Will different image width and height in training set cause errors?

[wang5566]

My training set images'width and height is different, (e.g. width is 80 and height is 200). what the parameter of imgsize should I choose?
and my training set A images' channel is 3 while B images' channel is 1, can I use "--A_channels 3 --B_channels 1 " to run the code?

[duxingren14]

The codes currently only supports images of width==height. For the latter case, "--A_channels 3 --B_channels 1 " is just what you need.

…

On 6 November 2017 at 23:59, wang5566 ***@***.***> wrote: My training set images'width and height is different, (e.g. width is 80 and height is 200). what the parameter of imgsize should I choose? and my training set A images' channel is 3 while B images' channel is 1, can I use "--A_channels 3 --B_channels 1 " to run the code? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#14>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACc6exIv-tZikEt9r1vcWSa9HeyzpbRVks5s0A3PgaJpZM4QUZBw> .

[wang5566]

Thank you for your reply. I resize my training set to the same size but when run the code, I got the error message "ValueError: Dimension 1 in both shapes must be equal, but are 0 and 1 for 'concat' (op: 'ConcatV2') with input shapes: [1,0,0,512], [1,1,1,512], [] and with computed input tensors: input[2] = <3>." I don't know what's wrong with my operations? Can you give me some advice?

[duxingren14]

The size must be 128, 256, 512, 2^n...

…

On 7 November 2017 at 00:51, wang5566 ***@***.***> wrote: Thank you for your reply. I resize my training set to the same size but when run the code, I got the error message "ValueError: Dimension 1 in both shapes must be equal, but are 0 and 1 for 'concat' (op: 'ConcatV2') with input shapes: [1,0,0,512], [1,1,1,512], [] and with computed input tensors: input[2] = <3>." I don't know what's wrong with my operations? Can you give me some advice? — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#14 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACc6e6QIsvRQaLFWmfQQ9RC3PA6fIOBkks5s0BoTgaJpZM4QUZBw> .

[wang5566]

Thank you so much. I got everything done.

[wang5566]

when I set --A_channels 3 --B_channels 1, I got message "Cannot feed value of shape (1, 256, 256, 3) for Tensor u'real_B:0', which has shape '(1, 256, 256, 1)'",so what can I do?

[o0t1ng0o]

@wang5566
I also met the problem.

But i check the images, they are all 2562563.
Could u please help us? @duxingren14

Traceback (most recent call last):
File "main.py", line 63, in
tf.app.run()
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/platform/app.py", line 48, in run
_sys.exit(main(_sys.argv[:1] + flags_passthrough))
File "main.py", line 58, in main
model.train(args)
File "/home/valentina/DualGAN-master/model.py", line 159, in train
self.run_optim(imgA_batch, imgB_batch, step, start_time)
File "/home/valentina/DualGAN-master/model.py", line 178, in run_optim
feed_dict = {self.real_A: batch_A_imgs, self.real_B: batch_B_imgs})
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 789, in run
run_metadata_ptr)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 975, in _run
% (np_val.shape, subfeed_t.name, str(subfeed_t.get_shape())))
ValueError: Cannot feed value of shape (1, 256, 256, 1) for Tensor u'real_B:0', which has shape '(1, 256, 256, 3)'

[allen-ash]

@o0t1ng0o Please check again images in you data set B to see whether it's channels equals to 1.

[xiaoshidai]

Hello, I want to know if the test sample can be different from the training sample size? If so, what should I do?

[duxingren14]

Sorry, unequal width and height is not supported yet. I will try add this feature in future version.

…

On 18 April 2018 at 07:56, xiaoshidai ***@***.***> wrote: Hello, I want to know if the test sample can be different from the training sample size? If so, what should I do? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#14 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACc6e7YedbS28UlhDjvMCiz1H7KAKcFNks5tp1QFgaJpZM4QUZBw> .