YOLOv5 with Deepstream 5

Please provide complete information as applicable to your setup.

**• Hardware Platform (Jetson / GPU)**Xavier
• DeepStream Version5
• 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)

In ultralytics/yolov5, use (python models/export.py --weights yolov5s.pt --img 640 --batch 1 ) to get onnx, And with config.txt

[property]
gpu-id=0
net-scale-factor=0.0039215697906911373
model-color-format=0
onnx-file=/home/nvidia/nvme/zhangsishuai.onnx
model-engine-file=/home/nvidia/nvme/zhangsishuai.onnx_b1_gpu0_fp16.engine.enc
labelfile-path=/home/nvidia/nvme/model/person/person.txt
batch-size=1
gie-unique-id=40001
#output-blob-names=prob
process-mode=1

0=FP32, 1=INT8, 2=FP16 mode

network-mode=2
num-detected-classes=80
interval=0
cluster-mode=2
maintain-aspect-ratio=1
output-tensor-meta=0
secondary-reinfer-interval=0
force-implicit-batch-dim=0
parse-bbox-func-name=NvDsInferParseCustomYoloV5

engine-create-func-name=BuildCustomYOLOv5Engine

custom-lib-path=/opt/nvidia/deepstream/deepstream-5.0/sources/libs/nvdsinfer_customparser/libnvds_infercustomparser.so

[class-attrs-all]
pre-cluster-threshold=0.2
post-cluster-threshold=0.3


parsebbox code is
Yolov5-in-Deepstream-5.0/nvdsparsebbox_Yolo.cpp at master · DanaHan/Yolov5-in-Deepstream-5.0 (github.com)

But after nms, the object is 0?
I observed that the output layer is:
1 OUTPUT kFLOAT 397 3x80x80x85
2 OUTPUT kFLOAT 458 3x40x40x85
3 OUTPUT kFLOAT 519 3x20x20x85
4 OUTPUT kFLOAT 561 25200x85
and used just one layer in code,

Does it work for you ?
What is your setup and what frames do you get ?
Do you use python or C++ to run the app

output shape of your model does not match with post processing code. Please check.

Hi, RayZhang

Do you fix this issue with the suggestion shared above?
Thanks.

Yes