Solved: openvino C++

Arthur123 · ‎09-11-2023

hello everyone!

I have a small question as to why model inference in python would use the
"net = ie.read_network(model,weights)" api to import model and parameters (.xml & .bin)

but In the c++ call samples:

"ie.compile_model" only the .xml file is imported, isn't the weights file (.bin) needed?

Expecting someone to tell me why

thanks in advance!

Hairul_Intel · ‎09-13-2023

Hi Arthur123,

Thank you for reaching out to us.

For your information, when loading a model in OpenVINO C++ API without specifying the bin_path, the core.read_model() will try to read a bin file with the same name as xml and if no bin file is found, it will load IR without weights.

You can refer the class ov::Core for more information on OpenVINO Runtime C++ API.

Hope this helps.

Regards,

Hairul

View solution in original post

Hairul_Intel · ‎09-13-2023

Hi Arthur123,

Thank you for reaching out to us.

For your information, when loading a model in OpenVINO C++ API without specifying the bin_path, the core.read_model() will try to read a bin file with the same name as xml and if no bin file is found, it will load IR without weights.

You can refer the class ov::Core for more information on OpenVINO Runtime C++ API.

Hope this helps.

Regards,

Hairul

Arthur123 · ‎09-17-2023

Get it, thanks a lot!

Hairul_Intel · ‎09-18-2023

Hi Arthur123,

This thread will no longer be monitored since this issue has been resolved. If you need any additional information from Intel, please submit a new question.

Regards,

Hairul