yaxingwang / transferring-gans Goto Github PK

and why do u choose conditional gan to pre_trained.

Sorry to bother you. But when I download the pre-train models form Tencent. There only has four folder. 'imagenet', 'places', 'celebA', 'bedroom', not including 'conditional' model you have mentioned?
Can you provide a link for me to download pretrained 'conditional' model from Tencent?
Thanks a lot ! Thanks for your great work!

Hello, Thanks for your work and for providing this code! I am wondering how you created the Bedroom subsets for Table 2 / Fig 2, as I would like to replicate these results. Do you just take the first N images of the Bedrooms training dataset? Also, I wonder if you have any code for doing the FID/IW evaluations. Thank you!

Hi, author,
I can't download the pretrained models, do you have any other way for getting it?
Thanks.

HI!
I got tow errors,and do not know to solve it.
my computer:
tf:1.13.1 python:3.7
GPU:1060TI 6G

2019-03-28 21:37:47.625690: E tensorflow/core/grappler/optimizers/dependency_optimizer.cc:704] Iteration = 0, topological sort failed with message: The graph couldn't be sorted in topological order.
2019-03-28 21:37:47.709614: E tensorflow/core/grappler/optimizers/dependency_optimizer.cc:704] Iteration = 1, topological sort failed with message: The graph couldn't be sorted in topological order.
2019-03-28 21:37:48.412559: E tensorflow/core/grappler/optimizers/dependency_optimizer.cc:704] Iteration = 0, topological sort failed with message: The graph couldn't be sorted in topological order.
2019-03-28 21:37:48.446716: E tensorflow/core/grappler/optimizers/dependency_optimizer.cc:704] Iteration = 1, topological sort failed with message: The graph couldn't be sorted in topological order.
2019-03-28 21:37:49.504099: E tensorflow/core/common_runtime/executor.cc:624] Executor failed to create kernel. Invalid argument: Conv2DCustomBackpropInputOp only supports NHWC.
[[{{node gradients/Discriminator.Res4.Shortcut_2/Conv2D_grad/Conv2DBackpropInput}}]]
Traceback (most recent call last):
File "D:\Anaconda3\lib\site-packages\tensorflow\python\client\session.py", line 1334, in _do_call
return fn(*args)
File "D:\Anaconda3\lib\site-packages\tensorflow\python\client\session.py", line 1319, in _run_fn
options, feed_dict, fetch_list, target_list, run_metadata)
File "D:\Anaconda3\lib\site-packages\tensorflow\python\client\session.py", line 1407, in _call_tf_sessionrun
run_metadata)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Conv2DCustomBackpropInputOp only supports NHWC.
[[{{node gradients/Discriminator.Res4.Shortcut_2/Conv2D_grad/Conv2DBackpropInput}}]]

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "C:\Users\Administrator\Desktop\Transferring-GANs-master\Transferring-GANs-master\transfer_gan.py", line 353, in
feed_dict={all_real_data_conv: _images, all_real_labels: _labels})
File "D:\Anaconda3\lib\site-packages\tensorflow\python\client\session.py", line 929, in run
run_metadata_ptr)
File "D:\Anaconda3\lib\site-packages\tensorflow\python\client\session.py", line 1152, in _run
feed_dict_tensor, options, run_metadata)
File "D:\Anaconda3\lib\site-packages\tensorflow\python\client\session.py", line 1328, in _do_run
run_metadata)
File "D:\Anaconda3\lib\site-packages\tensorflow\python\client\session.py", line 1348, in _do_call
raise type(e)(node_def, op, message)

tensorflow.python.framework.errors_impl.InvalidArgumentError: Conv2DCustomBackpropInputOp only supports NHWC.

so,hoping your reply.

Hi Yaxing,

I was just wondering how you pretrained your AC-GAN;

Did you train it on ImageNet with all 1000 classes?

Does that then mean you always have to prefix a one-hot vector of length 1000 to the noise vector even when fine-tuning on a small number of classes?

If not, do you mind letting me know how this process works?

Thanks in advance!