一下内容为笔者实际应用中遇到的问题，可能（必须）不全，后面会持续更新。
(1) tf.nn.run_cell 改为 tf.contrib.rnn
(2) tf.reduce_mean 改为 tf.reduce.sum
(3) tf.nn.seq2seq.sequence_loss_by_example 改为
tf.contrib.legacy_seq2seq.sequence_loss_by_example
(4) tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logit(y,y_)) 改为
tf.reduce_sum(tf.nn.softmax_cross_entropy_with_logit(logits=logits,labels=y_))
(5) tf.train.SummaryWriter 改为： tf.summary.FileWriter
(6) tf.merge_all_summary() 改为： tf.summary.merge_all()
(7) tf.histogram_summary() 改为： tf.summary.histogram()
(8) tf.scalar_summary() 改为： tf.summary.scalar()
(9) tf.image_summary() 改为： tf.summary.image()
(10) tf.concat(1,[indices,sparse_labels]) 改为： tf.concat([indices,sparse_labels],1)
(11) TypeError: Using a `tf.Tensor` as a Python `bool` is not allowed. Use `if t is not None:` instead of `if t:` to test if a tensor is defined, and use TensorFlow ops such as tf.cond to execute subgraphs conditioned on the value of a tensor.
if grad: 改为 if grad is not None: