Why does Tensorflow Reshape tf.reshape () break the flow of gradients?

Question

I create tf.Variable () and then create a simple function using this variable, then smooth the original variable with tf.reshape (), and then take tf.gradients () between the function and the flattened variable. Why does this return [No].

var = tf.Variable(np.ones((5,5)), dtype = tf.float32)
f = tf.reduce_sum(tf.reduce_sum(tf.square(var)))
var_f = tf.reshape(var, [-1])
print tf.gradients(f,var_f)

The above code block upon execution returns [No]. This is mistake? Please, help!

Answer 1

You find the derivative of fwith respect to var_f, but finstead is not a function var_f, but var. That is why you get [No]. Now, if you change the code to:

 var = tf.Variable(np.ones((5,5)), dtype = tf.float32)
 var_f = tf.reshape(var, [-1])
 f = tf.reduce_sum(tf.reduce_sum(tf.square(var_f)))
 grad = tf.gradients(f,var_f)
 print(grad)

your gradients will be determined:

tf.Tensor 'gradients_28/Square_32_grad/mul_1: 0' shape = (25,) dtype = float32 >

:

 var = tf.Variable(np.ones((5,5)), dtype = tf.float32, name='var')
 f = tf.reduce_sum(tf.reduce_sum(tf.square(var)), name='f')
 var_f = tf.reshape(var, [-1], name='var_f')
 grad_1 = tf.gradients(f,var_f, name='grad_1')
 grad_2 = tf.gradients(f,var, name='grad_2')

grad_1 , 'grad_2` . ( ) .