blade6570 / learningimage-to-imagetranslationusingpairedandunpairedtrainingsamples

Learning image-to-image translation using paired and unpaired training samples

learningimage-to-imagetranslationusingpairedandunpairedtrainingsamples's Issues

Error after 2 epochs

model [CycleGANModel] was created
Setting up a new session...
create web directory ./checkpoints/mygan_70/web...
End of epoch 1 / 200 Time Taken: 24 sec
learning rate = 0.0002000
(epoch: 2, iters: 37, time: 0.579) D_A: 0.282 G_A: 0.268 Cyc_A: 0.662 D_B: 0.284 G_B: 0.402 Cyc_B: 0.700 idt_A: 0.391 idt_B: 0.440
Traceback (most recent call last):
File "train.py", line 50, in
visualizer.plot_current_errors(epoch, float(epoch_iter)/dataset_size, opt, errors)
File "/media/rusica/0E3E135A0E3E135A/MoizBackup/cycgan/pairunpair/Learningimage-to-imagetranslationusingpairedandunpairedtrainingsamples/util/visualizer.py", line 113, in plot_current_errors
win=self.display_id)
File "/home/rusica/anaconda3/lib/python3.7/site-packages/visdom/init.py", line 389, in wrapped_f
return f(*args, **kwargs)
File "/home/rusica/anaconda3/lib/python3.7/site-packages/visdom/init.py", line 1715, in line
update=update, name=name)
File "/home/rusica/anaconda3/lib/python3.7/site-packages/visdom/init.py", line 389, in wrapped_f
return f(*args, **kwargs)
File "/home/rusica/anaconda3/lib/python3.7/site-packages/visdom/init.py", line 1640, in scatter
return self._send(data_to_send, endpoint=endpoint)
File "/home/rusica/anaconda3/lib/python3.7/site-packages/visdom/init.py", line 711, in _send
data=json.dumps(msg),
File "/home/rusica/anaconda3/lib/python3.7/json/init.py", line 231, in dumps
return _default_encoder.encode(obj)
File "/home/rusica/anaconda3/lib/python3.7/json/encoder.py", line 199, in encode
chunks = self.iterencode(o, _one_shot=True)
File "/home/rusica/anaconda3/lib/python3.7/json/encoder.py", line 257, in iterencode
return _iterencode(o, 0)
File "/home/rusica/anaconda3/lib/python3.7/json/encoder.py", line 179, in default
raise TypeError(f'Object of type {o.class.name} '
TypeError: Object of type Tensor is not JSON serializable

Kindly help me out!

naming convention for the paired data

Hi
Thanks for your nice work!
Can you please specify how to put up the paired data in the train sub-folder?
Any specific naming convention that we should follow? like image_a_1, image_a_2, image_b_1, image_b_2, ... etc ?

Any suggestions will be helpful for running the code.

Thanks
Devraj

Test questions

I have trained well and tested using the command you provided, but the result shows that the model parameters do not match, I have checked carefully, the training model and test model are cycleGAN, it is really strange.
(Error screenshot)

Getting the below error for cycleGAN mode

CustomDatasetDataLoader_super
dataset [AlignedDataset] was created
#training images = 300
cycle_gan
Traceback (most recent call last):
File "train.py", line 22, in
model = create_model(opt)
File "/home/kalai/exp/Learningimage-to-imagetranslationusingpairedandunpairedtrainingsamples/models/models.py", line 7, in create_model
from .cycle_gan_model import CycleGANModel
File "/home/kalai/exp/Learningimage-to-imagetranslationusingpairedandunpairedtrainingsamples/models/cycle_gan_model.py", line 97
input_A = input_A.cuda(self.gpu_ids[0], async=True)
^
SyntaxError: invalid syntax

getting training error Please help

I have added the dataset in the mentioned format ... but i got this error
Please Help

Where is the adversarial consistency loss?

Hello Mr. Tripathy,

I'm sorry if I overlooked it, but I can't find a model using the conditional discriminator for the adversarial consistency loss nor an implementation of it.

Greetings

Getting this error message while trying pix2pix mode

!python train.py --dataroot ./datasets --model pix2pix --dataset_mode aligned --which_model_netG resnet_9blocks --which_direction AtoB --super_epoch 50 --super_epoch_start 0 --super_mode aligned --super_start 1 --name mygan_70 --no_dropout

Traceback (most recent call last):
File "train.py", line 39, in
model.optimize_parameters(opt)
TypeError: optimize_parameters() takes 1 positional argument but 2 were given