Currently, I am trying to execute inference through deepstream-nvdsanalytics, but there is a problem. I want to apply a model created by learning the existing model and engine with Yolo V8, but it is said that onnx can be used, but it does not work if the model is actually applied and operated
In order to apply onnx, I tried a method in which the engine was acquired from an existing engine and operated with the same engine. I tried to make an engine that onnx operates with and put it in a folder of the existing path that it operates, but it didn’t work. Of course, the contents of the txt file were changed to write down the model to be applied and the path was written down, but it didn’t work. Is there a way to solve it??
Please provide complete information as applicable to your setup.
• Hardware Platform (Jetson / GPU)
• DeepStream Version
• JetPack Version (valid for Jetson only)
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs)
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)
• Jetson Orin nano, Ubuntu 22.04
• Deepstream Version 6.4
• JetPack Version (valid for Jetson only) 6.0
• TensorRT Version 8.6.2
/opt/nvidia/deepstream/deepstream-6.4/sources/deepstream_python_apps/apps/deepstream-nvdsanalytics
It operates deepstream-nvdsanalytics.py and tries to apply a self-learned role model when it operates
The existing one uses /opt/nvidia/deepstream/deepstream-6.4/samples/models/Primary_Detector resnet18_trafficamnet.etlt model and the engine is currently running through model-engine-file=…/…/…/samples/models/Primary_Detector/resnet18_trafficcamnet.etlt_b1_gpu0_int8.engine. I modified dsnvanalytics_pgie_config.txt to make it work as an onnx file.
ONNX: I was supposed to write down the name of the onnx-file, so I wrote down the contents including the extension, ex)XXX.onnx, just .onnx, included the path of the file, and wrote it down. I tried adding it to the folder where the Python file exists for operation, and it seemed to be a problem, so I took over the contents of in8 and created the engine with the onnx file attribute, but all of them did not run.
No, when I read the readme of NVDSAnalytics, it seemed to indicate that the onnx model is also applied and can be driven, but it doesn’t run when I actually make it work, so I left a question like this.
There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks
I don’t need this so. please check why there is log “Unable to create nvanalytics”. could you share the result of “gst-inspect-1.0 nvdsanalytics” and “ldd /opt/nvidia/deepstream/deepstream/lib/gst-plugins/libnvdsgst_dsanalytics.so”?