为什么我的tf_gradients返回None？

Question

# Defining the tf ops 
prob_placeholder = tf.placeholder(tf.float32, shape=(2)) 
log_placeholder = tf.log(prob_placeholder) 
grads_placeholder = tf.gradients(ys=tf.log(prob_placeholder), xs=model.weights) 


# t is some index into the holders (which are lists) 
# s is some state || p_a is some list of [p_1, 1 - p_1] || a_ is either 0 or 1 || r is 1 

prob_ = tf_sess.run(prob_placeholder, {prob_placeholder: p_a}) 
log_ = tf_sess.run(log_placeholder, {prob_placeholder: prob_}) 
print(prob_, log_) 
grads_ = tf_sess.run(grads_placeholder, {prob_placeholder: prob_}) 

基本上我不知道它为什么返回无。

TypeError: Fetch argument None has invalid type <type 'NoneType'> 

我试着添加打印语句，我可以看到prob_和LOG_出来就好了，但我不知道正在发生的事情，是造成上述问题的tf.gradients。

model.weights基本上是我正在使用的模型的权重。

Answer 1

prob_placeholder对model.weights没有任何明确的依赖性，即它在功能上不依赖于您定义它们的方式model.weights。

因此，尽管技术上梯度应为零，但由于technical reasons in TensorFlow，因此计算为None。