tensorflow中迭代产生数据时结果通过tensorboard显示

本方法从stack overflow获得

You can create a tf.Summary object in your Python program and write it to the same tf.summary.FileWriter object that takes your TensorFlow-produced summaries using the SummaryWriter.add_summary() method.

The tf.Summary class is a Python protocol buffer wrapper for the Summary protocol buffer. Each Summary contains a list of tf.Summary.Value protocol buffers, which each have a tag and a either a "simple" (floating-point scalar) value, an image, a histogram, or an audio snippet. For example, you can generate a scalar summary from a Python object as follows:

writer = tf.train.SummaryWriter(...)
value = 37.0
summary = tf.Summary(value=[
    tf.Summary.Value(tag="summary_tag", simple_value=value), 
])
writer.add_summary(summary）

 

   summary_writer2 = tf.summary.FileWriter("./test", sess.graph)

loss = 0.0
loss_summary = tf.Summary()
loss_summary.value.add(tag='sum cross-entropy', simple_value=loss)

for i in range(MAX_STEPS):

    for step1 in range(1, validation_prop):
        cross_entropy2 = sess.run(cross_entropy, feed_dict={
             x: X[(step1 - 1) * n_steps * batch_size: step1 * n_steps * batch_size],
             y_: Y[(step1 - 1) * n_steps * batch_size: step1 * n_steps * batch_size]})
             loss += cross_entropy2
    loss = loss/(validation_prop)
    print("loss: ", loss)
    loss_summary.value[0].simple_value = loss
    summary_writer2.add_summary(loss_summary, num_step)

这里给写乱了，如果直接使用不妨看我另一篇，直接copy然后使用即可。

tensorflow中针对迭代产生数据和对列表数据不使用feed_dict进行tensorboard表示