You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thanks to the Demo #3: SSD, I've successfully made a TensorRT optimized 'engine' of SSD.
However, I got the following error message when I tried to use the engine with NVIDIA's deepstream-app (the deepstream-app can dedicate the TensorRT engine to be used in its pipeline) :
I'm not familiar with the deepstream-app plugin structure and hope that any expert can explain what is the main cause of this problem and what I should do to use any TensorRT optimized engine in the deepstream pipeline.
(Before, I expected that the deepstream can perform inference by a user dedicated TensorRT optimized engine without any coding or plugin library building.. )
The text was updated successfully, but these errors were encountered:
@ahnjw72 Thanks for your comment. But I'm sorry. I don't currently use DeepStream SDK, and I don't have the time to look into this. I suggest you to post the issue onto NVIDIA Developer Forum and seek help from NVIDIA.
Thanks to the Demo #3: SSD, I've successfully made a TensorRT optimized 'engine' of SSD.
However, I got the following error message when I tried to use the engine with NVIDIA's deepstream-app (the deepstream-app can dedicate the TensorRT engine to be used in its pipeline) :
deepstream-app: nvdsiplugin_ssd.cpp:72: FlattenConcat::FlattenConcat(const void, size_t): Assertion `mConcatAxisID == 1 || mConcatAxisID == 2 || mConcatAxisID == 3' failed.
Aborted (core dumped)*
I'm not familiar with the deepstream-app plugin structure and hope that any expert can explain what is the main cause of this problem and what I should do to use any TensorRT optimized engine in the deepstream pipeline.
(Before, I expected that the deepstream can perform inference by a user dedicated TensorRT optimized engine without any coding or plugin library building.. )
The text was updated successfully, but these errors were encountered: