1.tf.train.piecewise_constant
走到一定步长更改学习率。
initial_learning_rate = 0.1 * params['batch_size'] / 128
batches_per_epoch = _NUM_IMAGES['train'] / params['batch_size']
global_step = tf.train.get_or_create_global_step()
boundaries = [int(batches_per_epoch * epoch) for epoch in [100, 150, 200]]
values = [initial_learning_rate * decay for decay in [1, 0.1, 0.01, 0.001]]
learning_rate = tf.train.piecewise_constant(
tf.cast(global_step, tf.int32), boundaries, values)2.tf.identity
https://blog.csdn.net/hu_guan_jie/article/details/78495297
没看懂。
tf.identity是返回了一个一模一样新的tensor，再control_dependencies的作用块下，需要增加一个新节点到gragh中。
3.tf.summary.scalar
对标量数据汇总和记录