Custom model in Deepstream

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) gpu
• DeepStream Version 7
• JetPack Version (valid for Jetson only)
• TensorRT Version 8.6.1
• NVIDIA GPU Driver Version (valid for GPU only) t4
**• Issue Type( questions, new requirements, bugs)**question
• 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)

Hi
I have an onnx model and i want to add it into deepstream
I have 2 question
I wanna know can i use my custom moden in deepstream using python?
Is there any example ?

3

Sure. To custom the model, you need to choose which inferencing method do you want. We have provided nvinfer DeepStream SDK FAQ - Intelligent Video Analytics / DeepStream SDK - NVIDIA Developer Forums and nvinferserver DeepStream SDK FAQ - Intelligent Video Analytics / DeepStream SDK - NVIDIA Developer Forums.

And you must know which preprocessing and postprocessing the model needs.

There are lots of samples in NVIDIA-AI-IOT/deepstream_tao_apps: Sample apps to demonstrate how to deploy models trained with TAO on DeepStream (github.com)

4

Thank for response
Did you mean if my model give me somthing like this prob-class1(0 or 1)-class 2(0 or 1)- bounding boxes
Then i can use it in deepstream by setting this items to metafile using preprocessing and post processing?

5

Do you mean your model is a detector?

6

A model with Both detector and classification

7

Then you need to customize the postprocessing.

We have postprocessing sample in /opt/nvidia/deepstream/deepstream/sources/objectDetector_Yolo

Or deepstream_python_apps/apps/deepstream-ssd-parser at master · NVIDIA-AI-IOT/deepstream_python_apps (github.com)

8

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