python - sess.run() 的 tensorflow 值错误

Question

我试着玩了一下 tensorflow，但似乎我做错了什么，我制作的小程序:

import tensorflow as tf

x = tf.placeholder(tf.float64)
y = tf.placeholder(tf.float64)

test = {"A":tf.Variable(tf.random_normal([20, 20])),
        "B":tf.Variable(tf.random_normal([20, 20]))}

math_stuff = tf.matmul(x,y)

with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    print(sess.run(math_stuff, feed_dict={x:test["A"], y:test["B"]}))

我想用两个 20x20 随机数组查看 tf.matmul(x,y) 的结果。它向我抛出的错误:

Traceback (most recent call last):
  File "C:\Users\Utilisateur\AppData\Local\Programs\Python\Python36\save\tensorflow_play.py", 
 line 15, in <module> print(sess.run(math_stuff, feed_dict={x:test["A"], y:test["B"]}))
  File "C:\Users\Utilisateur\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\client\session.py",
  line 889, in run run_metadata_ptr)
  File "C:\Users\Utilisateur\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\client\session.py", 
 line 1089, in _run np_val = np.asarray(subfeed_val, dtype=subfeed_dtype)
 File "C:\Users\Utilisateur\AppData\Local\Programs\Python\Python36\lib\site-packages\numpy\core\numeric.py", 
 line 531, in asarray return array(a, dtype, copy=False, order=order)
 ValueError: setting an array element with a sequence.

Answer 1

feed_dict 应该包含数值，不是 tf.Variable。将 test 的定义替换为:

test = {"A":np.random.randn(20,20),
        "B":np.random.randn(20,20)}

当然，您还应该在开头import numpy as np。然后代码会按您希望的方式运行。

为了进一步解释，您可以将 feed_dict 视为您赋予计算图的数值，而不是部分计算图(如 tf.Variable 那样)。