Running deepstream-ssd-parser Python sample code

[nawin.ks]

• Hardware Platform (GPU)
• DeepStream Version 6.2
• TensorRT Version 8.5.2
• NVIDIA GPU Driver Version 525.85.12
• Issue Type (questions)

I am trying to run deepstream-ssd-parser (Python) after installing deepstream SDK and Python Binding. I got model as per Example nvinferserver not working . SSD model is as

/opt/nvidia/deepstream/deepstream-6.2/samples/trtis_model_repo/ssd_inception_v2_coco_2018_01_28/1/model.graphdef

Still I am getting this error:

Traceback (most recent call last):
File “deepstream_ssd_parser.py”, line 454, in
sys.exit(main(sys.argv))
File “deepstream_ssd_parser.py”, line 377, in main
pgie.set_property(“config-file-path”, “dstest_ssd_nopostprocess.txt”)
AttributeError: ‘NoneType’ object has no attribute ‘set_property’

Please help me on this. Do I need to install NVIDIA Triton Inference Server.

[yuweiw]

It seems that you create pgie failed. Did you configure your dgpu environment according to the following guide?
https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_Quickstart.html#dgpu-setup-for-ubuntu

[nawin.ks]

Yes, I am able to run sample apps also. I think, this error is related to nvinferserver. Do I need docker necessarily to run deepstream-ssd-parser (Python)? I am running in AWS EC2 instance with T4 GPU.

[yuweiw]

If you run the nvinferserver demo on your GPU directly, you should install the triton on your GPU. Or you can use our triton-docker. Please refer to the link below:
https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_troubleshooting.html#triton