caffe2_cpp_tutorial
caffe2_cpp_tutorial copied to clipboard
[Question] Switch workspace
Hi,
Thanks a lot for your sample they were really helpful to build my own network. Now I have a new problem as I want to run the same model on different cameras, therefore I would love to re-use the same Workspace if it's already existing, so I won't run out of memory by loading the model for every run.
Do you have any suggestions how this could be done please or if it possible?
Thanks very much :)
If I understand correctly, you can get this done by simply keeping the same Workspace around. Just provide your model with new input and run inference. Even if images come from different camera, this should work.