InvalidArgumentError: Input to reshape is a tensor with 3200 values, but the requested shape requires a multiple of 49 [[{{node decoder/Reshape}}]] about deep-learning-with-python

[u112358]

No description provided.

change x = tf.layers.dense(x, units=inputs_decoder * 2 + 1, activation=lrelu) to x = tf.layers.dense(x, units=inputs_decoder * 2, activation=lrelu) in decoder() will solve this problem.

[AdityaAmrutiya]

I am also getting the same issue

[erdalalim]

@AdityaAmrutiya @alla15747 you solve it?
Iam trying to run face creation, its prepared to generate 40x40 pixel images.. I want to generate 200x200 pixel so I change 40,40,3 values to 200,200,3 to get 200pixel images but Iam facing this error :

InvalidArgumentError: Input to reshape is a tensor with 307200 values, but the requested shape requires a multiple of 120000
[[Node: discriminator_1/Reshape = Reshape[T=DT_FLOAT, Tshape=DT_INT32, _device="/job:localhost/replica:0/task:0/device:CPU:0"](generator/conv2d_transpose_3/Sigmoid, discriminator/Reshape/shape)]]

any suggestions,In which line I need to make correction?

[erdalalim]

Hi @u112358 ,

i can't find the line x = tf.layers.dense(x, units=inputs_decoder * 2 + 1, activation=lrelu) in script (DCGAN-face-creation.ipynb), the code you mentioned is in VAE.ipynb.

i only need to change the size of the generated images from default size 40x40 to any other size example between: 100x100 or 200x200 pixels. When i tried to change the sizes i'm getting the error above. Do you have any idea how to change the sizes of images ?

thanks

[arnouri]

No description provided.

change x = tf.layers.dense(x, units=inputs_decoder * 2 + 1, activation=lrelu) to x = tf.layers.dense(x, units=inputs_decoder * 2, activation=lrelu) in decoder() will solve this problem.

This is a full year later so sorry for reviving this but I was wondering if you could provide any insight on why this fix works?
Thanks!