老旧tensorflow的api对比
tf.nn.seq2seq.sequence_loss_by_example -> tf.contrib.legacy_seq2seq.sequence_loss_by_example
tf.nn.rnn_cell -> tf.contrib.rnn
tensorflow.Variable与 get_variable的区别
简而言之, get_variable用户共享的变量, 所以使用了两次一模一样的参数的时候, 就会提示错误;
而 tf.variable_scope("scope1", reuse=True)后, 下面再次出现一模一样的参数, 就会去那个共享的variable。
import tensorflow as tf
with tf.variable_scope("scope1"):
w1 = tf.get_variable("w1", shape=[])
w2 = tf.Variable(0.0, name="w2")
with tf.variable_scope("scope1", reuse=True):
w1_p = tf.get_variable("w1", shape=[])
w2_p = tf.Variable(1.0, name="w2")
print(w1 is w1_p, w2 is w2_p)
#输出
#True False