Need to know Deepstream App startup time (python)

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) - dGPU
• DeepStream Version - 5.0.1
• TensorRT Version - 7.0.0.11
• NVIDIA GPU Driver Version (valid for GPU only) - 440

Hi
Is there a way to know that the deepstream app has initialized properly and started running, because it takes some time for app to get started and there are some operation we need to do once its started and not while the app is starting.

Add delay to our operation is one way to go but not optimum.

Also need to calculate the time required from starting the app to displaying the rendered frame i.e. after running this line “loop.run()”.

Thanks

loop.run() is a loop, it only ended after the pipeline stop to work. Since the whole pipeline works in asynchronized threads, it is not possible to measure the exact frame latency(from the frame is generated to it is shown on screen) with python.

The only latency you may measure is the latency between source plugin src pad to the sink pad of your sink(I don’t know which sink you are using). You can add probe function to record the time you received the GstBuffer. You can refer to the deepstream_test_3.py, the way it calculate the fps may be a reference for you.