Model parallelism

[pedropgusmao]

Hi,
I am trying to generate HD images, but results are not so good and I guess it is because I am training the models using lower resolution images (256x256).
Unfortunately, if I try to train these networks using higher resolution images I run out of memory. So, I was wondering if it is possible to split the networks over multiple GPUs and run the model over multiple devices.

Thanks

[junyanz]

This multi-gpu support would be a great feature.
With the current implementation, you can also crop patches from high-res images (e.g. crop 256x256 patches from 512x512 images. Set loadSize=512, and fineSize=256 for both training and test). This will save you gpu memory during training, while still allows you to test your model on high-res images.