我正在尝试将变量'theta'的值初始化:
n=tf.constant(8)
theta = tf.Variable(tf.random_uniform([n + 1, 1],-1.0, 1.0))
with tf.Session() as sess:
theta.initializer.run()
执行上述代码时遇到以下错误:
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-34-6b7fabbea243> in <module>()
1 n=tf.constant(8)
----> 2 theta = tf.Variable(tf.random_uniform([n + 1, 1],-1.0, 1.0))
3 with tf.Session() as sess:
4 theta.initializer.run()
答案 0 :(得分:0)
这应该有效。该常数应该被求值。
n = tf.constant(8)
with tf.Session() as sess:
init_n = sess.run(n) # Evaluate
theta = tf.Variable(tf.random_uniform([init_n, 1], -1.0, 1.0))
theta.initializer.run()
print( sess.run(theta) )