Deepstream Presentation

[joeshiref]

Hello,

I am asked to make a presentation about the usage of Deepstream SDK in Python. But my laptop doesn’t have a Nvidia GPU. Am I still able to run deepstream tests (2,3) without NVIDIA GPU?

My GPU is intel iRIS.

When I tried to run deepstream test 2, it asked me to install cuda and pyds. I used pip install cuda-python and pip install pyds inside the container. But it gives me this error now:

Creating Pipeline 

Creating Source 

Creating H264Parser

Creating Decoder

 Unable to create Nvv4l2 Decoder
 Unable to create NvStreamMux
 Unable to create pgie
 Unable to create tracker
 Unable to make sgie1
 Unable to make sgie2
 Unable to create nvvidconv
Traceback (most recent call last):
  File "/opt/nvidia/deepstream/deepstream-7.0/sources/deepstream_python_apps/apps/deepstream-test2/deepstream_test_2.py", line 350, in <module>
    sys.exit(main(sys.argv))
  File "/opt/nvidia/deepstream/deepstream-7.0/sources/deepstream_python_apps/apps/deepstream-test2/deepstream_test_2.py", line 236, in main
    if platform_info.is_integrated_gpu():
  File "/opt/nvidia/deepstream/deepstream-7.0/sources/deepstream_python_apps/apps/deepstream-test2/../common/platform_info.py", line 61, in is_integrated_gpu
    cuda_init_result, = cuda.cuInit(0)
  File "cuda/cuda.pyx", line 15991, in cuda.cuda.cuInit
  File "cuda/ccuda.pyx", line 17, in cuda.ccuda.cuInit
  File "cuda/_cuda/ccuda.pyx", line 2684, in cuda._cuda.ccuda._cuInit
  File "cuda/_cuda/ccuda.pyx", line 490, in cuda._cuda.ccuda.cuPythonInit
RuntimeError: Failed to dlopen libcuda.so.1

Thanks.

[yingliu]

No. Some plugins in deepstream_test_2 use Nvidia GPU HW, such as nvv4l2decoder, nvinfer, which will have initialization issue on non-Nvidia platform.

[joeshiref]

Thanks!