You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thanks for sharing the code base and awesome work in view synthesis for Fashion! I tried training/finetuning the model with Batch size = 2, which is essentially one pair of data with a custom dataset (Image 512 x 352 and pose obtained and formatted from Openpose). I am getting CUDA out of memory at the lowest possible settings. I use an Nvidia RTX GPU with 20Gb size x 8 (number of GPUs). Could you please suggest some tips or tricks that can reduce memory utilization?
The text was updated successfully, but these errors were encountered:
Hi authors,
Thanks for sharing the code base and awesome work in view synthesis for Fashion! I tried training/finetuning the model with Batch size = 2, which is essentially one pair of data with a custom dataset (Image 512 x 352 and pose obtained and formatted from Openpose). I am getting CUDA out of memory at the lowest possible settings. I use an Nvidia RTX GPU with 20Gb size x 8 (number of GPUs). Could you please suggest some tips or tricks that can reduce memory utilization?
The text was updated successfully, but these errors were encountered: