How to debug dlstreamer

[MikeVMWEdge]

Hi - I am trying to get dlstreamer working with a custom trained Yolov5s model.   My pipeline is as follows:

GST PIPELINE::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
gst-launch-1.0 urisourcebin buffer-size=4096 uri=rtsp://user:password@192.168.0.33/axis-media/media.amp?videocodec=h264 ! decodebin ! gvadetect model=../models/cards-yolov5-ov.xml model_proc=blackjack-model-proc-v2.json device=GPU ! queue ! gvametaconvert ! gvametapublish file-format=json-lines file-path=output.json ! fakesink async=false

Model Proc is as follows

{

"json_schema_version": "2.2.0",

"input_preproc": [

{

"layer_name": "images",

"format": "image",

"params": {

"resize": "aspect-ratio"

}

}

],

"output_postproc": [

{

"do_cls_softmax": true,

"classes": 52,

"labels": [

"10C", "10D", "10H", "10S", "2C", "2D", "2H", "2S", "3C", "3D", "3H", "3S", "4C", "4D", "4H", "4S", "5C", "5D", "5H", "5S", "6C", "6D", "6H", "6S", "7C", "7D", "7H", "7S", "8C", "8D", "8H", "8S", "9C", "9D", "9H", "9S", "AC", "AD", "AH", "AS", "JC", "JD", "JH", "JS", "KC", "KD", "KH", "KS", "QC", "QD", "QH", "QS"

],

"converter": "yolo_v5"

}

]

}

Output indicates an error upon start:

ERROR: from element /GstPipeline:pipeline0/GstGvaDetect:gvadetect0: base_inference based element initialization has been failed.
Additional debug info:
/home/dlstreamer/dlstreamer/src/monolithic/gst/inference_elements/base/gva_base_inference.cpp(796): gva_base_inference_set_caps (): /GstPipeline:pipeline0/GstGvaDetect:gvadetect0:

post-processing is NULL.
Execution ended after 0:00:04.637141310

How can I debug this and find what is causing this error?  What does "post-processing is NULL." mean?

Thanks!

[Iffa_Intel]

Hi,

could you share more information:

1. Which OS are you using?
2. Which tutorial are you referring to? (please share url or etc)
3. Steps that you successfully did until point of error

Cordially,

Iffa

[Iffa_Intel]

Hi,

Thank you for your question. If you need any additional information from Intel, please submit a new question as Intel is no longer monitoring this thread.

Cordially,

Iffa