I want to convert the pix2pix Image to Image transform model ( https://github.com/phillipi/pix2pix ), which is built using nngraph.
When I try to convert it to a caffe model using the torch for the caffe tool ( https://github.com/facebook/fb-caffe-exts#torch2caffe ), I get the error "unknown class nn. GModule".
I am also trying to load a torch model into pytorch using load_lua from torch.utils.serialization, where I get a similar error.
Since, it seems, nngraph is not supported for conversion. Is there a way to convert an nngraph model to an nn model without the need for rewriting or retraining? In the lineup, I want to convert the pix2pix model to various formats, such as caffe, caffe2, coreML and others, to test the performance in a mobile environment.
thanks
source share