Question

我正在关注tensorboard tutorial以了解如何使用摘要，并且我不断收到以下错误。它在调用时抛出错误：

summary_op = tf.merge_all_summaries()

但问题似乎在这里：

def variable_summaries(var):
  with tf.name_scope('summaries'):
    mean = tf.reduce_mean(var)
    tf.scalar_summary('mean', mean)

任何线索为什么会发生这种情况？

InvalidArgumentError（参见上面的回溯）：重复标记的意思在摘要输入中找到[[Node：MergeSummary / MergeSummary = MergeSummary [N = 7， _device = “/作业：本地主机/复制：0 /任务：0 / CPU：0”]（层1 /权重/摘要/ ScalarSummary，层1 /偏压/摘要/ ScalarSummary，二层/权重/摘要/ ScalarSummary，二层/偏压/摘要/ ScalarSummary，出/重量/摘要/ ScalarSummary， out / biases / summary / ScalarSummary，ScalarSummary）]]

Answer 1

问题在于我获得代码的源代码。 tensorboard link将我们带到代码错误的主版本：

def variable_summaries(var):
  """Attach a lot of summaries to a Tensor (for TensorBoard visualization)."""
  with tf.name_scope('summaries'):
    mean = tf.reduce_mean(var)
    tf.summary.scalar('mean', mean)
    with tf.name_scope('stddev'):
      stddev = tf.sqrt(tf.reduce_mean(tf.square(var - mean)))
    tf.summary.scalar('stddev', stddev)
    tf.summary.scalar('max', tf.reduce_max(var))
    tf.summary.scalar('min', tf.reduce_min(var))
    tf.summary.histogram('histogram', var)

其他版本具有以下正确的代码：

def variable_summaries(var, name):
  """Attach a lot of summaries to a Tensor."""
  with tf.name_scope('summaries'):
    mean = tf.reduce_mean(var)
    tf.scalar_summary('mean/' + name, mean)
    with tf.name_scope('stddev'):
      stddev = tf.sqrt(tf.reduce_mean(tf.square(var - mean)))
    tf.scalar_summary('stddev/' + name, stddev)
    tf.scalar_summary('max/' + name, tf.reduce_max(var))
    tf.scalar_summary('min/' + name, tf.reduce_min(var))
    tf.histogram_summary(name, var)

Tensorflow：scalar_summary重复异常

1 个答案: