Images with different input shape

[fm94]

When changing img_size = 64 to count for images with different width and height i.e., also changing the process_img to:

def process_img(file_path, img_w, img_h):
  img = tf.io.read_file(file_path)
  img = tf.image.decode_jpeg(img, channels=3)
  img = tf.image.convert_image_dtype(img, tf.float32)
  img = tf.image.resize(img, size=(img_h, img_w), method='area')
  img = tf.image.convert_image_dtype(img, tf.uint8)
  return img

and the definition of the discriminator to:

def make_discriminator_model():
  disc = tf.keras.Sequential(
    [
      layers.Input(shape=(img_h, img_w, 3)),
...

and line 391 to: all_images.append(process_img(item, img_w, img_h))

The training crashes after the first epoch with: ValueError: Cannot reshape a tensor with 49152 elements to shape [32,128] (4096 elements) for '{{node Reshape}} = Reshape[T=DT_FLOAT, Tshape=DT_INT32](discriminator/flatten/Reshape, Reshape/shape)' with input shapes: [32,1536], [2] and with input tensors computed as partial shapes: input[1] = [32,128].

Apparently, there is a problem here: real_first_output = tf.reshape(self.discriminator(images[i,...], training=True), (batch_size, disc_dim)). Somehow, the discriminator is not outputting correct dimension for reshaping. I have doubts that something is wrong with disc_dim but couldn't solve it.

I am using:

img_w = 256
img_h = 144

[muxamilian]

The neural network's size is fixed. If you feed larger input images, also the output of the neural network is going to be larger. Then the shapes don't match anymore.

There are two options:

• Manually changing the neural network by, for example, adding an additional layer, since the input images are now larger.
• Resizing the input to 64x64

In general, any input size that isn't a square is going to be difficult because then the convolutional layers aren't symmetric anymore and it can be quite tedious to design them so that the shapes match everywhere.