+size. To include tensor shapes in the `GraphDef` pass the actual graph object
+(as in `sess.graph`) to the `SummaryWriter` when serializing the graph.
+The images below show the CIFAR-10 model with tensor shape information:
 <table width="100%;">
   <tr>
     <td style="width: 100%;">
diff --git a/tensorflow/g3doc/how_tos/summaries_and_tensorboard/index.md b/tensorflow/g3doc/how_tos/summaries_and_tensorboard/index.md
index 5059f02a73..39a5661f24 100644
--- a/tensorflow/g3doc/how_tos/summaries_and_tensorboard/index.md
+++ b/tensorflow/g3doc/how_tos/summaries_and_tensorboard/index.md
@@ -55,11 +55,10 @@ Finally, to write this summary data to disk, pass the summary protobuf to a
 
 The `SummaryWriter` takes a logdir in its constructor - this logdir is quite
 important, it's the directory where all of the events will be written out.
-Also, the `SummaryWriter` can optionally take a `GraphDef` in its constructor.
-If it receives one, then TensorBoard will visualize your graph as well.
-To include tensor shape information in the `GraphDef`, pass
-`sess.graph.as_graph_def(add_shapes=True)` to the `SummaryWriter`. This will
-give you a much better sense of what flows through the graph: see
+Also, the `SummaryWriter` can optionally take a `Graph` in its constructor.
+If it receives a `Graph` object, then TensorBoard will visualize your graph
+along with tensor shape information. This will give you a much better sense of
+what flows through the graph: see
 [Tensor shape information](../../how_tos/graph_viz/index.md#tensor-shape-information).
 
 Now that you've modified your graph and have a `SummaryWriter`, you're ready to
@@ -106,8 +105,7 @@ with tf.name_scope("test") as scope:
 
 # Merge all the summaries and write them out to /tmp/mnist_logs
 merged = tf.merge_all_summaries()
-writer = tf.train.SummaryWriter("/tmp/mnist_logs",
-                                sess.graph.as_graph_def(add_shapes=True))
+writer = tf.train.SummaryWriter("/tmp/mnist_logs", sess.graph)
 tf.initialize_all_variables().run()
 
 # Train the model, and feed in test data and record summaries every 10 steps
diff --git a/tensorflow/models/embedding/word2vec.py b/tensorflow/models/embedding/word2vec.py
index 196d15e9e4..cf30548e14 100644
--- a/tensorflow/models/embedding/word2vec.py
+++ b/tensorflow/models/embedding/word2vec.py
@@ -395,8 +395,7 @@ class Word2Vec(object):
     initial_epoch, initial_words = self._session.run([self._epoch, self._words])
 
     summary_op = tf.merge_all_summaries()
-    summary_writer = tf.train.SummaryWriter(opts.save_path,
-                                            graph_def=self._session.graph_def)
+    summary_writer = tf.train.SummaryWriter(opts.save_path, self._session.graph)
     workers = []
     for _ in xrange(opts.concurrent_steps):
       t = threading.Thread(target=self._train_thread_body)
diff --git a/tensorflow/models/image/cifar10/cifar10_eval.py b/tensorflow/models/image/cifar10/cifar10_eval.py
index 57eb94c3c9..e94cc3f2f5 100644
--- a/tensorflow/models/image/cifar10/cifar10_eval.py
+++ b/tensorflow/models/image/cifar10/cifar10_eval.py
@@ -115,7 +115,7 @@ def eval_once(saver, summary_writer, top_k_op, summary_op):
 
 def evaluate():
   """Eval CIFAR-10 for a number of steps."""
-  with tf.Graph().as_default():
+  with tf.Graph().as_default() as g:
     # Get images and labels for CIFAR-10.
     eval_data = FLAGS.eval_data == 'test'
     images, labels = cifar10.inputs(eval_data=eval_data)
@@ -136,9 +136,7 @@ def evaluate():
     # Build the summary operation based on the TF collection of Summaries.
     summary_op = tf.merge_all_summaries()
 
-    graph_def = tf.get_default_graph().as_graph_def(add_shapes=True)
-    summary_writer = tf.train.SummaryWriter(FLAGS.eval_dir,
-                                            graph_def=graph_def)
+    summary_writer = tf.train.SummaryWriter(FLAGS.eval_dir, g)
 
     while True:
       eval_once(saver, summary_writer, top_k_op, summary_op)
diff --git a/tensorflow/models/image/cifar10/cifar10_multi_gpu_train.py b/tensorflow/models/image/cifar10/cifar10_multi_gpu_train.py
index 3e3d21cb53..da4565ff6c 100644
--- a/tensorflow/models/image/cifar10/cifar10_multi_gpu_train.py
+++ b/tensorflow/models/image/cifar10/cifar10_multi_gpu_train.py
@@ -239,9 +239,7 @@ def train():
     # Start the queue runners.
     tf.train.start_queue_runners(sess=sess)
 
-    graph_def = sess.graph.as_graph_def(add_shapes=True)
-    summary_writer = tf.train.SummaryWriter(FLAGS.train_dir,
-                                            graph_def=graph_def)
+    summary_writer = tf.train.SummaryWriter(FLAGS.train_dir, sess.graph)
 
     for step in xrange(FLAGS.max_steps):
       start_time = time.time()
diff --git a/tensorflow/models/image/cifar10/cifar10_train.py b/tensorflow/models/image/cifar10/cifar10_train.py
index cf59894a81..a224ecbf2e 100644
--- a/tensorflow/models/image/cifar10/cifar10_train.py
+++ b/tensorflow/models/image/cifar10/cifar10_train.py
@@ -93,9 +93,7 @@ def train():
     # Start the queue runners.
     tf.train.start_queue_runners(sess=sess)
 
-    graph_def = sess.graph.as_graph_def(add_shapes=True)
-    summary_writer = tf.train.SummaryWriter(FLAGS.train_dir,
-                                            graph_def=graph_def)
+    summary_writer = tf.train.SummaryWriter(FLAGS.train_dir, sess.graph)
 
     for step in xrange(FLAGS.max_steps):
       start_time = time.time()
diff --git a/tensorflow/python/summary/event_accumulator_test.py b/tensorflow/python/summary/event_accumulator_test.py
index 2050c346cd..8dafaad0db 100644
--- a/tensorflow/python/summary/event_accumulator_test.py
+++ b/tensorflow/python/summary/event_accumulator_test.py
@@ -596,9 +596,11 @@ class RealisticEventAccumulatorTest(EventAccumulatorTest):
     gfile.MkDir(directory)
 
     writer = tf.train.SummaryWriter(directory, max_queue=100)
-    graph_def = tf.GraphDef(node=[tf.NodeDef(name='A', op='Mul')])
+
+    with tf.Graph().as_default() as graph:
+      _ = tf.constant([2.0, 1.0])
     # Add a graph to the summary writer.
-    writer.add_graph(graph_def)
+    writer.add_graph(graph)
 
     run_metadata = tf.RunMetadata()
     device_stats = run_metadata.step_stats.dev_stats.add()
@@ -651,7 +653,7 @@ class RealisticEventAccumulatorTest(EventAccumulatorTest):
       self.assertEqual(i * 5, sq_events[i].step)
       self.assertEqual(i, id_events[i].value)
       self.assertEqual(i * i, sq_events[i].value)
-    self.assertProtoEquals(graph_def, acc.Graph())
+    self.assertProtoEquals(graph.as_graph_def(add_shapes=True), acc.Graph())
 
 
 if __name__ == '__main__':
diff --git a/tensorflow/python/training/summary_writer_test.py b/tensorflow/python/training/summary_writer_test.py
index fcc095020d..79a7ecef24 100644
--- a/tensorflow/python/training/summary_writer_test.py
+++ b/tensorflow/python/training/summary_writer_test.py
@@ -81,8 +81,7 @@ class SummaryWriterTestCase(tf.test.TestCase):
                    20)
     with tf.Graph().as_default() as g:
       tf.constant([0], name="zero")
-    gd = g.as_graph_def()
-    sw.add_graph(gd, global_step=30)
+    sw.add_graph(g, global_step=30)
 
     run_metadata = tf.RunMetadata()
     device_stats = run_metadata.step_stats.dev_stats.add()
@@ -124,7 +123,7 @@ class SummaryWriterTestCase(tf.test.TestCase):
     self.assertEquals(30, ev.step)
     ev_graph = tf.GraphDef()
     ev_graph.ParseFromString(ev.graph_def)
-    self.assertProtoEquals(gd, ev_graph)
+    self.assertProtoEquals(g.as_graph_def(add_shapes=True), ev_graph)
 
     # The next event should have metadata for the run.
     ev = next(rr)