Question

我在运行sess.run（不参考训练步长）时，模型中的权重得到更新的事实困扰着我。

我尝试用变量填充模型以获取估计的输出，但是当我运行sess.run时，权重得到更新。

### in the training phase ####
X_eval, Y_eval, O_eval, W_eval, cost_eval, train_step_eval = sess.run([X, Y, O_out, W, cost, train_step], feed_dict={X:x_batch , Y:y_batch})

### when the training is finished (closed for loop) ###
print(W_eval)
Y_out, W_eval2 = sess.run([O_out, W], feed_dict = {X:labeled_features[:,: - n_labels], Y:labeled_features[:,- n_labels :]})
print(W_eval2)

当我比较W_eval和W_eval2时，它们并不相同，我不明白为什么。您能指出我正确的方向吗，为什么权重不一样？

'w3': array([[-2.9685912],
   [-3.215485 ],
   [ 3.8806837],
   [-3.331745 ],
   [-3.3904853]], dtype=float32

'w3': array([[-2.9700036],
   [-3.2168453],
   [ 3.8804765],
   [-3.3330843],
   [-3.3922129]], dtype=float32

谢谢。

编辑。添加了W_eval分配。

Answer 1

您的代码

### in the training phase ####
X_eval, Y_eval, O_eval, W_eval, cost_eval, train_step_eval = sess.run([X, Y, O_out, W, cost, train_step], feed_dict={X:x_batch , Y:y_batch})

### when the training is finished (closed for loop) ###
print(W_eval)
Y_out, W_eval2 = sess.run([O_out, W], feed_dict = {X:labeled_features[:,: - n_labels], Y:labeled_features[:,- n_labels :]})
print(W_eval2)

仍执行train_step。理解发生了什么的一个更简单的版本是：

import tensorflow as tf

a = tf.get_variable('a', initializer=42.)

train_step = a.assign(a + 1)

with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    value, _ = sess.run([a, train_step])  # will update a
    print(value)
    value = sess.run([a])  # will not update a
    print(value)
    value = sess.run([a])  # will not update a
    print(value)

提供输出

42.0
[43.0]
[43.0]

要检查的另一件事是x_batch == labeled_features[:,: - n_labels]是否成立。

Tensorflow模型的权重已在sess.run

1 个答案: