yiranran / qmupd Goto Github PK

Hi,

Just wondering, what is the difference between the pre-trained model in this repo and the one in Unpaired-Portrait-Drawing? They seem to have the same file and size.

Thanks

Hi,
can you give some basic info on how to train a model?
How big were your datasets?
Can you provide a ample command?
Thanks

excuse me, I can't find DistModel in cycle_gan_cls_model.py. Could u help me?

Your idea is so great！

But when I change the imgsize from 512 to 1024 I met this error:
File "/home/ubuntu/cyh/QMUPD/models/networks.py", line 856, in forward
y1 = torch.cat([f1, input2], 1)
RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 256 but got size 128 for tensor number 1 in the list.

Hello！Could you please tell me the following questions? Are you train the regression network for the APDrawing quality with manually annotated data？If so, how exactly are they labeled? If not, what are the negative samples for training？

Hi,
great work. I don't have access to your paper (it is behind a paywall).
Can you elaborate on how I would train a model?