tf.compat.v1.summary.merge_all  |  TensorFlow v2.16.1 (original) (raw)

tf.compat.v1.summary.merge_all

Stay organized with collections Save and categorize content based on your preferences.

Merges all summaries collected in the default graph.

tf.compat.v1.summary.merge_all(
    key=_ops.GraphKeys.SUMMARIES, scope=None, name=None
)

Migrate to TF2

This API is not compatible with eager execution or tf.function. To migrate to TF2, this API can be omitted entirely, because in TF2 individual summary ops, like tf.summary.scalar(), write directly to the default summary writer if one is active. Thus, it's not necessary to merge summaries or to manually add the resulting merged summary output to the writer. See the usage example shown below.

For a comprehensive tf.summary migration guide, please followMigrating tf.summary usage to TF 2.0.

TF1 & TF2 Usage Example

TF1:

dist = tf.compat.v1.placeholder(tf.float32, [100])
tf.compat.v1.summary.histogram(name="distribution", values=dist)
writer = tf.compat.v1.summary.FileWriter("/tmp/tf1_summary_example")
summaries = tf.compat.v1.summary.merge_all()

sess = tf.compat.v1.Session()
for step in range(100):
  mean_moving_normal = np.random.normal(loc=step, scale=1, size=[100])
  summ = sess.run(summaries, feed_dict={dist: mean_moving_normal})
  writer.add_summary(summ, global_step=step)

TF2:

writer = tf.summary.create_file_writer("/tmp/tf2_summary_example")
for step in range(100):
  mean_moving_normal = np.random.normal(loc=step, scale=1, size=[100])
  with writer.as_default(step=step):
    tf.summary.histogram(name='distribution', data=mean_moving_normal)

Description