Open SaharR1372 opened 2 years ago
ckkelvinchan
commented
2 years ago (1) The error log said CUDA out of memory. You may need to reduce the batch size or patch size.
(2) and (4) I will need to check whether I still have it. I may have accidentally deleted that during a cleaning in the past. But the training settings are basically identical.
(3) No, since they use different StyleGANs.
SaharR1372
commented
2 years ago I have another question. The lowest input size to train GLEAN would be just 32x32 (according to figure in your paper)? we just can feed input size of 32x32 because weight of styleGAN is available for 256? Therefore, we can not train the network input size of 16x16?
If there are any details regarding this, please let me know because I am interested in using images with the resolution of 16 and 32.
On Wed, Apr 13, 2022 at 9:23 PM Kelvin C.K. Chan @.***> wrote:
(1) The error log said CUDA out of memory. You may need to reduce the batch size or patch size.
(2) and (4) I will need to check whether I still have it. I may have accidentally deleted that during a cleaning in the past. But the training settings are basically identical.
(3) No, since they use different StyleGANs.
— Reply to this email directly, view it on GitHub https://github.com/ckkelvinchan/GLEAN/issues/14#issuecomment-1098620221, or unsubscribe https://github.com/notifications/unsubscribe-auth/AVMDQQDEC7GCBAHBAHEVP6LVE5XRXANCNFSM5TLYKXUQ . You are receiving this because you authored the thread.Message ID: @.***>
Hi, I am really interested in your work and thanks for all the provided materials. 1-would u please help me to solve this problem along training I have attached the screenshot of it?
I have checked all the installation multiple times.
2-would you please share the weight for when your input size is 16x16?
3- is it possible to just change the input size to 16 and out to 256 and using your x16 pretained weight without any other modification?
4- would u please share the x8 weights for face images (u have just provided x16)?