Testing in deepstream_nvdsanalytics.py

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

• Hardware Platform (Jetson / GPU) Jetson
• DeepStream Version 6.2
• JetPack Version (valid for Jetson only) 5.1.1
• TensorRT Version 8.5.2.2
• Issue Type( questions, new requirements, bugs) Questions

Hi, I’m using the test.py example to test an app based on deepstream_nvdsanalytics. I have the following questions. I would like to build on test2, i.e. end-to-end testing of my application.

-Should I modify the pipeline_fakesink_tracker class to accept multiple video streams?
-Also, should I add reading the config_nvdsanalycs.py configuration file?
-Maybe there is something already developed?

Thanks a lot for the help!

3

deepstream_nvdsanalytics.py support multiple video streams as input.
Track and analyze independently in each video stream

what are pipeline_fakesink_tracker and config_nvdsanalycs.py ?

Can you share example code and config file ？

4

These are deepstream_pythons_apps files:

deepstream_python_apps/pipeline_fakesink_tracker.py at master · NVIDIA-AI-IOT/deepstream_python_apps · GitHub

deepstream_python_apps/config_nvdsanalytics.txt at master · NVIDIA-AI-IOT/deepstream_python_apps · GitHub

I would like to write a test like the example test,py but for deepstream_nvdsanalytics.py

5

Do you mean add a function like test_nvdsanalytics in test.py. then use pytest for automated test？

Depends on the usage scenario

Yes，you can use same configure file for test

not yet

6

In the end I have generated a test that checks the application thanks to the console output, taking advantage of the fact that in each nvdsanalytics event I print on screen the frame and the object.

So, I have solved it this way and for now, it is valid for me.

7

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.