JDAI-CV / dabnn

dabnn is an accelerated binary neural networks inference framework for mobile platform
Other
769 stars 102 forks source link

When using onnx2bnn, the network channels c should must satisfy c %64 == 0, otherwise, there will be a problem. #23

Closed Sakulaki closed 5 years ago

Sakulaki commented 5 years ago

I0517 20:34:54.114783 15334 OnnxConverter.cpp:145] c % 64 == 0 is not satisfied! [48, 3, 3, 48] terminate called after throwing an instance of 'std::runtime_error' what(): c % 64 == 0 is not satisfied! [48, 3, 3, 48] Aborted (core dumped)

daquexian commented 5 years ago

Fixed in #49 and #51