一口酥 issues

Results 9 issues of


                                            一口酥

In factor vae: an inplace problem occurred when using loss.backward

the error message is: `one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [4096, 6]], which is output 0 of TBackward, is at...

Inplace problem when calling D_tc_loss.backward()

the error message is: `one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [4096, 6]], which is output 0 of TBackward, is at...

how to segement the image with a mask?

hey there, now I have a mask(binary), which indicates possible foreground of the image, and one original image corresponding to the mask. How can i use this for segmented region...

I tried DeepNextFace myself, but landmark loss decreases slowly.

Hey there, thanks for your impressive work! I tried Nextface first and now implemented DeepNextFace by myself, using default resnet152 in pytorch(pretrained on ImageNet), but during step 1, landmark loss...

No module named 'circus_zoos'

Hi, I encountered with the error as title said when running ur test code, but I can't find `circus_zoos` in your code or pypi

can this model predict gender?

Thanks for sharing your work, my question is as title said.

what's the difference between v1.4 and v1.3? how can it's result improve so much?

thanks for your great work! i found no specific explaination of why v1.4 improved. Is it possible to share some details?

is it possible to open closed eyes or close opened eyes?

hi there, thanks for your great work! couldn't find any demo on editing eyes (open/close), quiet interested about it. waiting for your response.