pix2pix without B side image,how to test about pytorch-cyclegan-and-pix2pix HOT 4 CLOSED

I just committed the new code for doing that. When generating images, please use

--model=one_direction_test --max_dataset_size=20

This will load [...]_net_G.pth and generate 20 samples without loading the model of the other direction.

from pytorch-cyclegan-and-pix2pix.

hi, thank u for your reply and your new code,
i retrain with the your new code using the following cmd
training cmd:

python train.py --dataroot ./datasets/faceid/folderAB --name facaid_pix2pix --model pix2pix --which_model_netG unet_256 --which_direction BtoA --lambda_A 100 --align_data --use_dropout --no_lsgan --niter 3000 --continue_train

and the test cmd I use is this:
python test.py --dataroot datasets/9 --name facaid_pix2pix --model one_direction_test --which_direction BtoA --align_data --use_dropout
is this cmd right？

And errors occured :

• model.test()
• File "/pytorch/GAN_NEW/pytorch-CycleGAN-and-pix2pix/models/one_direction_test_model.py", line 41, in test
• self.fake_B = self.netG_A.forward(self.real_A)
• File "/home//GAN_NEW/pytorch-CycleGAN-and-pix2pix/models/networks.py", line 225, in forward
• return nn.parallel.data_parallel(self.model, input, self.gpu_ids)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/parallel/data_parallel.py", line 100, in data_parallel
• return module(*inputs[0], **module_kwargs[0])
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line 206, in call
• result = self.forward(*input, **kwargs)
• File "/home/public/pytorch/GAN_NEW/pytorch-CycleGAN-and-pix2pix/models/networks.py", line 276, in forward
• return self.model(x)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line 206, in call
• result = self.forward(*input, **kwargs)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/container.py", line 64, in forward
• input = module(input)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line 206, in call
• result = self.forward(*input, **kwargs)
• File "/home/public/itorch/zsl_exercise/pytorch/GAN_NEW/pytorch-CycleGAN-and-
• input = module(input)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line 206, in call
• result = self.forward(*input, **kwargs)
• File "/home/public/itorch/zsl_exercise/pytorch/GAN_NEW/pytorch-CycleGAN-and-pix2pix/models/networks.py", line 278, in forward
• return torch.cat([self.model(x), x], 1)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line 206, in call
• result = self.forward(*input, **kwargs)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/container.py", line 64, in forward
• input = module(input)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line 206, in call
• result = self.forward(*input, **kwargs)
• File "/home/public/itorch/zsl_exercise/pytorch/GAN_NEW/pytorch-CycleGAN-and-pix2pix/models/networks.py", line 278, in forward
• return torch.cat([self.model(x), x], 1)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line 206, in call
• result = self.forward(*input, **kwargs)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/container.py", line 64, in forward
• input = module(input)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line 206, in call
• result = self.forward(*input, **kwargs)
• File "/home/public/itorch/zsl_exercise/pytorch/GAN_NEW/pytorch-CycleGAN-and-pix2pix/models/networks.py", line 278, in forward
• return torch.cat([self.model(x), x], 1)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line 206, in call
• result = self.forward(*input, **kwargs)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/container.py", line 64, in forward
• input = module(input)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line 206, in call
• result = self.forward(*input, **kwargs)
• File "/home/public/itorch/zsl_exercise/pytorch/GAN_NEW/pytorch-CycleGAN-and-pix2pix/models/networks.py", line 278, in forward
• return torch.cat([self.model(x), x], 1)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line 206, in call
• result = self.forward(*input, **kwargs)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/container.py", line 64, in forward
• input = module(input)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line 206, in call
• result = self.forward(*input, **kwargs)
• File "/home/public/itorch/zsl_exercise/pytorch/GAN_NEW/pytorch-CycleGAN-and-pix2pix/models/networks.py", line 278, in forward
• return torch.cat([self.model(x), x], 1)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line 206, in call
• result = self.forward(*input, **kwargs)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/container.py", line 64, in forward
• input = module(input)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line 206, in call
• result = self.forward(*input, **kwargs)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/conv.py", line 237, in forward
• self.padding, self.dilation, self.groups)
• File "/usr/local/lib/python2.7/dist-packages/torch/nn/functional.py", line 39, in conv2d
• return f(input, weight, bias)
• RuntimeError: CUDNN_STATUS_BAD_PARAM

the A and B image size is 128128
in AB folder images are 256128
training is ok。
but A Image in my testset is 128*128

my test folder is datasets/9 A image is dataset/9/test is this right？
thank u so much

from pytorch-cyclegan-and-pix2pix.

Will this commit be merged? e5b2fd6

from pytorch-cyclegan-and-pix2pix.

It has been added. You can use --model test. See here for more details.

from pytorch-cyclegan-and-pix2pix.