When use batch_size>2 about mobilestylegan.pytorch HOT 9 CLOSED

@zhongtao93 it is an out of memory error

from mobilestylegan.pytorch.

But I have four 32G GPUs, it's also not work when use --gpus 4

from mobilestylegan.pytorch.

@zhongtao93 try to decrease batch size

from mobilestylegan.pytorch.

Thank you, decrease to 3 is work. Another question, the Prediction head of MobileStyleGAN is only useful for training, when model forward, those part can skip. It's correct?

from mobilestylegan.pytorch.

@zhongtao93 auxiliary heads will be skipped when MobileStyleGAN converts to ONNX or CoreML

from mobilestylegan.pytorch.

Thank you. Do you think this model is likely to run realtime with the output size of 256x256 on the mobile phone? After reduce the number of channels or other Pruning method.

from mobilestylegan.pytorch.

@zhongtao93 it works on iPhone 12 for 4fps with the output size of 1024x1024, we didn't test for less resolution.

from mobilestylegan.pytorch.

Thank you! I will continue to study the work of improving speed.

from mobilestylegan.pytorch.

Thank you. Do you think this model is likely to run realtime with the output size of 256x256 on the mobile phone? After reduce the number of channels or other Pruning method.

Hi,I have a same question.Did you have any results?

from mobilestylegan.pytorch.