Model Optimizer is able to convert .caffemodel to IR format for inference engine to read it.
I wonder whether you have some way to convert these IR models back to .caffemodel.
The LPRNet model `\license-plate-recognition-barrier-0001.xml` inprovided have very good performance on Chinese License Plate Recognition,
If I want to fine-tune this LPRNet on my own dataset, I wonder whether there is a way to convert LPRNet model saved as IR to caffemodel, so that it is possible to load the pretrained parameters in caffe?
Thank you very much.
The non-public pre-trained models such as LPRNet cannot be converted back to source code as we don't have a tool to do that and we aren't planning to give the source code away for these models currently.