Tensorboard draw a wrong graph when I use `tf.layers.conv2d_transpose` without pass the name parameter.

Tensorboard draw a wrong graph when I use tf.layers.conv2d_transpose without pass the name parameter.

Here is my code:

import tensorflow as tf

x = tf.placeholder(tf.float32, shape=(None, None, None, 1), name="Input")
x = tf.layers.conv2d_transpose(x, kernel_size=2, filters=10)
x = tf.layers.conv2d_transpose(x, kernel_size=2, filters=10)

with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    summary_writer = tf.summary.FileWriter("tensorflow_logs/graph", graph=tf.get_default_graph())

It will draw a right graph when I fill the name parameter in the two tf.layers.conv2d_transpose.

Reproduce the Python snippet using two tf.layers.conv2d_transpose calls and inspect the resulting TensorBoard graph. Compare the graph with and without explicit name parameters, then trace the graph visualization entry point to identify the discrepancy. Done means both versions render the correct graph structure.