Conversation
| { | ||
| if (outputLayer == "detection_out") | ||
| { | ||
| if (backend == DNN_BACKEND_INFERENCE_ENGINE_NN_BUILDER_2019) |
There was a problem hiding this comment.
Can you please check why nGraph path works correctly this way? I thought outputs from both APIs should be the same.
There was a problem hiding this comment.
It strange, but nGraph works in both cases
| virtual Ptr<BackendNode> initInfEngine(const std::vector<Ptr<BackendWrapper> > &inputs); | ||
|
|
||
| virtual Ptr<BackendNode> initNgraph(const std::vector<Ptr<BackendWrapper> > &inputs, const std::vector<Ptr<BackendNode> >& nodes); | ||
| virtual Ptr<BackendNode> initNgraph(const std::vector<Ptr<BackendNode> >& nodes); |
There was a problem hiding this comment.
What do you think, it's better to rename it to initial initInfEngine? But let's discuss before change :)
modules/dnn/src/dnn.cpp
Outdated
| { | ||
| return backendId == DNN_BACKEND_OPENCV || | ||
| (backendId == DNN_BACKEND_INFERENCE_ENGINE_NN_BUILDER_2019 && inputsData.size() == 1); | ||
| return backendId == DNN_BACKEND_OPENCV; |
There was a problem hiding this comment.
We need to support this layer with nGraph as for me
There was a problem hiding this comment.
Ddo we need inputData.size () == 1condition?
There was a problem hiding this comment.
At least match DNN Builder implementation
|
Lets hold this activity for now. |
0b5860f to
d506147
Compare
d506147 to
9ac98f2
Compare
|
@alalek is there a way to compile current |
Pull Request Readiness Checklist
See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request
Patch to opencv_extra has the same branch name.