UniversalStyleTransfer icon indicating copy to clipboard operation
UniversalStyleTransfer copied to clipboard

Published 20 hours ago • verified publisher icon Yijunmaverick

CPU mode?

Open CJHFUTURE opened this issue 7 years ago • 5 comments

Hi is there a CPU mode only flag available?

CJHFUTURE avatar Oct 22 '17 06:10 CJHFUTURE

@Sugarbank Thanks for your interests about our work.

Yes, we provide the option for running in CPU mode:

th test_wct.lua -gpu -1

Yijunmaverick avatar Oct 22 '17 06:10 Yijunmaverick

Could someone provide model snapshots for CPU? (or can they be converted without cuda support?

jjmontesl avatar Dec 10 '17 19:12 jjmontesl

@jjmontesl

Hi, do you mean that you cannot load my model trained under GPU? Did you install cudnn?

Yijunmaverick avatar Dec 13 '17 08:12 Yijunmaverick

Hello. Thanks for responding. I don't have a Cuda GPU, hence I cannot install cudnn and wanted to run from CPU.

When loading models, torch complains because the models are serialized including Cuda classes that are not available in my runtime. The issue seems well known (https://groups.google.com/forum/#!topic/torch7/ugBCwaoXw_s ).

I am entirely new to Torch and maybe this has an easy solution, but I have since tried the Tensorflow implementation you referenced which doesn't raise this issue.

Best,

jjmontesl avatar Dec 13 '17 11:12 jjmontesl

@Sugarbank @jjmontesl Sorry for the late reply. Since I use the GPU during the training, the decoders require the cudnn for the loading.

I need to explicitly transfer the models to float so that you can load in CPU mode. Please find them here.

Yijunmaverick avatar Feb 05 '18 23:02 Yijunmaverick