support mutil-gpus?how to solve batchnormalization?

Paper99 / SRFBN_CVPR19

Pytorch code for our paper "Feedback Network for Image Super-Resolution" (CVPR2019)

MIT License

551 stars 126 forks source link

support mutil-gpus?how to solve batchnormalization? #19

Open Peterisfar opened 5 years ago

Paper99 commented 5 years ago

We will test the compatibility of multi-gpu and give your feedback in a few days. If you want to inject a BN layer in one conv/deconv block, you can change norm_type in your architecture definition file to bn. See the ConvBlock declaration for details.

sararaza534 commented 5 years ago

can we test model only on cpu.. if we can ..? what changes we have to do?